7.5-commit

to unsigned short

CHANGES

@@ -6,6 +6,558 @@
 
                                History of Changes
 
+Version 7.5
+
+Daniel (1 December 2000)
+- Craig Davison gave us his updates on the VC++ makefiles, so now curl should
+  build fine with the Microsoft compiler on windows too.
+
+- Fixed the libcurl versioning so that we don't ruin old programs when
+  releasing new shared library interfaces.
+
+Daniel (30 November 2000)
+- Renamed docs/README.curl to docs/MANUAL to better reflect what the document
+  actually contains.
+
+Daniel (29 November 2000)
+- I removed a bunch of '#if 0' sections from the code. They only make things
+  harder to follow. After all, we do have all older versions in the CVS.
+
+Version 7.5-pre5
+
+Daniel (28 November 2000)
+- I filled in more error codes in the man page error code list that had been
+  lagging.
+
+- James Griffiths mailed me a fine patch that introduces the CURLOPT_MAXREDIRS
+  libcurl option. When used, it'll prevent location following more than the
+  set number of times. It is useful to break out of endless redirect-loops.
+
+Daniel (27 November 2000)
+- Added two test cases for file://.
+
+Daniel (22 November 2000)
+- Added the libcurl CURLOPT_FILETIME setopt, when set it tries to get the
+  modified time of the remote document. This is a special option since it
+  involves an extra set of commands on FTP servers. (Using the MDTM command
+  which is not in the RFC959)
+
+  curl_easy_getinfo() got a corresponding CURLINFO_FILETIME to get the time
+  after a transfer. It'll return a zero if CURLOPT_FILETIME wasn't used or if
+  the time wasn't possible to get.
+
+  --head/-I used on a FTP server will now present a 'Last-Modified:' header
+  if curl could get the time of the specified file.
+
+- Added the option '--cacert [file]' to curl, which allows a specified PEM
+  file to be used to verify the peer's certificate when doing HTTPS
+  connections. This has been requested, rather recently by Hulka Bohuslav but
+  others have asked for it before as well.
+
+Daniel (21 November 2000)
+- Numerous fixes the test suite has brought into the daylight:
+
+   * curl_unescape() could return a too long string
+   * on ftp transfer failures, there could be memory leaks
+   * ftp CWD could use bad directory names
+   * memdebug now uses the mprintf() routines for better portability
+   * free(NULL) removed when doing resumed transfers
+
+- Added a bunch of test cases for FTP.
+
+- General cleanups to make less warnings with gcc -Wall -pedantic.
+
+- I made the tests/ftpserver.pl work with the most commonly used ftp
+  operations. PORT, PASV, RETR, STOR, LIST, SIZE, USER, PASS all work now. Now
+  all I have to do is integrate the ftp server doings in the runtests.pl
+  script so that ftp tests can be run the same way http tests already run.
+
+Daniel (20 November 2000)
+- Made libcurl capable of dealing with any-length URLs. The former limit of
+  4096 bytes was a bit annoying when people wanted to use curl to really make
+  life tough on a web server. Now, the command line limit is the most annoying
+  but that can be circumvented by using a config file.
+
+  NOTE: there is still a 4096-byte limit on URLs extracted from Location:
+  headers.
+
+- Corrected the spelling of 'resolve' in two error messages.
+
+- Alexander Kourakos posted a bug report and a patch that corrected it! It
+  turned out that lynx and wget support lowercase environment variable names
+  where curl only looked for the uppercase versions. Now curl will use the
+  lowercase versions if they exist, but if they don't, it'll use the uppercase
+  versions.
+
+Daniel (17 November 2000)
+- curl_formfree() was added. How come no one missed that one before? I ran the
+  test suite with the malloc debug enabled and got lots of "nice" warnings on
+  memory leaks. The most serious one was this. There were also leaks in the
+  cookie handling, and a few errors when curl failed to connect and similar
+  things. More tests cases were added to cover up and to verify that these
+  problems have been removed.
+
+- Mucho updated config file parser (I'm dead tired of all the bug reports and
+  weird behaviour I get on the former one). It works slightly differently now,
+  although I doubt many people will notice the differences. The main
+  difference being that if you use options that require parameters, they must
+  both be specified on the same line. With this new parser, you can also
+  specify long options without '--' and you may separate options and
+  parameters with : or =. It makes a config file line could look like:
+
+        user-agent = "foobar and something"
+
+  Parameters within quotes may contain spaces. Without quotes, they're
+  expected to be a single non-space word.
+
+  Had to patch the command line argument parser a little to make this work.
+
+- Added --url as an option to allow the URL to be specified this way. It makes
+  way nicer config files. The previous way of specifying URLs in the config
+  file doesn't work anymore.
+
+Daniel (15 November 2000)
+- Using certain characters in usernames or passwords for HTTP authentication
+  failed. This was due to the mprintf() that had a silly check for letters,
+  and if they weren't isprint() they weren't outputed "as-is". This caused
+  passwords and usernames using '<27>' (for example) to fail.
+
+Version 7.4.2
+
+Daniel (15 November 2000)
+- 'tests/runtests.pl' now sorts the test cases properly when 'all' is used.
+
+Daniel (14 November 2000)
+- I fell over the draft-ietf-ftpext-mlst-12.txt Internet Draft titled
+  "Extensions to FTP" that contains a defined way how the ftp command SIZE
+  could be assumed to work.
+
+- Laurent Papier posted a bug report about using "-C -" and FTP uploading a
+  file that isn't prsent on the server. The server might then return a 550 and
+  curl will fail. Should it instead as Laurent Papier suggests, start
+  uploading from the beginning as a normal upload?
+
+Daniel (13 November 2000)
+- Fixed a crash with the followlocation counter.
+
+- While writing test cases for the test suite, I discovered an old limitation
+  that prevented -o and -T to be used at the same time. I removed this
+  immediately as this has no relevance in the current libcurl.
+  
+- Chris Faherty fixed a free-twice problem in lib/file.c
+
+- I fixed the perl http server problem in the test suite.
+
+Version 7.4.2 pre4
+
+Daniel (10 November 2000)
+- I've (finally) started working on the curl test suite. It is in the new
+  tests/ directory. It requires sh and perl. There's a TCP server in perl and
+  most of the other stuff running a pretty simple shell script.
+
+  I've only made four test cases so far, but it proves the system can work.
+
+- Laurent Papier noticed that curl didn't set TYPE when doing --head checks
+  for sizes on FTP servers. Some servers seem to return different sizes
+  depending on whether ASCII or BINARY is used!
+
+- Laurent Papier detected that if you appended a FTP upload and everything was
+  already uploaded, curl would hang.
+
+- Angus Mackay's getpass_r() in lib/getpass.c is now compliant with the
+  getpass_r() function it seems some systems actually have.
+  
+- Venkataramana Mokkapati detected a bug in the cookie parser and corrected
+  it.  If the cookie was set for the full host name (domain=full.host.com),
+  the cookie was never sent back because of a faulty length comparison between
+  the set domain length and the current host name.
+
+Daniel (9 November 2000)
+- Added a configure check for gethostbyname in -lsocket (OS/2 seems to need
+  it). Added a check for RSAglue/rsaref for the cases where libcrypto is found
+  but libssl isn't. I haven't verified this fix yet though, as I have no
+  system that requires those libs to build.
+  
+Version 7.4.2 pre3
+
+Daniel (7 November 2000)
+- Removed perror() outputs from getpass.c. Angus Mackay also agreed to a
+  slightly modified license of the getpass.c file as the prototype was changed.
+
+Daniel (6 November 2000)
+- Added possibility to set a password callback to use instead of the built-in.
+  They're controled with curl_easy_setopt() of course, the tags are
+  CURLOPT_PASSWDFUNCTION and CURLOPT_PASSWDDATA.
+
+- Used T. Bharath's thinking and fixed the timers that showed terribly wrong
+  times when location: headers were followed.
+
+- Emmanuel Tychon discovered that curl didn't really like user names only in
+  the URL. I corrected this and I also fixed the since long living problem
+  with URL encoded user names and passwords in the URLs. They should work now.
+  
+Daniel (2 November 2000)
+- When I added --interface, the new error code that was added with it was
+  inserted in the wrong place and thus all error codes from 35 and upwards got
+  increased one step. This is now corrected, we're back at the previous
+  numbers. All new exit codes should be added at the end.
+
+Daniel (1 November 2000)
+- Added a check for signal() in the configure script so that if sigaction()
+  isn't present, we can use signal() instead.
+
+- I'm having a license discussion going on privately. The issue is yet again
+  GPL-licensed programs that have problems with MPL. I am leaning towards
+  making a kind of dual-license that will solve this once and for all...
+
+Daniel (31 October 2000)
+- Added the packages/ directory. I intend to let this contain some docs and
+  templates on how to generate custom-format packages for various platforms.
+  I've now removed the RPM related curl.spec files from the archive root.
+
+Daniel (30 October 2000)
+- T. Bharath brought a set of patches that bring new functionality to
+  curl_easy_getinfo() and curl_easy_setopt(). Now you can request peer
+  certificate verification with the *setopt() CURLOPT_SSL_VERIFYPEER option
+  and then use the CURLOPT_CAINFO to set the certificate to verify the remote
+  peer against. After an such an operation with a verification request, the
+  *_getinfo() option CURLINFO_SSL_VERIFYRESULT will return information about
+  whether the verification succeeded or not.  
+
+Daniel (27 October 2000)
+- Georg Horn brought us a splendid patch that solves the long-standing
+  annoying problem with timeouts that made curl exit with silly exit codes
+  (which as been commented out lately). This solution is sigaction() based and
+  of course then only works for unixes (and only those unixes that actually
+  have the sigaction() function).
+
+Daniel (26 October 2000)
+- Bj<42>rn Stenberg supplied a patch that fixed the flaw mentioned by Kevin Roth
+  that made the password get echoed when prompted for interactively. The
+  getpass() function (now known as my_getpass()) was also fixed to not use any
+  static buffers. This also means we cannot use the "standard" getpass()
+  function even for those systems that have it, since it isn't thread-safe.
+  
+- Kevin Roth found out that if you'd write a config file with '-v url', the
+  url would not be used as "default URL" as documented, although if you wrote
+  it 'url -v' it worked! This has been corrected now.
+
+- Kevin Roth's idea of using multiple -d options on the same command line was
+  just brilliant, and I couldn't really think of any reason why we shouldn't
+  support it! The append function always append '&' and then the new -d
+  chunk. This enables constructs like the following:
+
+        curl -d name=daniel -d age=unknown foobarsite.com
+
+Daniel (24 October 2000)
+- I fixed the lib/memdebug.c source so that it compiles on Linux and other
+  systems. It will be useful one day when someone else but me wants to run the
+  memory debugging system.
+
+Daniel (23 October 2000)
+- I modified the maketgz and configure scripts, so that the configure script
+  will fetch the version number from the include/curl/curl.h header files, and
+  then the maketgz doesn't have to rebuild the configure script when I build
+  release-archives.
+
+- Bj<42>rn Stenberg and Linus Nielsen correctly pointed out that curl was silly
+  enough to not allow @-letters in passwords when they were specified with the
+  -u or -U flags (CURLOPT_USERPWD and CURLOPT_PROXYUSERPWD). This also
+  suggests that curl probably should url-decode the password piece of an URL
+  so that you could pass an encoded @-letter there...
+  
+Daniel (20 October 2000)
+- Yet another http server barfed on curl's request that include the port
+  number in the Host: header always. I now only include the port number if it
+  isn't the default (80 for HTTP, 443 for HTTPS). www.perl.com turned out to
+  run one of those nasty servers.
+
+- The PHP4 module for curl had problems with referer that seems to have been
+  corrected just yesterday. (Sterling Hughes of the PHP team confirmed this)
+
+Daniel (17 October 2000)
+- Vladimir Oblomov reported that the -Y and -y options didn't work. They
+  didn't work for me either. This once again proves we should have that test
+  suite...
+  
+- I finally changed the error message libcurl returns if you try a https://
+  URL when the library wasn't build with SSL enabled. It will now return this
+  error:
+        "libcurl was built with SSL disabled, https: not supported!"
+
+  I really hope it will make it a bit clearer to users where the actual
+  problem lies.
+
+Version 7.4.1
+
+Daniel (16 October 2000)
+- I forgot to remove some of the malloc debug defines from the makefiles in
+  the release archive (of course).
+
+Version 7.4
+
+Daniel (16 October 2000)
+- The buffer overflow mentioned below was posted to bugtraq on Friday 13th.
+
+Daniel (12 October 2000)
+- Colin Robert Phipps elegantly corrected a buffer overflow. It could be used
+  by an evil ftp server to crash curl. I took the opportunity of replacing a
+  few other sprintf()s into snprintf()s as well.
+
+Daniel (11 October 2000)
+- Found some more memory leaks. This new simple memory debugger has turned out
+  really useful!
+
+Version 7.4 pre6
+
+Daniel (9 October 2000)
+- Florian Koenig pointed out that the bool typedef in the curl/curl.h include
+  file was breaking PHP 4.0.3 compiling. The bool typedef is not used in the
+  public interface and was wrongly inserted in that header file.
+
+- J<>rg Hartroth corrected a minor memory leak in the src/urlglob.c stuff. It
+  didn't harm anyone since the memory is free()ed on exit anyway.
+
+- Corrected the src/main.c. We use the _MPRINTF_REPLACE #define to use our
+  libcurl-printf() functions. This gives us snprintf() et al on all
+  platforms. I converted the allocated useragent string to one that uses a
+  local buffer.
+
+- I've set an #if 0 section around the Content-Transfer-Encoding header
+  generated in lib/formdata.c. This will hopefully make curl do more
+  PHP-friendly multi-part posts.
+
+Version 7.4 pre5
+
+Daniel (9 October 2000)
+- Nico Baggus found out that curl's ability to force a ASCII download when
+  using FTP was no longer working! I corrected this. This problem was probably
+  introduced when I redesigned libcurl for version 7.
+
+- Georg Horn provided a source example that proved a memory leak in libcurl.
+  I added simple memory debugging facilities and now we can make libcurl log
+  all memory fiddling functions. An additional perl script is used to analyze
+  the output logfile and to match malloc()s with free()s etc. The memory leak
+  Georg found turned out to be the main cookie struct that cookie_cleanup()
+  didn't free! The perl script is named memanalyze.pl and it is available in
+  the CVS respository, not in the release archive.
+
+Daniel (8 October 2000)
+- Georg Horn found a GetHost() problem. It turned out it never assigned the
+  pointer in the third argument properly! This could make a crash, or at best
+  a memory leak!
+
+Version 7.4 pre4
+
+Daniel (6 October 2000)
+- Is the -F post following the RFC 1867 spec? We had this dicussion on the
+  mailing list since it appears curl can't post -F form posts to a PHP
+  receiver... I've been in touch with the PHP developers about this.
+
+- Domenico Andreoli found out that the long option '--proxy' wasn't working
+  anymore! The option parser got confused when I added the --proxytunnel for
+  7.3. This was indeed a very old flaw that hasn't turned up until now...
+
+- J<>rn Hartroth provided patches, updated makefiles and two new files for DLL
+  stuff on win32. He also pointed out that lib source files were compiled with
+  -I../src which isn't only wrong but plain stupid!
+
+- Troels Walsted Hansen fixed a problem with HTTP resume. Curl previously used
+  a local variable badly, that could lead to crashes.
+
+Version 7.4 pre3
+
+Daniel (4 October 2000)
+- More docs written. The curl_easy_getinfo.3 man page is now pretty accurate,
+  as is the -w section in curl.1. I added two options to enable the user to
+  get information about the received headers' size and the size of the HTTP
+  request. T. Bharath requested them.
+  
+Daniel (3 October 2000)
+- Corrected a sever free() before use in the new add_buffer_send()! ;-)
+
+Version 7.4 pre2
+
+Daniel (3 October 2000)
+- Jason S. Priebe sent me patches that changed the way curl issues HTTP
+  requests. The entire request is now issued in one single shot. It didn't do
+  this previously, and it has turned out that since the common browsers do it
+  this way, some sites have turned out to work with browsers but not with
+  curl! Although this is not a client-side problem, we want to be able to
+  fully emulate browsers, and thus we have now adjusted the networking layer
+  to slightly more appear as a browser. I adjusted Jason's patch, the faults
+  are probably mine.
+
+Daniel (2 October 2000)
+- Anyone who ever uploaded data with curl on a slow link has noticed that the
+  progess meter is updated very infrequently. That is due to the large buffer
+  size curl is using. It reads 50Kb and sends it, updates the progress meter
+  and loops. 50Kb is very much on a slow link, although it is pretty neat to
+  use on a fast one.
+
+  I've now made an adjustment that makes curl use a 2Kb buffer for uploads to
+  start with. If curl's average upload speed is faster than buffer size bytes
+  per second, curl will increase the used buffer size up to max 50Kb. It
+  should make the progress meter work better.
+  
+Version 7.4 pre1
+
+Daniel (29 September 2000)
+- Ripped out the -w stuff from the library and put in the curl tool. It gets
+  all the relevant info from the library using the new curl_easy_getinfo()
+  function.
+
+- brad at openbsd.org mailed me a patch that corrected my kerberos mistake and
+  removed a compiler warning from hostip.c that OpenBSD people get.
+
+Daniel (28 September 2000)
+- Of course (I should probably get punished somehow) I didn't properly correct
+  the #include lines for the base64 stuff in the kerberos sources in the just
+  released 7.3 package. They still include the *_krb.h files! Now, the error
+  is sooo very easy to spot and fix so I won't bother with a quick bug fix
+  release. I'll post a patch whenever one is needed instead. It'll be
+  available in the CVS in a few minutes anyway.
+
+Version 7.3
+
+Daniel (28 September 2000)
+- Removed the base64_krb.[ch] files. They've now replaced the former
+  base64.[ch] files.
+
+Daniel (26 September 2000)
+- Updated some docs.
+
+- I changed the OpenSSL fix to work with older versions as well. The posted
+  patch was only working with 0.9.6 and no older ones.
+  
+Version 7.3-pre8
+
+Daniel (25 September 2000)
+- Erdmut Pfeifer informed us that curl didn't build with OpenSSL 0.9.6 and
+  showed us what needed to get patched in order to make it build properly
+  again.
+
+- Dirk Kruschewski found a bug in the cookie parser. I made an alternative
+  approach to the solution Dirk himself suggested. The bug made a cookie
+  header that didn't end with a trailing semicolon to not get parsed.
+
+- I've marked -c and -t deprecated now. If you use any of them, curl will tell
+  you to use "-C -" or "-T -" instead. I don't think occupying two letters for
+  nearly identical functions is good use. Also, -T - kind of follows the curl
+  tradition of using - for stdin where a file name is expected.
+
+Daniel (23 September 2000)
+- Martin Hedenfalk provided the patch that finally made the krb4 ftp upload
+  work!
+
+Daniel (21 September 2000)
+- The kerberos code is not quite thread-safe yet. There are a few more globals
+  that need to be take care of. Let's get the upload working first!
+
+Daniel (20 September 2000)
+- Richard Prescott solved another name lookup buffer size problem. I took this
+  opportunity to rewrite the GetHost() function. With these large buffer
+  sizes, I think keeping them as local arrays quickly turn ugly. I now use
+  malloc() to get the buffer memory. Thanks to this, I now can realloc() to a
+  large buffer in case of demand (errno == ERANGE) in case a solution like
+  that would become necessary. I still want to avoid that kind of nastiness.
+
+- Tried to compile and run curl on Linux for alpha and FreeBSD for alpha. Went
+  as smooth as it could.
+
+- Added a docs/examples directory with two tiny example sources that show how
+  to use libcurl. I hope users will supply me with more useful examples
+  further on.
+
+- Applied a patch by J<>rn Hartroth to no longer use the word 'inteface' in the
+  config struct in the src/main.c file since certain compilers have that word
+  "reservered".  I figure that is some kind of C++ decease.
+
+- Updated the curl.1 man page with --interface and --krb4.
+
+- Modified the base64Encode() function to work like the kerberos one, so that
+  I could remove the use of that. There is no need for *two* base64 encoding
+  functions! ;-)
+
+Version 7.3pre5
+
+Daniel (19 September 2000)
+- The kerberos4-layer source code that is much "influenced" by the original
+  krb4 source code, through yafc into curl, was using quite a lot of global
+  variables. libcurl can't work properly with globals like that why I had to
+  clean up almost every function in the new security.c to make them use
+  connection specific variables instead of the globals. I just hope I didn't
+  destroy anything now... :-) configure updated, version string now reflects
+  krb4 built-in. It almost works now. Only uploads are still being naughty.
+
+Version 7.3pre3
+
+Daniel (18 September 2000)
+- Martin Hedenfalk supplied a major patch that introduces krb4-ftp support to
+  curl. Martin is the primary author of the ftp client named yafc and he did
+  not hesitate to help us implement this when I asked him. Many and sincere
+  thanks to a splendid effort. It didn't even take many hours!
+
+- Stephen Kick supplied a big patch that introduces the --interface flag to
+  the curl tool and CURLOPT_INTERFACE for libcurl. It allows you to specify an
+  outgoing interface to use for your request. This may not work on all
+  platforms. This needs testing.
+
+- Richard Prescott noticed that curl on Tru64 unix could core dumped if the
+  name didn't resolve properly. This was due to the GetHost() function not
+  returning an error even though it failed on some platforms!
+
+Daniel (15 September 2000)
+- Updated all sorts of documents in regards to the new proxytunnel support.
+
+Version 7.3pre2
+
+Daniel (15 September 2000)
+- Kai-Uwe Rommel pointed out a problem in the httpproxytunnel stuff for ftp.
+  Adjusted it. Added better info message when setting up the tunnel and the
+  pasv message when doing the second connect.
+  
+Version 7.3pre1
+
+Daniel (15 September 2000)
+- libcurl now allows "httpproxytunnel" to an arbitrary host and port name. The
+  second connection on ftp needed that.
+
+- TheArtOfHTTPScripting was corrected all over. I both type and spell really
+  bad at times!
+  
+Daniel (14 September 2000)
+- -p/--proxytunnel was added to 'curl'. It uses the new
+  CURLOPT_HTTPPROXYTUNNEL libcurl option that allows "any" protocol to tunnel
+  through the specified http proxy. At the moment, this should work with ftp.
+
+Daniel (13 September 2000)
+- Jochen Schaeuble found that file:// didn't work as expected. Corrected this
+  and mailed the patch to the mailing list.
+
+Daniel (7 September 2000)
+- I changed the #define T() in curl.h since it turned out it wasn't really
+  a good symbol to use (when you compiled PHP with curl as a module, that
+  define collided with some IMAP define or something). This was posted to the
+  PHP bug tracker.
+
+- I added extern "C" stuff in two header files to better allow libcurl usage
+  in C++ sorces. Discussions on the libcurl list with Danny Horswell lead to
+  this.
+
+Version 7.2.1
+
+Daniel (31 August 2000)
+- Albert Chin-A-Young fixed the configure script *again* and now it seems to
+  detect Linux name resolving properly! (heard that before?)
+
+- Troels Walsted Hansen pointed out that downloading a file containing the
+  letter '+' from an ftp server didn't work. It did work from HTTP though and
+  the reason was my lame URL decoder.
+
+- I happened to notice that -I didn't at all work on ftp anymore. I corrected
+  that.
+
 Version 7.2
 
 Daniel (30 August 2000)
@@ -2139,7 +2691,7 @@ Version 4 (1998-03-20)
 
 ***************************************************************************
 
-Version 3.12
+Version 3.12 (14 March 1998)
  Daniel Stenberg
  - End-of-header tracking still lacked support for \r\n or just \n at the
    end of the last header line.

FILES

@@ -3,15 +3,13 @@ FILES
 LEGAL
 MPL-1.0.txt
 README
-*spec
-*spec.in
 docs/BUGS
 docs/CONTRIBUTE
 docs/FAQ
 docs/FEATURES
 docs/INSTALL
 docs/INTERNALS
-docs/README.curl
+docs/MANUAL
 docs/README.win32
 docs/README.libcurl
 docs/RESOURCES
@@ -21,6 +19,8 @@ docs/Makefile.in
 docs/Makefile.am
 docs/TheArtOfHttpScripting
 docs/*.3
+docs/examples/README
+docs/examples/*.c
 maketgz
 Makefile.in
 Makefile.am
@@ -40,9 +40,17 @@ reconf
 stamp-h.in
 ltconfig
 ltmain.sh
-src/*.[ch]
-src/*in
-src/*am
+src/config-win32.h
+src/hugehelp.c
+src/main.c
+src/setup.h
+src/urlglob.c
+src/urlglob.h
+src/version.h
+src/writeout.c
+src/writeout.h
+src/*.in
+src/*.am
 src/mkhelp.pl
 src/Makefile.vc6
 src/*m32
@@ -52,10 +60,23 @@ lib/*in
 lib/*am
 lib/Makefile.vc6
 lib/*m32
+lib/libcurl.def
 include/README
 include/Makefile.in
 include/Makefile.am
 include/curl/*.h
 include/curl/Makefile.in
 include/curl/Makefile.am
-
+packages/Linux/RPM/curl-ssl.spec
+packages/Linux/RPM/curl.spec
+packages/Linux/RPM/make_curl_rpm
+packages/Linux/RPM/README
+packages/Win32/README
+packages/README
+tests/Makefile.am
+tests/Makefile.in
+tests/runtests.pl
+tests/README
+tests/httpserver.pl
+tests/ftpserver.pl
+tests/data/*.txt

Makefile.am

@@ -6,5 +6,7 @@ AUTOMAKE_OPTIONS = foreign no-dependencies
 
 EXTRA_DIST = curl.spec curl-ssl.spec
 
-SUBDIRS = docs lib src include
+SUBDIRS = docs lib src include tests
 
+test:
+	@(cd tests; make quiet-test)

Makefile.dist

@@ -58,8 +58,16 @@ mingw32-ssl:
 	cd src; make -f Makefile.m32 SSL=1
 
 vc:
-	cd lib; nmake -f Makefile.vc6
-	cd src; nmake -f Makefile.vc6
+	cd lib
+	nmake -f Makefile.vc6
+	cd ..\src
+	nmake -f Makefile.vc6
+
+vc-ssl:
+	cd lib
+	nmake -f Makefile.vc6 release-ssl
+	cd ..\src
+	nmake -f Makefile.vc6
 
 cygwin:
 	./configure

README

@@ -8,7 +8,7 @@ README
 
   Curl is a command line tool for transfering data specified with URL
   syntax. Find out how to use Curl by reading the curl.1 man page or the
-  README.curl document. Find out how to install Curl by reading the INSTALL
+  MANUAL document. Find out how to install Curl by reading the INSTALL
   document.
 
   libcurl is a library that Curl is using to do its job. It is readily
@@ -41,3 +41,7 @@ README
         cvs -d :pserver:anonymous@cvs.curl.sourceforge.net:/cvsroot/curl logout
         
      (you're off the hook!)
+
+  Curl contains pieces of source code that is Copyright (c) 1998, 1999
+  Kungliga Tekniska H<>gskolan. This notice is included here to comply with the
+  distribution terms.

acconfig.h

@@ -33,3 +33,6 @@
 
 /* Define if you need the _REENTRANT define for some functions */
 #undef NEED_REENTRANT
+
+/* Define if you have the Kerberos4 libraries (including -ldes) */
+#undef KRB4

config.h.in

@@ -49,6 +49,9 @@
 /* Define if you need the _REENTRANT define for some functions */
 #undef NEED_REENTRANT
 
+/* Define if you have the Kerberos4 libraries (including -ldes) */
+#undef KRB4
+
 /* The number of bytes in a long double.  */
 #undef SIZEOF_LONG_DOUBLE
 
@@ -76,8 +79,8 @@
 /* Define if you have the gethostname function.  */
 #undef HAVE_GETHOSTNAME
 
-/* Define if you have the getpass function.  */
-#undef HAVE_GETPASS
+/* Define if you have the getpass_r function.  */
+#undef HAVE_GETPASS_R
 
 /* Define if you have the getservbyname function.  */
 #undef HAVE_GETSERVBYNAME
@@ -94,6 +97,9 @@
 /* Define if you have the inet_ntoa_r function.  */
 #undef HAVE_INET_NTOA_R
 
+/* Define if you have the krb_get_our_ip_for_realm function.  */
+#undef HAVE_KRB_GET_OUR_IP_FOR_REALM
+
 /* Define if you have the localtime_r function.  */
 #undef HAVE_LOCALTIME_R
 
@@ -106,6 +112,12 @@
 /* Define if you have the setvbuf function.  */
 #undef HAVE_SETVBUF
 
+/* Define if you have the sigaction function.  */
+#undef HAVE_SIGACTION
+
+/* Define if you have the signal function.  */
+#undef HAVE_SIGNAL
+
 /* Define if you have the socket function.  */
 #undef HAVE_SOCKET
 
@@ -124,6 +136,9 @@
 /* Define if you have the stricmp function.  */
 #undef HAVE_STRICMP
 
+/* Define if you have the strlcpy function.  */
+#undef HAVE_STRLCPY
+
 /* Define if you have the strstr function.  */
 #undef HAVE_STRSTR
 
@@ -145,6 +160,9 @@
 /* Define if you have the <crypto.h> header file.  */
 #undef HAVE_CRYPTO_H
 
+/* Define if you have the <des.h> header file.  */
+#undef HAVE_DES_H
+
 /* Define if you have the <dlfcn.h> header file.  */
 #undef HAVE_DLFCN_H
 
@@ -160,6 +178,9 @@
 /* Define if you have the <io.h> header file.  */
 #undef HAVE_IO_H
 
+/* Define if you have the <krb.h> header file.  */
+#undef HAVE_KRB_H
+
 /* Define if you have the <malloc.h> header file.  */
 #undef HAVE_MALLOC_H
 
@@ -256,6 +277,9 @@
 /* Define if you have the nsl library (-lnsl).  */
 #undef HAVE_LIBNSL
 
+/* Define if you have the resolv library (-lresolv).  */
+#undef HAVE_LIBRESOLV
+
 /* Define if you have the resolve library (-lresolve).  */
 #undef HAVE_LIBRESOLVE
 

configure.in

@@ -2,7 +2,9 @@ dnl $Id$
 dnl Process this file with autoconf to produce a configure script.
 AC_INIT(lib/urldata.h)
 AM_CONFIG_HEADER(config.h src/config.h)
-AM_INIT_AUTOMAKE(curl,"7.2")
+
+VERSION=`sed -ne 's/^#define LIBCURL_VERSION "\(.*\)"/\1/p' include/curl/curl.h`
+AM_INIT_AUTOMAKE(curl,$VERSION)
 AM_PROG_LIBTOOL
 
 dnl
@@ -24,148 +26,126 @@ dnl The install stuff has already been taken care of by the automake stuff
 dnl AC_PROG_INSTALL
 AC_PROG_MAKE_SET
 
+AC_DEFUN(CURL_CHECK_LOCALTIME_R,
+[
+  dnl check for a few thread-safe functions
+  AC_CHECK_FUNCS(localtime_r,[
+    AC_MSG_CHECKING(whether localtime_r is declared)
+    AC_EGREP_CPP(localtime_r,[
+#include <time.h>],[
+      AC_MSG_RESULT(yes)],[
+      AC_MSG_RESULT(no)
+      AC_MSG_CHECKING(whether localtime_r with -D_REENTRANT is declared)
+      AC_EGREP_CPP(localtime_r,[
+#define _REENTRANT
+#include <time.h>],[
+	AC_DEFINE(NEED_REENTRANT)
+	AC_MSG_RESULT(yes)],
+	AC_MSG_RESULT(no))])])
+])
 
-dnl **********************************************************************
-dnl Checks for libraries.
-dnl **********************************************************************
+AC_DEFUN(CURL_CHECK_INET_NTOA_R,
+[
+  dnl determine if function definition for inet_ntoa_r exists.
+  AC_CHECK_FUNCS(inet_ntoa_r,[
+    AC_MSG_CHECKING(whether inet_ntoa_r is declared)
+    AC_EGREP_CPP(inet_ntoa_r,[
+#include <arpa/inet.h>],[
+      AC_DEFINE(HAVE_INET_NTOA_R_DECL)
+      AC_MSG_RESULT(yes)],[
+      AC_MSG_RESULT(no)
+      AC_MSG_CHECKING(whether inet_ntoa_r with -D_REENTRANT is declared)
+      AC_EGREP_CPP(inet_ntoa_r,[
+#define _REENTRANT
+#include <arpa/inet.h>],[
+	AC_DEFINE(HAVE_INET_NTOA_R_DECL)
+	AC_DEFINE(NEED_REENTRANT)
+	AC_MSG_RESULT(yes)],
+	AC_MSG_RESULT(no))])])
 
-dnl nsl lib?
-AC_CHECK_FUNC(gethostbyname, , AC_CHECK_LIB(nsl, gethostbyname))
+])
 
-dnl At least one system has been identified to require BOTH nsl and
-dnl socket libs to link properly.
-if test "$ac_cv_lib_nsl_gethostbyname" = "$ac_cv_func_gethostbyname"; then
-  AC_MSG_CHECKING([trying both nsl and socket libs])
-  my_ac_save_LIBS=$LIBS
-  LIBS="-lnsl -lsocket $LIBS"
-  AC_TRY_LINK( ,
-             [gethostbyname();],
-             my_ac_link_result=success,
-             my_ac_link_result=failure )
+AC_DEFUN(CURL_CHECK_GETHOSTBYADDR_R,
+[
+  dnl check for number of arguments to gethostbyaddr_r. it might take
+  dnl either 5, 7, or 8 arguments.
+  AC_CHECK_FUNCS(gethostbyaddr_r,[
+    AC_MSG_CHECKING(if gethostbyaddr_r takes 5 arguments)
+    AC_TRY_COMPILE([
+#include <sys/types.h>
+#include <netdb.h>],[
+char * address;
+int length;
+int type;
+struct hostent h;
+struct hostent_data hdata;
+int rc;
+rc = gethostbyaddr_r(address, length, type, &h, &hdata);],[
+      AC_MSG_RESULT(yes)
+      AC_DEFINE(HAVE_GETHOSTBYADDR_R_5)
+      ac_cv_gethostbyaddr_args=5],[
+      AC_MSG_RESULT(no)
+      AC_MSG_CHECKING(if gethostbyaddr_r with -D_REENTRANT takes 5 arguments)
+      AC_TRY_COMPILE([
+#define _REENTRANT
+#include <sys/types.h>
+#include <netdb.h>],[
+char * address;
+int length;
+int type;
+struct hostent h;
+struct hostent_data hdata;
+int rc;
+rc = gethostbyaddr_r(address, length, type, &h, &hdata);],[
+	AC_MSG_RESULT(yes)
+	AC_DEFINE(HAVE_GETHOSTBYADDR_R_5)
+	AC_DEFINE(NEED_REENTRANT)
+	ac_cv_gethostbyaddr_args=5],[
+	AC_MSG_RESULT(no)
+	AC_MSG_CHECKING(if gethostbyaddr_r takes 7 arguments)
+	AC_TRY_COMPILE([
+#include <sys/types.h>
+#include <netdb.h>],[
+char * address;
+int length;
+int type;
+struct hostent h;
+char buffer[8192];
+int h_errnop;
+struct hostent * hp;
 
-  if test "$my_ac_link_result" = "failure"; then
-    AC_MSG_RESULT([no])
-    AC_MSG_ERROR([couldn't find libraries for gethostbyname()])
-    dnl restore LIBS
-    LIBS=$my_ac_save_LIBS
-  else
-    AC_MSG_RESULT([yes])
-  fi
-fi
+hp = gethostbyaddr_r(address, length, type, &h,
+                     buffer, 8192, &h_errnop);],[
+	  AC_MSG_RESULT(yes)
+	  AC_DEFINE(HAVE_GETHOSTBYADDR_R_7)
+	  ac_cv_gethostbyaddr_args=7],[
+	  AC_MSG_RESULT(no)
+	  AC_MSG_CHECKING(if gethostbyaddr_r takes 8 arguments)
+	  AC_TRY_COMPILE([
+#include <sys/types.h>
+#include <netdb.h>],[
+char * address;
+int length;
+int type;
+struct hostent h;
+char buffer[8192];
+int h_errnop;
+struct hostent * hp;
+int rc;
 
-dnl resolve lib?
-AC_CHECK_FUNC(strcasecmp, , AC_CHECK_LIB(resolve, strcasecmp))
-
-if test "$ac_cv_lib_resolve_strcasecmp" = "$ac_cv_func_strcasecmp"; then
-  AC_CHECK_LIB(resolve, strcasecmp,
-              [LIBS="-lresolve $LIBS"],
-               ,
-               -lnsl)
-fi
-
-dnl socket lib?
-AC_CHECK_FUNC(connect, , AC_CHECK_LIB(socket, connect))
-
-dnl ucb lib?
-AC_CHECK_FUNC(gethostname, , AC_CHECK_LIB(ucb, gethostname))
-
-dnl dl lib?
-AC_CHECK_FUNC(dlopen, , AC_CHECK_LIB(dl, dlopen))
-
-dnl **********************************************************************
-dnl Check for the presence of SSL libraries and headers
-dnl **********************************************************************
-
-dnl Default to compiler & linker defaults for SSL files & libraries.
-OPT_SSL=off
-AC_ARG_WITH(ssl,dnl
-[  --with-ssl[=DIR]        where to look for SSL [compiler/linker default paths]
-                          DIR points to the SSL installation [/usr/local/ssl]],
-  OPT_SSL=$withval
-)
-
-if test X"$OPT_SSL" = Xno
-then
-  AC_MSG_WARN(SSL/https support disabled)  
-else
-
-  dnl Check for & handle argument to --with-ssl.
-
-  AC_MSG_CHECKING(where to look for SSL)
-  if test X"$OPT_SSL" = Xoff
-  then
-  	AC_MSG_RESULT([defaults (or given in environment)])
-  else
-	test X"$OPT_SSL" = Xyes && OPT_SSL=/usr/local/ssl
-	LIBS="$LIBS -L$OPT_SSL/lib"
-	CPPFLAGS="$CPPFLAGS -I$OPT_SSL/include/openssl -I$OPT_SSL/include"
-	AC_MSG_RESULT([$OPT_SSL])
-  fi
-
-  dnl check for crypto libs (part of SSLeay)
-  AC_CHECK_LIB(crypto, CRYPTO_lock)
-
-  if test $ac_cv_lib_crypto_CRYPTO_lock = yes; then
-    dnl This is only reasonable to do if crypto actually is there: check for
-    dnl SSL libs NOTE: it is important to do this AFTER the crypto lib
-    AC_CHECK_LIB(ssl, SSL_connect)
-
-    dnl Check for SSLeay headers
-    AC_CHECK_HEADERS(openssl/x509.h openssl/rsa.h openssl/crypto.h openssl/pem.h openssl/ssl.h openssl/err.h)
-
-    if test $ac_cv_header_openssl_x509_h = no; then
-      AC_CHECK_HEADERS(x509.h rsa.h crypto.h pem.h ssl.h err.h)
-    fi
-
-  fi
-fi
-
-dnl **********************************************************************
-dnl Check for the presence of ZLIB libraries and headers
-dnl **********************************************************************
-
-dnl Default to compiler & linker defaults for files & libraries.
-dnl OPT_ZLIB=no
-dnl AC_ARG_WITH(zlib,dnl
-dnl [  --with-zlib[=DIR]  where to look for ZLIB [compiler/linker default paths]
-dnl                      DIR points to the ZLIB installation prefix [/usr/local]],
-dnl  OPT_ZLIB=$withval,
-dnl )
-
-dnl Check for & handle argument to --with-zlib.
-dnl
-dnl NOTE:  We *always* look for ZLIB headers & libraries, all this option
-dnl        does is change where we look (by adjusting LIBS and CPPFLAGS.)
-dnl
-
-dnl AC_MSG_CHECKING(where to look for ZLIB)
-dnl if test X"$OPT_ZLIB" = Xno
-dnl then
-dnl 	AC_MSG_RESULT([defaults (or given in environment)])
-dnl else
-dnl	test X"$OPT_ZLIB" = Xyes && OPT_ZLIB=/usr/local
-dnl	LIBS="$LIBS -L$OPT_ZLIB/lib"
-dnl	CPPFLAGS="$CPPFLAGS -I$OPT_ZLIB/include"
-dnl	AC_MSG_RESULT([$OPT_ZLIB])
-dnl fi
-
-dnl z lib?
-dnl AC_CHECK_FUNC(gzread, , AC_CHECK_LIB(z, gzread))
+rc = gethostbyaddr_r(address, length, type, &h,
+                     buffer, 8192, &hp, &h_errnop);],[
+	    AC_MSG_RESULT(yes)
+	    AC_DEFINE(HAVE_GETHOSTBYADDR_R_8)
+	    ac_cv_gethostbyaddr_args=8],[
+	    AC_MSG_RESULT(no)
+	    have_missing_r_funcs="$have_missing_r_funcs gethostbyaddr_r"])])])])])
 
 
-dnl Default is to try the thread-safe versions of a few functions
-OPT_THREAD=on
-AC_ARG_ENABLE(thread,dnl
-[  --disable-thread       tell configure to not look for thread-safe functions],
-  OPT_THREAD=off
-)
+])
 
-if test X"$OPT_THREAD" = Xoff
-then
-  AC_MSG_WARN(libcurl will not get built using thread-safe functions)
-  AC_DEFINE(DISABLED_THREADSAFE, 1, \
-Set to explicitly specify we don't want to use thread-safe functions)
-else
+AC_DEFUN(CURL_CHECK_GETHOSTBYNAME_R,
+[
   dnl check for number of arguments to gethostbyname_r. it might take
   dnl either 3, 5, or 6 arguments.
   AC_CHECK_FUNCS(gethostbyname_r,[
@@ -255,113 +235,273 @@ exit (rc != 0 ? 1 : 0); }],[
 	[ac_cv_gethostbyname_args=0])],
       [ac_cv_gethostbyname_args=0])])
 
-  dnl check for number of arguments to gethostbyaddr_r. it might take
-  dnl either 5, 7, or 8 arguments.
-  AC_CHECK_FUNCS(gethostbyaddr_r,[
-    AC_MSG_CHECKING(if gethostbyaddr_r takes 5 arguments)
-    AC_TRY_COMPILE([
-#include <sys/types.h>
-#include <netdb.h>],[
-char * address;
-int length;
-int type;
-struct hostent h;
-struct hostent_data hdata;
-int rc;
-rc = gethostbyaddr_r(address, length, type, &h, &hdata);],[
-      AC_MSG_RESULT(yes)
-      AC_DEFINE(HAVE_GETHOSTBYADDR_R_5)
-      ac_cv_gethostbyaddr_args=5],[
-      AC_MSG_RESULT(no)
-      AC_MSG_CHECKING(if gethostbyaddr_r with -D_REENTRANT takes 5 arguments)
-      AC_TRY_COMPILE([
-#define _REENTRANT
-#include <sys/types.h>
-#include <netdb.h>],[
-char * address;
-int length;
-int type;
-struct hostent h;
-struct hostent_data hdata;
-int rc;
-rc = gethostbyaddr_r(address, length, type, &h, &hdata);],[
-	AC_MSG_RESULT(yes)
-	AC_DEFINE(HAVE_GETHOSTBYADDR_R_5)
-	AC_DEFINE(NEED_REENTRANT)
-	ac_cv_gethostbyaddr_args=5],[
-	AC_MSG_RESULT(no)
-	AC_MSG_CHECKING(if gethostbyaddr_r takes 7 arguments)
-	AC_TRY_COMPILE([
-#include <sys/types.h>
-#include <netdb.h>],[
-char * address;
-int length;
-int type;
-struct hostent h;
-char buffer[10];
-int buflen;
-int h_errnop;
-struct hostent * hp;
+])
 
-hp = gethostbyaddr_r(address, length, type, &h,
-                     buffer, buflen, &h_errnop);],[
+
+dnl **********************************************************************
+dnl Checks for libraries.
+dnl **********************************************************************
+
+dnl gethostbyname in the nsl lib?
+AC_CHECK_FUNC(gethostbyname, , AC_CHECK_LIB(nsl, gethostbyname))
+
+if test "$ac_cv_lib_nsl_gethostbyname" != "yes" -a "$ac_cv_func_gethostbyname" != "yes"; then
+  dnl gethostbyname in the socket lib?
+  AC_CHECK_FUNC(gethostbyname, , AC_CHECK_LIB(socket, gethostbyname))
+fi
+
+dnl At least one system has been identified to require BOTH nsl and
+dnl socket libs to link properly.
+if test "$ac_cv_lib_nsl_gethostbyname" = "$ac_cv_func_gethostbyname"; then
+  AC_MSG_CHECKING([trying both nsl and socket libs])
+  my_ac_save_LIBS=$LIBS
+  LIBS="-lnsl -lsocket $LIBS"
+  AC_TRY_LINK( ,
+             [gethostbyname();],
+             my_ac_link_result=success,
+             my_ac_link_result=failure )
+
+  if test "$my_ac_link_result" = "failure"; then
+    AC_MSG_RESULT([no])
+    AC_MSG_ERROR([couldn't find libraries for gethostbyname()])
+    dnl restore LIBS
+    LIBS=$my_ac_save_LIBS
+  else
+    AC_MSG_RESULT([yes])
+  fi
+fi
+
+dnl resolve lib?
+AC_CHECK_FUNC(strcasecmp, , AC_CHECK_LIB(resolve, strcasecmp))
+
+if test "$ac_cv_lib_resolve_strcasecmp" = "$ac_cv_func_strcasecmp"; then
+  AC_CHECK_LIB(resolve, strcasecmp,
+              [LIBS="-lresolve $LIBS"],
+               ,
+               -lnsl)
+fi
+
+dnl socket lib?
+AC_CHECK_FUNC(connect, , AC_CHECK_LIB(socket, connect))
+
+dnl ucb lib?
+AC_CHECK_FUNC(gethostname, , AC_CHECK_LIB(ucb, gethostname))
+
+dnl dl lib?
+AC_CHECK_FUNC(dlopen, , AC_CHECK_LIB(dl, dlopen))
+
+dnl **********************************************************************
+dnl Check for the presence of Kerberos4 libraries and headers
+dnl **********************************************************************
+
+AC_ARG_WITH(krb4-includes,
+ [  --with-krb4-includes[=DIR]   Specify location of kerberos4 headers],[
+ CPPFLAGS="$CPPFLAGS -I$withval"
+ KRB4INC="$withval"
+ want_krb4=yes
+ ])
+
+AC_ARG_WITH(krb4-libs,
+ [  --with-krb4-libs[=DIR]   Specify location of kerberos4 libs],[
+ LDFLAGS="$LDFLAGS -L$withval"
+ KRB4LIB="$withval"
+ want_krb4=yes
+ ])
+
+
+OPT_KRB4=off
+AC_ARG_WITH(krb4,dnl
+[  --with-krb4[=DIR]       where to look for Kerberos4],[
+  OPT_KRB4="$withval"
+  if test X"$OPT_KRB4" != Xyes
+  then
+    LDFLAGS="$LDFLAGS -L$OPT_KRB4/lib"
+    KRB4LIB="$OPT_KRB4/lib"
+    CPPFLAGS="$CPPFLAGS -I$OPT_KRB4/include"
+    KRB4INC="$OPT_KRB4/include"
+  fi
+  want_krb4="yes"
+ ])
+
+AC_MSG_CHECKING([if Kerberos4 support is requested])
+
+if test "$want_krb4" = yes
+then
   AC_MSG_RESULT(yes)
-	  AC_DEFINE(HAVE_GETHOSTBYADDR_R_7)
-	  ac_cv_gethostbyaddr_args=7],[
-	  AC_MSG_RESULT(no)
-	  AC_MSG_CHECKING(if gethostbyaddr_r takes 8 arguments)
-	  AC_TRY_COMPILE([
-#include <sys/types.h>
-#include <netdb.h>],[
-char * address;
-int length;
-int type;
-struct hostent h;
-char buffer[10];
-int buflen;
-int h_errnop;
-struct hostent * hp;
-int rc;
 
-rc = gethostbyaddr_r(address, length, type, &h,
-                     buffer, buflen, &hp, &h_errnop);],[
+  dnl Check for & handle argument to --with-krb4
+
+  AC_MSG_CHECKING(where to look for Kerberos4)
+  if test X"$OPT_KRB4" = Xyes
+  then
+    AC_MSG_RESULT([defaults])
+  else
+    AC_MSG_RESULT([libs in $KRB4LIB, headers in $KRB4INC])
+  fi
+
+  dnl Check for DES library
+  AC_CHECK_LIB(des, des_pcbc_encrypt,
+  [
+    AC_CHECK_HEADERS(des.h)
+
+    dnl resolv lib?
+    AC_CHECK_FUNC(res_search, , AC_CHECK_LIB(resolv, res_search))
+
+    dnl Check for the Kerberos4 library
+    AC_CHECK_LIB(krb, krb_net_read,
+    [
+      dnl Check for header files
+      AC_CHECK_HEADERS(krb.h)
+
+      dnl we found the required libraries, add to LIBS
+      LIBS="-lkrb -ldes $LIBS"
+
+      dnl Check for function krb_get_our_ip_for_realm
+      dnl this is needed for NAT networks
+      AC_CHECK_FUNCS(krb_get_our_ip_for_realm)
+
+      dnl add define KRB4
+      AC_DEFINE(KRB4)
+
+      dnl the krb4 stuff needs a strlcpy()
+      AC_CHECK_FUNCS(strlcpy)
+
+    ])
+  ])
+else
+  AC_MSG_RESULT(no)
+fi
+
+dnl **********************************************************************
+dnl Check for the presence of SSL libraries and headers
+dnl **********************************************************************
+
+dnl Default to compiler & linker defaults for SSL files & libraries.
+OPT_SSL=off
+AC_ARG_WITH(ssl,dnl
+[  --with-ssl[=DIR]        where to look for SSL [compiler/linker default paths]
+                          DIR points to the SSL installation [/usr/local/ssl]],
+  OPT_SSL=$withval
+)
+
+if test X"$OPT_SSL" = Xno
+then
+  AC_MSG_WARN(SSL/https support disabled)  
+else
+
+  dnl Check for & handle argument to --with-ssl.
+
+  AC_MSG_CHECKING(where to look for SSL)
+  if test X"$OPT_SSL" = Xoff
+  then
+  	AC_MSG_RESULT([defaults (or given in environment)])
+  else
+	test X"$OPT_SSL" = Xyes && OPT_SSL=/usr/local/ssl
+	LIBS="$LIBS -L$OPT_SSL/lib"
+	CPPFLAGS="$CPPFLAGS -I$OPT_SSL/include/openssl -I$OPT_SSL/include"
+	AC_MSG_RESULT([$OPT_SSL])
+  fi
+
+  dnl check for crypto libs (part of SSLeay)
+  AC_CHECK_LIB(crypto, CRYPTO_lock)
+
+  if test $ac_cv_lib_crypto_CRYPTO_lock = yes; then
+    dnl This is only reasonable to do if crypto actually is there: check for
+    dnl SSL libs NOTE: it is important to do this AFTER the crypto lib
+    AC_CHECK_LIB(ssl, SSL_connect)
+
+    if test "$ac_cv_lib_ssl_SSL_connect" != yes; then
+        dnl we didn't find the SSL lib, try the RSAglue/rsaref stuff
+        AC_MSG_CHECKING(for ssl with RSAglue/rsaref libs in use);
+        OLIBS=$LIBS
+        LIBS="$LIBS -lRSAglue -lrsaref"
+        AC_CHECK_LIB(ssl, SSL_connect)
+        if test "$ac_cv_lib_ssl_SSL_connect" != yes; then
+            dnl still no SSL_connect
+            AC_MSG_RESULT(no)
+            LIBS=$OLIBS
+        else
             AC_MSG_RESULT(yes)
-	    AC_DEFINE(HAVE_GETHOSTBYADDR_R_8)
-	    ac_cv_gethostbyaddr_args=8],[
-	    AC_MSG_RESULT(no)
-	    have_missing_r_funcs="$have_missing_r_funcs gethostbyaddr_r"])])])])])
+        fi
+    fi
 
-  dnl determine if function definition for inet_ntoa_r exists.
-  AC_CHECK_FUNCS(inet_ntoa_r,[
-    AC_MSG_CHECKING(whether inet_ntoa_r is declared)
-    AC_EGREP_CPP(inet_ntoa_r,[
-#include <arpa/inet.h>],[
-      AC_DEFINE(HAVE_INET_NTOA_R_DECL)
-      AC_MSG_RESULT(yes)],[
-      AC_MSG_RESULT(no)
-      AC_MSG_CHECKING(whether inet_ntoa_r with -D_REENTRANT is declared)
-      AC_EGREP_CPP(inet_ntoa_r,[
-#define _REENTRANT
-#include <arpa/inet.h>],[
-	AC_DEFINE(HAVE_INET_NTOA_R_DECL)
-	AC_DEFINE(NEED_REENTRANT)
-	AC_MSG_RESULT(yes)],
-	AC_MSG_RESULT(no))])])
 
-  dnl check for a few thread-safe functions
-  AC_CHECK_FUNCS(localtime_r,[
-    AC_MSG_CHECKING(whether localtime_r is declared)
-    AC_EGREP_CPP(localtime_r,[
-#include <time.h>],[
-      AC_MSG_RESULT(yes)],[
-      AC_MSG_RESULT(no)
-      AC_MSG_CHECKING(whether localtime_r with -D_REENTRANT is declared)
-      AC_EGREP_CPP(localtime_r,[
-#define _REENTRANT
-#include <time.h>],[
-	AC_DEFINE(NEED_REENTRANT)
-	AC_MSG_RESULT(yes)],
-	AC_MSG_RESULT(no))])])
+
+    dnl Check for SSLeay headers
+    AC_CHECK_HEADERS(openssl/x509.h openssl/rsa.h openssl/crypto.h openssl/pem.h openssl/ssl.h openssl/err.h)
+
+    if test $ac_cv_header_openssl_x509_h = no; then
+      AC_CHECK_HEADERS(x509.h rsa.h crypto.h pem.h ssl.h err.h)
+    fi
+
+  fi
+
+  dnl these can only exist if openssl exists
+
+  AC_CHECK_FUNCS( RAND_status \
+                  RAND_screen )
+
+fi
+
+dnl **********************************************************************
+dnl Check for the presence of ZLIB libraries and headers
+dnl **********************************************************************
+
+dnl Default to compiler & linker defaults for files & libraries.
+dnl OPT_ZLIB=no
+dnl AC_ARG_WITH(zlib,dnl
+dnl [  --with-zlib[=DIR]  where to look for ZLIB [compiler/linker default paths]
+dnl                      DIR points to the ZLIB installation prefix [/usr/local]],
+dnl  OPT_ZLIB=$withval,
+dnl )
+
+dnl Check for & handle argument to --with-zlib.
+dnl
+dnl NOTE:  We *always* look for ZLIB headers & libraries, all this option
+dnl        does is change where we look (by adjusting LIBS and CPPFLAGS.)
+dnl
+
+dnl AC_MSG_CHECKING(where to look for ZLIB)
+dnl if test X"$OPT_ZLIB" = Xno
+dnl then
+dnl 	AC_MSG_RESULT([defaults (or given in environment)])
+dnl else
+dnl	test X"$OPT_ZLIB" = Xyes && OPT_ZLIB=/usr/local
+dnl	LIBS="$LIBS -L$OPT_ZLIB/lib"
+dnl	CPPFLAGS="$CPPFLAGS -I$OPT_ZLIB/include"
+dnl	AC_MSG_RESULT([$OPT_ZLIB])
+dnl fi
+
+dnl z lib?
+dnl AC_CHECK_FUNC(gzread, , AC_CHECK_LIB(z, gzread))
+
+
+dnl Default is to try the thread-safe versions of a few functions
+OPT_THREAD=on
+AC_ARG_ENABLE(thread,dnl
+[  --disable-thread       tell configure to not look for thread-safe functions],
+  OPT_THREAD=off
+)
+
+if test X"$OPT_THREAD" = Xoff
+then
+  AC_MSG_WARN(libcurl will not get built using thread-safe functions)
+  AC_DEFINE(DISABLED_THREADSAFE, 1, \
+Set to explicitly specify we don't want to use thread-safe functions)
+else
+
+  dnl dig around for gethostbyname_r()
+  CURL_CHECK_GETHOSTBYNAME_R()
+
+  dnl dig around for gethostbyaddr_r()
+  CURL_CHECK_GETHOSTBYADDR_R()
+
+  dnl poke around for inet_ntoa_r()
+  CURL_CHECK_INET_NTOA_R()
+
+  dnl is there a localtime_r()
+  CURL_CHECK_LOCALTIME_R()
+
 fi
 
 dnl **********************************************************************
@@ -439,13 +579,15 @@ AC_CHECK_FUNCS( socket \
                 tcsetattr \
                 tcgetattr \
                 perror \
-                getpass \
                 closesocket \
                 setvbuf \
-                RAND_status \
-                RAND_screen
+                sigaction \
+                signal \
+                getpass_r
 )
 
+dnl removed 'getpass' check on October 26, 2000
+
 if test "$ac_cv_func_select" != "yes"; then
   AC_MSG_ERROR(Can't work without an existing socket() function)
 fi
@@ -465,13 +607,12 @@ dnl   $PATH:/usr/bin/:/usr/local/bin )
 dnl AC_SUBST(RANLIB)
 
 AC_OUTPUT( Makefile \
-           curl.spec \
-           curl-ssl.spec \
 	   docs/Makefile \
 	   include/Makefile \
 	   include/curl/Makefile \
 	   src/Makefile \
-           lib/Makefile )
+           lib/Makefile \
+           tests/Makefile)
 dnl	   perl/checklinks.pl \
 dnl	   perl/getlinks.pl \
 dnl        perl/formfind.pl \

docs/CONTRIBUTE

@@ -20,7 +20,10 @@ The License Issue
  GNU Public License. We can never re-use sources from a GPL program in curl.
  If you add a larger piece of code, you can opt to make that file or set of
  files to use a different license as long as they don't enfore any changes to
- the rest of the package. Such "separate parts" can not be GPL either.
+ the rest of the package and they make sense. Such "separate parts" can not be
+ GPL either (although they should use "GPL compatible" licenses).
+
+ Curl and libcurl will soon become dual licensed, MozPL/MITX!
 
 Naming
 
@@ -82,3 +85,12 @@ Write Access to CVS Repository
  course get write access to the CVS repository and then you'll be able to
  check-in all your changes straight into the CVS tree instead of sending all
  changes by mail as patches. Just ask if this is what you'd want.
+
+Test Cases
+
+ Since the introduction of the test suite, we will get the possibility to
+ quickly verify that the main features are working as supposed to. To maintain
+ this situation and improve it, all new features and functions that are added
+ need tro be tested. Every feature that is added should get at least one valid
+ test case that verifies that it works as documented. If every submitter also
+ post a few test cases, it won't end up as a heavy burden on a single person!

docs/FAQ

@@ -1,4 +1,4 @@
-Updated: August 22, 2000 (http://curl.haxx.se/docs/faq.shtml)
+Updated: November 22, 2000 (http://curl.haxx.se/docs/faq.shtml)
                                   _   _ ____  _     
                               ___| | | |  _ \| |    
                              / __| | | | |_) | |    
@@ -43,9 +43,12 @@ FAQ
    4.5.4 "404 Not Found"
    4.5.5 "405 Method Not Allowed"
   4.6 Can you tell me what error code 142 means?
+  4.7 How do I keep usernames and passwords secret in Curl command lines?
+  4.8 I found a bug!
 
  5. libcurl Issues
   5.1 Is libcurl thread safe?
+  5.2 How can I receive all data into a large memory chunk?
 
  6. License Issues
   6.1 I have a GPL program, can I use the libcurl library?
@@ -63,7 +66,8 @@ FAQ
 
   cURL (or simply just 'curl') is a command line tool for getting or sending
   files using URL syntax. The name is a play on 'Client for URLs', originally
-  with URL spelled in uppercase to make it obvious it deals with URLs.
+  with URL spelled in uppercase to make it obvious it deals with URLs. The
+  fact it can also be pronounced 'see URL' also helped.
 
   Curl supports a range of common internet protocols, currently including
   HTTP, HTTPS, FTP, GOPHER, LDAP, DICT and FILE.
@@ -76,7 +80,7 @@ FAQ
   transfer library.
 
   Any application is free to use libcurl, even commercial or closed-source
-  ones. Just make sure changes to the lib itself is made public.
+  ones. Just make sure changes to the lib itself are made public.
 
   1.3 What is cURL not?
   
@@ -97,7 +101,7 @@ FAQ
   or with PHP.
 
   Curl is not a single-OS program. Curl exists, compiles, builds and runs
-  under a long range of operating systems, including all modern Unixes,
+  under a wide range of operating systems, including all modern Unixes,
   Windows, Amiga, BeOS, OS/2, OS X, QNX etc.
 
   1.4 When will you make curl do XXXX ?
@@ -117,7 +121,7 @@ FAQ
     program or redirect to another file for the next program to interpret.
 
   * I focus on protocol related issues and improvements. If you wanna do more
-    magic with the supported protocols than curl currently does, changes are
+    magic with the supported protocols than curl currently does, chances are
     big I will agree. If you wanna add more protocols, I may very well
     agree.
 
@@ -141,6 +145,7 @@ FAQ
       Solaris (native cc compiler)
       HPUX (native cc compiler)
       SGI IRIX (native cc compiler)
+      SCO UNIX (native cc compiler)
 
     When configuring curl, I specify --with-ssl. OpenSSL is installed in
     /usr/local/ssl Configure reports SSL in /usr/local/ssl, but fails to find
@@ -193,7 +198,6 @@ FAQ
   brings this functionality.
 
 
-
 3. Usage problems
 
   3.1. curl: (1) SSL is disabled, https: not supported
@@ -352,9 +356,9 @@ FAQ
   4.6. Can you tell me what error code 142 means?
 
   All error codes that are larger than the highest documented error code means
-  that curl has existed due to a timeout. There is currentl no nice way for
+  that curl has existed due to a timeout. There is currently no nice way for
   curl to abort from such a condition and that's why it gets this undocumented
-  error. This is planned to change in a future release.
+  error. This should be changed in releases after 7.4.1.
 
   4.7. How do I keep usernames and passwords secret in Curl command lines?
 
@@ -370,27 +374,77 @@ FAQ
   at least hide them from being read by human eyes, but that is not what
   anyone would call security.
 
+  4.8 I found a bug!
+
+  It is not a bug if the behaviour is documented. Read the docs first.
+
+  If it is a problem with a binary you've downloaded or a package for your
+  particular platform, try contacting the person who built the package/archive
+  you have.
+
+  If there is a bug, post a bug report in the Curl Bug Track System over at
+  http://sourceforge.net/bugs/?group_id=976 or mail a detailed bug description
+  to curl-bug@haxx.se.
+
+  Always include as many details you can think of, including curl version,
+  operating system name and version and complete instructions how to repeat
+  the bug.
+
 
 5. libcurl Issues
 
   5.1. Is libcurl thread safe?
 
-  As version seven is slowly marching in as the libcurl version to use, we
-  have made a serious attempt to address all places in the code where we could
-  forsee problems for multi-threaded programs. If your system has them, curl
-  will attempt to use threadsafe functions instead of non-safe ones.
+  We have attempted to write the entire code adjusted for multi-threaded
+  programs. If your system has such, curl will attempt to use threadsafe
+  functions instead of non-safe ones.
 
   I am very interested in once and for all getting some kind of report or
   README file from those who have used libcurl in a threaded environment,
   since I haven't and I get this question more and more frequently!
 
+  5.2 How can I receive all data into a large memory chunk?
+
+  You are in full control of the callback function that gets called every time
+  there is data received from the remote server. You can make that callback do
+  whatever you want. You do not have to write the receivied data to a file.
+
+  One solution to this problem could be to have a pointer to a struct that you
+  pass to the callback function. You set the pointer using the
+  curl_easy_setopt(CURLOPT_FILE) function. Then that pointer will be passed to
+  the callback instead of a FILE * to a file:
+
+        /* imaginary struct */
+        struct MemoryStruct {
+          char *memory;
+          size_t size;
+        };
+
+        /* imaginary callback function */
+        size_t
+        WriteMemoryCallback(void *ptr, size_t size, size_t nmemb, void *data)
+        {
+          register int realsize = size * nmemb;
+          struct MemoryStruct *mem = (struct MemoryStruct *)data;
+        
+          mem->memory = (char *)realloc(mem->memory, mem->size + realsize + 1);
+          if (mem->memory) {
+            memcpy(&(mem->memory[mem->size]), ptr, realsize);
+            mem->size += realsize;
+            mem->memory[mem->size] = 0;
+          }
+          return realsize;
+        }
+
+
 6. License Issues
 
   Curl and libcurl are released under the MPL, the Mozilla Public License. To
-  get a really good answer to this or other licensing questions, you should
+  get a really good answer to your license conflict questions, you should
   study the MPL license and the license you are about to use and check for
-  clashes yourself. This is a brief summary for the cases we get the most
-  questions. (Parts of this section was enhanced by Bjorn Reese.)
+  clashes yourself. This section is just a brief summary for the cases we get
+  the most questions. (Parts of this section was much enhanced by Bjorn
+  Reese.)
 
   6.1. I have a GPL program, can I use the libcurl library?
 
@@ -427,19 +481,19 @@ FAQ
   6.2. I have a closed-source program, can I use the libcurl library?
 
   Yes, libcurl does not put any restrictions on the program that uses the
-    library. If you end up doing changes to the library, only those changes
-    must be made available, not the ones to your program.
+  library. If you end up doing changes to the library, only those changes must
+  be made available, not the ones to your program.
 
   6.3. I have a BSD licensed program, can I use the libcurl library?
 
   Yes, libcurl does not put any restrictions on the program that uses the
-    library. If you end up doing changes to the library, only those changes
-    must be made available, not the ones to your program.
+  library. If you end up doing changes to the library, only those changes must
+  be made available, not the ones to your program.
 
   6.4. I have a program that uses LGPL libraries, can I use libcurl?
 
-    Yes you can. LGPL libraries don't spread to other libraries the same way
-    GPL ones do.
+  Yes you can. LGPL libraries don't spread to other libraries the same way GPL
+  ones do.
 
   However, when you read paragraph (3) of the LGPL license, you'll see that
   anyone - at will - may at any time convert that LGPL program into GPL. And
@@ -452,7 +506,7 @@ FAQ
 
   6.6. Can you please change the curl/libcurl license to XXXX?
 
-    No. We carefully picked this license years ago and a large amount of
-    people have contributed with source code knowing that this is the license
-    we use. This license puts the restrictions we want on curl/libcurl and it
-    does not spread to other programs or libraries that use it.
+  No. We carefully picked this license years ago and a large amount of people
+  have contributed with source code knowing that this is the license we
+  use. This license puts the restrictions we want on curl/libcurl and it does
+  not spread to other programs or libraries that use it.

docs/FEATURES

@@ -30,7 +30,7 @@ HTTP
  - follow redirects
  - custom HTTP request
  - cookie get/send
- - understands the netscape cookie file
+ - understands the netscape cookie file format
  - custom headers (that can replace/remove internally generated headers)
  - custom user-agent string
  - custom referer string
@@ -38,6 +38,7 @@ HTTP
  - proxy authentication
  - time conditions
  - via http-proxy
+ - specify interface device/port
 
 HTTPS (*1)
  - (all the HTTP features)
@@ -47,6 +48,7 @@ HTTPS (*1)
 FTP
  - download
  - authentication
+ - kerberos security
  - PORT or PASV
  - single file size information (compare to HTTP HEAD)
  - 'type=' URL support
@@ -60,6 +62,7 @@ FTP
  - custom ftp commands (before and/or after the transfer)
  - simple "range" support
  - via http-proxy
+ - all operations can be tunneled through a http-proxy
 
 TELNET
  - connection negotiation

docs/INSTALL

@@ -11,35 +11,8 @@ way to proceed is mainly divided in two different ways: the unix way or the
 windows way.
 
 If you're using Windows (95, 98, NT) or OS/2, you should continue reading from
-the Win32 header below. All other systems should be capable of being installed
-as described in the the UNIX header.
-
-PORTS
-=====
-   Just to show off, this is a probably incomplete list of known hardware and
-   operating systems that curl has been compiled for:
-
-        - Ultrix
-        - SINIX-Z v5
-        Alpha DEC OSF 4
-        Alpha Digital UNIX V3.2D-1 (rev 41)
-        HP-PA HP-UX 10.X 11.X
-        MIPS IRIX 6.2, 6.5
-        Power AIX 4.2, 4.3.1, 4.3.2
-        PowerPC Darwin 1.0
-        PowerPC Mac OS X
-        Sparc Solaris 2.4, 2.5, 2.5.1, 2.6, 7
-        Sparc SunOS 4.1.*
-        i386 BeOS
-        i386 FreeBSD
-        i386 Linux 1.3, 2.0, 2.2
-        i386 NetBSD
-        i386 OS/2
-        i386 OpenBSD
-        i386 Solaris 2.7
-        i386 Windows 95, 98, NT
-        m68k AmigaOS 3
-        m68k OpenBSD
+the Win32 or OS/2 headers further down. All other systems should be capable of
+being installed as described below.
 
 UNIX
 ====
@@ -47,7 +20,9 @@ UNIX
    The configure script *always* tries to find a working SSL library unless
    explicitly told not to.  If you have OpenSSL installed in the default
    search path for your compiler/linker, you don't need to do anything
-   special.
+   special:
+
+        ./configure
 
    If you have OpenSSL installed in /usr/local/ssl, you can run configure
    like:
@@ -95,9 +70,17 @@ UNIX
 
    Use the executable `curl` in src/ directory.
 
-   'make install' copies the curl file to /usr/local/bin/ (or $prefix/bin
-   if you used the --prefix option to configure) and copies the curl.1
-   man page to a suitable place too.
+   To install curl on your system, run
+
+     make install
+
+   This will copy curl to /usr/local/bin/ (or $prefix/bin if you used the
+   --prefix option to configure) and it copies the man pages, the lib and the
+   include files to suitable places.
+
+   To make sure everything runs as supposed, run the test suite:
+
+     make test
 
    KNOWN PROBLEMS
 
@@ -119,6 +102,27 @@ UNIX
          or
        env Cc=cc ./configure
 
+     To force a static library compile, disable the shared library creation
+     by running configure like:
+
+       ./configure --disable-shared
+
+     To tell the configure script to skip searching for thread-safe functions,
+     add an option like:
+
+       ./configure --disable-thread
+
+     To build curl with kerberos4 support enabled, curl requires the krb4 libs
+     and headers installed. You can then use a set of options to tell
+     configure where those are:
+
+          --with-krb4-includes[=DIR]   Specify location of kerberos4 headers
+          --with-krb4-libs[=DIR]       Specify location of kerberos4 libs
+          --with-krb4[=DIR]            where to look for Kerberos4
+
+     In most cases, /usr/athena is the install prefix and then it works with
+
+       ./configure --with-krb4=/usr/athena
 
 Win32
 =====
@@ -232,6 +236,39 @@ IBM OS/2
    If you're getting huge binaries, probably your makefiles have the -g in
    CFLAGS.
 
+PORTS
+=====
+   Just to show off, this is a probably incomplete list of known hardware and
+   operating systems that curl has been compiled for:
+
+        - Ultrix
+        - SINIX-Z v5
+        - Alpha DEC OSF 4
+        - Alpha Digital UNIX v3.2
+        - Alpha FreeBSD 4.1
+        - Alpha Linux 2.2.16
+        - Alpha Tru64 v5.0 5.1
+        - HP-PA HP-UX 9.X 10.X 11.X
+        - MIPS IRIX 6.2, 6.5
+        - Power AIX 4.2, 4.3.1, 4.3.2
+        - PowerPC Darwin 1.0
+        - PowerPC Linux
+        - PowerPC Mac OS X
+        - Sparc Linux
+        - Sparc Solaris 2.4, 2.5, 2.5.1, 2.6, 7, 8
+        - Sparc SunOS 4.1.*
+        - i386 BeOS
+        - i386 FreeBSD
+        - i386 Linux 1.3, 2.0, 2.2, 2.3, 2.4
+        - i386 NetBSD
+        - i386 OS/2
+        - i386 OpenBSD
+        - i386 Solaris 2.7
+        - i386 Windows 95, 98, NT, 2000
+        - ia64 Linux 2.3.99
+        - m68k AmigaOS 3
+        - m68k OpenBSD
+
 OpenSSL
 =======
 
@@ -245,7 +282,7 @@ MingW32/Cygwin
 
    You'll find MingW32 and Cygwin information at:
 
-      http://www.xraylith.wisc.edu/~khan/software/gnu-win32/index.html
+      http://www.mingw.org
 
 OpenLDAP
 ========

docs/INTERNALS

@@ -12,6 +12,17 @@ INTERNALS
 
  Thus, the largest amount of code and complexity is in the library part.
 
+CVS
+===
+
+ All changes to the sources are committed to the CVS repository as soon as
+ they're somewhat verified to work. Changes shall be commited as independently
+ as possible so that individual changes can be easier spotted and tracked
+ afterwards.
+
+ Tagging shall be used extensively, and by the time we release new archives we
+ should tag the sources with a name similar to the released version number.
+
 Windows vs Unix
 ===============
 
@@ -34,8 +45,7 @@ Windows vs Unix
 
  (3) is simply avoided by not trying any funny tricks on file descriptors.
 
- (4) is left alone, giving windows users problems when they pipe binary data
- through stdout...
+ (4) we set stdout to binary under windows
 
  Inside the source code, I do make an effort to avoid '#ifdef WIN32'. All
  conditionals that deal with features *should* instead be in the format
@@ -48,9 +58,9 @@ Library
 =======
 
  As described elsewhere, libcurl is meant to get two different "layers" of
- interface. At the present point only the high-level, the "easy", interface
- has been fully implemented and thus documented. We assume the easy-interface
- in this description, the low-level interface will be documented when fully
+ interfaces. At the present point only the high-level, the "easy", interface
+ has been fully implemented and documented. We assume the easy-interface in
+ this description, the low-level interface will be documented when fully
  implemented.
 
  There are plenty of entry points to the library, namely each publicly defined
@@ -58,11 +68,14 @@ Library
  rather small and easy-to-follow. All the ones prefixed with 'curl_easy' are
  put in the lib/easy.c file.
 
+ curl_easy_init() allocates an internal struct and makes some initializations.
+ The returned handle does not revail internals.
+
  curl_easy_setopt() takes a three arguments, where the option stuff must be
  passed in pairs, the parameter-ID and the parameter-value. The list of
  options is documented in the man page.
 
- curl_easy_perform() does a whole lot of things.
+ curl_easy_perform() does a whole lot of things:
 
  The function analyzes the URL, get the different components and connects to
  the remote host. This may involve using a proxy and/or using SSL. The
@@ -84,10 +97,6 @@ Library
  called). The speedcheck functions in lib/speedcheck.c are also used to verify
  that the transfer is as fast as required.
 
- When the operation is done, the writeout() function in lib/writeout.c may be
- called to report about the operation as specified previously in the arguments
- to curl_easy_setopt().
-
  When completed curl_easy_cleanup() should be called to free up used
  resources.
 
@@ -136,11 +145,12 @@ Library
 
  lib/getenv.c offers curl_getenv() which is for reading environment variables
  in a neat platform independent way. That's used in the client, but also in
- lib/url.c when checking the PROXY variables.
+ lib/url.c when checking the proxy environment variables.
 
- lib/netrc.c keeps the .netrc parser
+ lib/netrc.c holds the .netrc parser
 
  lib/timeval.c features replacement functions for systems that don't have
+ gettimeofday().
  
  A function named curl_version() that returns the full curl version string is
  found in lib/version.c.
@@ -148,13 +158,31 @@ Library
 Client
 ======
 
- main() resides in src/main.c together with most of the client
- code. src/hugehelp.c is automatically generated by the mkhelp.pl perl script
- to display the complete "manual" and the src/urlglob.c file holds the
- functions used for the multiple-URL support.
+ main() resides in src/main.c together with most of the client code.
+ src/hugehelp.c is automatically generated by the mkhelp.pl perl script to
+ display the complete "manual" and the src/urlglob.c file holds the functions
+ used for the multiple-URL support.
 
  The client mostly mess around to setup its config struct properly, then it
  calls the curl_easy_*() functions of the library and when it gets back
  control after the curl_easy_perform() it cleans up the library, checks status
  and exits.
 
+ When the operation is done, the ourWriteOut() function in src/writeout.c may
+ be called to report about the operation. That function is using the
+ curl_easy_getinfo() function to extract useful information from the curl
+ session.
+
+Test Suite
+==========
+
+ During November 2000, a test suite has evolved. It is placed in its own
+ subdirectory directly off the root in the curl archive tree, and it contains
+ a bunch of scripts and a lot of test case data.
+
+ The main test script is runtests.pl that will invoke the two servers
+ httpserver.pl and ftpserver.pl before all the test cases are performed. The
+ test suite currently only runs on unix-like platforms.
+
+ You'll find a complete description of the test case data files in the README
+ file in the test directory.

docs/MANUAL

@@ -139,9 +139,11 @@ UPLOADING
 
         curl -T localfile -a ftp://ftp.upload.com/remotefile
 
-  NOTE: Curl does not support ftp upload through a proxy! The reason for this
-  is simply that proxies are seldomly configured to allow this and that no
-  author has supplied code that makes it possible!
+  Curl also supports ftp upload through a proxy, but only if the proxy is
+  configured to allow that kind of tunneling. If it does, you can run curl in
+  a fashion similar to:
+
+        curl --proxytunnel -x proxy:port -T localfile ftp.upload.com
 
  HTTP
 
@@ -280,6 +282,8 @@ REFERER
 
         curl -e www.coolsite.com http://www.showme.com/
 
+  NOTE: The referer field is defined in the HTTP spec to be a full URL.
+
 USER AGENT
 
   A HTTP request has the option to include information about the browser
@@ -398,17 +402,26 @@ SPEED LIMIT
 CONFIG FILE
 
   Curl automatically tries to read the .curlrc file (or _curlrc file on win32
-  systems) from the user's home dir on startup. The config file should be
-  made up with normal command line switches. Comments can be used within the
-  file. If the first letter on a line is a '#'-letter the rest of the line
-  is treated as a comment.
+  systems) from the user's home dir on startup.
+
+  The config file could be made up with normal command line switches, but you
+  can also specify the long options without the dashes to make it more
+  readable. You can separate the options and the parameter with spaces, or
+  with = or :. Comments can be used within the file. If the first letter on a
+  line is a '#'-letter the rest of the line is treated as a comment.
+
+  If you want the parameter to contain spaces, you must inclose the entire
+  parameter within double quotes ("). Within those quotes, you specify a
+  quote as \".
+
+  NOTE: You must specify options and their arguments on the same line.
 
   Example, set default time out and proxy in a config file:
 
         # We want a 30 minute timeout:
         -m 1800
         # ... and we use a proxy for all accesses:
-        -x proxy.our.domain.com:8080
+        proxy = proxy.our.domain.com:8080
 
   White spaces ARE significant at the end of lines, but all white spaces
   leading up to the first characters of each line are ignored.
@@ -422,14 +435,14 @@ CONFIG FILE
   without URL by making a config file similar to:
 
         # default url to get
-        http://help.with.curl.com/curlhelp.html
+        url = "http://help.with.curl.com/curlhelp.html"
 
   You can specify another config file to be read by using the -K/--config
   flag. If you set config file name to "-" it'll read the config from stdin,
   which can be handy if you want to hide options from being visible in process
   tables etc:
 
-        echo "-u user:passwd" | curl -K - http://that.secret.site.com
+        echo "user = user:passwd" | curl -K - http://that.secret.site.com
 
 EXTRA HEADERS
 
@@ -480,13 +493,14 @@ FTP and firewalls
   connect to the client on the given (as parameters to the PORT command) IP
   number and port.
 
-  The -P flag to curl allows for different options. Your machine may have
+  The -P flag to curl supports a few different options. Your machine may have
   several IP-addresses and/or network interfaces and curl allows you to select
   which of them to use. Default address can also be used:
 
         curl -P - ftp.download.com
 
-  Download with PORT but use the IP address of our 'le0' interface:
+  Download with PORT but use the IP address of our 'le0' interface (this does
+  not work on windows):
 
         curl -P le0 ftp.download.com
 
@@ -494,6 +508,16 @@ FTP and firewalls
 
         curl -P 192.168.0.10 ftp.download.com
 
+NETWORK INTERFACE
+
+  Get a web page from a server using a specified port for the interface:
+
+	curl --interface eth0:1 http://www.netscape.com/
+
+  or
+
+	curl --interface 192.168.1.10 http://www.netscape.com/
+
 HTTPS
 
   Secure HTTP requires SSL libraries to be installed and used when curl is
@@ -689,13 +713,26 @@ CUSTOM OUTPUT
 
         curl -w 'We downloaded %{size_download} bytes\n' www.download.com
 
+KERBEROS4 FTP TRANSFER
+
+  Curl supports kerberos4 for FTP transfers. You need the kerberos package
+  installed and used at curl build time for it to be used.
+
+  First, get the krb-ticket the normal way, like with the kauth tool. Then use
+  curl in way similar to:
+
+        curl --krb4 private ftp://krb4site.com -u username:fakepwd
+
+  There's no use for a password on the -u switch, but a blank one will make
+  curl ask for one and you already entered the real password to kauth.
+
 MAILING LIST
 
   We have an open mailing list to discuss curl, its development and things
   relevant to this.
 
-  To subscribe, mail curl-request@contactor.se with "subscribe <your email
-  address>" in the body.
+  To subscribe, mail curl-request@contactor.se with "subscribe <fill in your
+  email address>" in the body.
 
   To post to the list, mail curl@contactor.se.
 

docs/Makefile.am

@@ -7,6 +7,7 @@ AUTOMAKE_OPTIONS = foreign no-dependencies
 man_MANS = \
 	curl.1 \
 	curl_easy_cleanup.3 \
+	curl_easy_getinfo.3 \
 	curl_easy_init.3 \
 	curl_easy_perform.3 \
 	curl_easy_setopt.3 \

docs/README.win32

@@ -17,3 +17,8 @@ README.win32
   freely available nroff binary for win32 (*pointers appriciated*), convert
   the files into plain-text on your neighbor's unix machine or run over to the
   curl web site and view them as plain HTML.
+
+  The main curl.1 man page is "built-in". Use a command line similar to this
+  in order to extract a separate text file:
+
+        curl -M >manual.txt

docs/TODO

@@ -13,6 +13,32 @@ For the future
  product! (Yes, you may add things not mentioned here, these are just a
  few teasers...)
 
+ * Improve the command line option parser to accept '-m300' as well as the '-m
+   300' convention. It should be able to work if '-m300' is considered to be
+   space separated to the next option.
+
+ * Make the curl tool support URLs that start with @ that would then mean that
+   the following is a plain list with URLs to download. Thus @filename.txt
+   reads a list of URLs from a local file. A fancy option would then be to
+   support @http://whatever.com that would first load a list and then get the
+   URLs mentioned in the list. I figure -O or something would have to be
+   implied by such an action.
+
+ * Make curl with multiple URLs, even outside of {}-letters. I could also
+   imagine an optional fork()ed system that downloads each URL in its own
+   thread. It should of course have a maximum amount of simultaneous fork()s.
+
+ * Improve the regular progress meter with --continue is used. It should be
+   noticable when there's a resume going on.
+
+ * Add a command line option that allows the output file to get the same time
+   stamp as the remote file. This requires some fiddling on FTP but comes
+   almost free for HTTP.
+
+ * Make the SSL layer option capable of using the Mozilla Security Services as
+   an alternative to OpenSSL:
+   http://www.mozilla.org/projects/security/pki/nss/
+
  * Make sure the low-level interface works. highlevel.c should basically be
    possible to write using that interface. Document the low-level interface
 
@@ -22,9 +48,40 @@ For the future
  * Move non-URL related functions that are used by both the lib and the curl
    application to a separate "portability lib".
 
- * Add support for other languages than C (not important)
+ * Add support for other languages than C.  C++ (rumours have been heard about
+   something being worked on in this area) and perl (we have seen the first
+   versions of this!) comes to mind. Python anyone?
 
- * Improve the -K config file parser.
+ * Improve the -K config file parser (the parameter following the flag should
+   be possible to get specified *exactly* as it is done on a shell command
+   line).
+
+   Alternatively, and preferably, we rewrite the entire config file to become
+   a true config file that uses its own format instead of the currently
+   crippled and stupid format:
+
+        [option] = [value]
+
+   Where [option] would be the same as the --long-option and [value] would
+   either be 'on/off/true/false' for booleans or a plain value for [option]s
+   that accept variable input (such as -d, -o, -H, -d, -F etc).
+
+   [value] could be written as plain text, and then the initial and trailing
+   white spaces would be stripped off, or it can be specified within quotes
+   and then all white spaces within the quotes will count.
+
+   [value] could then be made to accept some format to specify an environment
+   variable. I could even think of supporting
+
+        [option] += [value]
+
+   for appending stuff to an option.
+
+   As has been suggested, ${name} could be used to read environment variables
+   and possibly other options. That could then be used instead of += operators
+   like:
+
+        bar = "foo ${bar}"
 
  * rtsp:// support -- "Real Time Streaming Protocol" (RFC 2326)
 
@@ -35,28 +92,13 @@ For the future
    started in October 1999 but halted again since it proved more work than we
    thought. It is still a good idea to implement though.
 
- * HTTP Pipelining/persistant connections
-
- - We should introduce HTTP "pipelining". Curl could be able to request for
-   several HTTP documents in one connect. It would be the beginning for
-   supporing more advanced functions in the future, like web site
-   mirroring. This will require that the urlget() function supports several
-   documents from a single HTTP server, which it doesn't today.
-
- - When curl supports fetching several documents from the same server using
-   pipelining, I'd like to offer that function to the command line. Anyone has
-   a good idea how? The current way of specifying one URL with the output sent
-   to the stdout or a file gets in the way. Imagine a syntax that supports
-   "additional documents from the same server" in a way similar to:
-
-     curl <main URL> --more-doc <path> --more-doc <path>
-
-   where --more-doc specifies another document on the same server. Where are
-   the output files gonna be put and how should they be named? Should each
-   "--more-doc" parameter require a local file name to store the result in?
-   Like "--more-file" as in:
-
-     curl <URL> --more-doc <path> --more-file <file>
+ * Authentication: NTLM. It would be cool to support that MS crap called NTLM
+   authentication. MS proxies and servers sometime require that. Since that
+   protocol is a proprietary one, it involves reverse engineering and network
+   sniffing. This should however be a library-based functionality. There are a
+   few different efforts "out there" to make open source HTTP clients support
+   this and it should be possible to take advantage of other people's hard
+   work.
 
  * RFC2617 compliance, "Digest Access Authentication"
    A valid test page seem to exist at:
@@ -70,12 +112,6 @@ For the future
    sends the password in cleartext over the network, this "Digest" method uses
    a challange-response protocol which increases security quite a lot.
 
- * Different FTP Upload Through Web Proxy
-   I don't know any web proxies that allow CONNECT through on port 21, but
-   that would be the best way to do ftp upload. All we would need to do would
-   be to 'CONNECT <host>:<port> HTTP/1.0\r\n\r\n' and then do business as
-   usual. I least I think so. It would be fun if someone tried this...
-
  * Multiple Proxies?
    Is there anyone that actually uses serial-proxies? I mean, send CONNECT to
    the first proxy to connect to the second proxy to which you send CONNECT to
@@ -86,8 +122,7 @@ For the future
    Ftp-kind proxy, Socks5, whatever kind of proxies are there?
 
  * IPv6 Awareness and support
-   Where ever it would fit. I am not that into v6 yet to fully grasp what we
-   would need to do, but letting the autoconf search for v6-versions of a few
+   Where ever it would fit.  configure search for v6-versions of a few
    functions and then use them instead is of course the first thing to do...
    RFC 2428 "FTP Extensions for IPv6 and NATs" will be interesting. PORT
    should be replaced with EPRT for IPv6, and EPSV instead of PASV.
@@ -97,8 +132,3 @@ For the future
 
  * HTTP POST resume using Range:
 
- * Make curl capable of verifying the server's certificate when connecting
-   with HTTPS://.
-
- * Kerberos-FTP
-

docs/TheArtOfHttpScripting

@@ -1,6 +1,6 @@
 Author:  Daniel Stenberg <daniel@haxx.se>
-Date:    August 7, 2000
-Version: 0.2
+Date:    September 15, 2000
+Version: 0.3
 
                 The Art Of Scripting HTTP Requests Using Curl
                 =============================================
@@ -31,7 +31,7 @@ Version: 0.2
 1. The HTTP Protocol
 
  HTTP is the protocol used to fetch data from web servers. It is a very simple
- protocol that is built upon TCP/IP. The protocol also allow information to
+ protocol that is built upon TCP/IP. The protocol also allows information to
  get sent to the server from the client using a few different methods, as will
  be shown here.
 
@@ -130,12 +130,12 @@ Version: 0.2
         curl -d "birthyear=1905&press=OK" www.hotmail.com/when/junk.cgi
 
   This kind of POST will use the Content-Type
-  application/x-www-form-urlencoded and is the most widly used POST kind.
+  application/x-www-form-urlencoded and is the most widely used POST kind.
 
  4.3 FILE UPLOAD POST
 
-  Back in late 1995 they defined a new to post data over HTTP. It was
-  documented in the RFC 1867, why this method sometimes are refered to as
+  Back in late 1995 they defined a new way to post data over HTTP. It was
+  documented in the RFC 1867, why this method sometimes is refered to as
   a rfc1867-posting.
 
   This method is mainly designed to better support file uploads. A form that
@@ -165,7 +165,7 @@ Version: 0.2
 
     <form method="POST" action="foobar.cgi">
       <input type=text name="birthyear">
-      <input type=text name="person" value="daniel">
+      <input type=hidden name="person" value="daniel">
       <input type=submit name="press" value="OK">
     </form>
 
@@ -209,17 +209,18 @@ Version: 0.2
 
  Do note that when a program is run, its parameters are possible to see when
  listing the running processes of the system. Thus, other users may be able to
- watch your passwords if you pass them as plain command line options.
+ watch your passwords if you pass them as plain command line options. There
+ are ways to circumvent this.
 
 7. REFERER
 
- A HTTP request has the ability to feature a 'referer' field, which can be
- used to tell which URL that causes the client to get this particular
- resource. Some programs/scripts check the referer field of requests to verify
- that this wasn't arriving from an external site or unknown page. While this
- is a stupid way to check something so easily forged, many scripts still do
- it. Using curl, you can put anything you want in the referer-field and thus
- more easily being able to fool the server into serving your request.
+ A HTTP request may include a 'referer' field, which can be used to tell from
+ which URL the client got to this particular resource. Some programs/scripts
+ check the referer field of requests to verify that this wasn't arriving from
+ an external site or an unknown page. While this is a stupid way to check
+ something so easily forged, many scripts still do it. Using curl, you can put
+ anything you want in the referer-field and thus more easily be able to fool
+ the server into serving your request.
 
  Use curl to set the referer field with:
 
@@ -278,10 +279,10 @@ Version: 0.2
  specified in a received cookie, the client sends back the cookies and their
  contents to the server, unless of course they are expired.
 
- Many applications and server use this method to connect a series of request
+ Many applications and servers use this method to connect a series of requests
  into a single logical session. To be able to use curl in such occations, we
- must be able to record and send back cookies in the way that the web
- application expects them. The same way browsers deal with them.
+ must be able to record and send back cookies the way the web application
+ expects them. The same way browsers deal with them.
 
  The simplest way to send a few cookies to the server when getting a page with
  curl is to add them on the command line like:
@@ -307,15 +308,15 @@ Version: 0.2
 
  There are a few ways to do secure HTTP transfers. The by far most common
  protocol for doing this is what is generally known as HTTPS, HTTP over
- SSL. SSL encrypts all the data that is send and received over the network and
+ SSL. SSL encrypts all the data that is sent and received over the network and
  thus makes it harder for attackers to spy on sensitive information.
 
  SSL (or TLS as the latest version of the standard is called) offers a
  truckload of advanced features to allow all those encryptions and key
- infrastructure mechanisms ecnrypted HTTP requires.
+ infrastructure mechanisms encrypted HTTP requires.
 
- Curl supports enscrypted fetches thanks to the freely available OpenSSL
- libraries. To get a pafe from a https server, simply run curl like:
+ Curl supports encrypted fetches thanks to the freely available OpenSSL
+ libraries. To get a page from a HTTPS server, simply run curl like:
 
         curl https://that.secure.server.com
 

docs/curl.1

@@ -2,7 +2,7 @@
 .\" nroff -man curl.1
 .\" Written by Daniel Stenberg
 .\"
-.TH curl 1 "24 August 2000" "Curl 7.2" "Curl Manual"
+.TH curl 1 "22 November 2000" "Curl 7.5" "Curl Manual"
 .SH NAME
 curl \- get a URL with FTP, TELNET, LDAP, GOPHER, DICT, FILE, HTTP or
 HTTPS syntax.
@@ -72,6 +72,7 @@ Use ASCII transfer when getting an FTP file or LDAP info. For FTP, this can
 also be enforced by using an URL that ends with ";type=A". This option causes
 data sent to stdout to be in text mode for win32 systems.
 .IP "-c/--continue"
+.B Deprecated. Use '-C -' instead.
 Continue/Resume a previous file transfer. This instructs curl to
 continue appending data on the file where it was previously left,
 possibly because of a broken connection to the server. There must be
@@ -92,11 +93,16 @@ HTTP resume is only possible with HTTP/1.1 or later servers.
 that the data is sent exactly as specified with no extra processing (with all
 newlines cut off).  The data is expected to be "url-encoded". This will cause
 curl to pass the data to the server using the content-type
-application/x-www-form-urlencoded. Compare to -F.
+application/x-www-form-urlencoded. Compare to -F. If more than one -d/--data
+option is used on the same command line, the data pieces specified will be
+merged together with a separating &-letter. Thus, using '-d name=daniel -d
+skill=lousy' would generate a post chunk that looks like
+'name=daniel&skill=lousy'.
 
 If you start the data with the letter @, the rest should be a file name to
-read the data from, or - if you want curl to read the data from stdin.
-The contents of the file must already be url-encoded.
+read the data from, or - if you want curl to read the data from stdin.  The
+contents of the file must already be url-encoded. Multiple files can also be
+specified.
 
 To post data purely binary, you should instead use the --data-binary option.
 
@@ -130,6 +136,9 @@ with HTTPS. The certificate must be in PEM format.
 If the optional password isn't specified, it will be queried for on
 the terminal. Note that this certificate is the private key and the private
 certificate concatenated!
+.IP "--cacert <CA certificate>"
+(HTTPS) Tells curl to use the specified certificate file to verify the
+peer. The certificate must be in PEM format.
 .IP "-f/--fail"
 (HTTP)
 Fail silently (no output at all) on server errors. This is mostly done
@@ -172,11 +181,20 @@ prevent that header from appearing.
 (HTTP)
 Include the HTTP-header in the output. The HTTP-header includes things
 like server-name, date of the document, HTTP-version and more...
+.IP "--interface <name>"
+Perform an operation using a specified interface. You can enter interface
+name, IP address or host name. An example could look like:
+
+.B "curl --interface eth0:1 http://www.netscape.com/"
 .IP "-I/--head"
 (HTTP/FTP)
 Fetch the HTTP-header only! HTTP-servers feature the command HEAD
 which this uses to get nothing but the header of a document. When used
 on a FTP file, curl displays the file size only.
+.IP "--krb4 <level>"
+(FTP) Enable kerberos4 authentication and use. The level must be entered and
+should be one of 'clear', 'safe', 'confidential' or 'private'. Should you use
+a level that is not one of these, 'private' will instead be used.
 .IP "-K/--config <config file>"
 Specify which config file to read curl arguments from. The config
 file is a text file in which command line arguments can be written
@@ -241,6 +259,12 @@ or use several variables like:
 .IP "-O/--remote-name"
 Write output to a local file named like the remote file we get. (Only
 the file part of the remote file is used, the path is cut off.)
+.IP "-p/--proxytunnel"
+When an HTTP proxy is used, this option will cause non-HTTP protocols to
+attempt to tunnel through the proxy instead of merely using it to do HTTP-like
+operations. The tunnel approach is made with the HTTP proxy CONNECT request
+and requires that the proxy allows direct connect to the remote port number
+curl wants to tunnel through to.
 .IP "-P/--ftpport <address>"
 (FTP)
 Reverses the initiator/listener roles when connecting with ftp. This
@@ -317,6 +341,7 @@ Curl mute.
 .IP "-S/--show-error"
 When used with -s it makes curl show error message if it fails.
 .IP "-t/--upload"
+.B Deprecated. Use '-T -' instead.
 Transfer the stdin data to the specified file. Curl will read
 everything from stdin until EOF and store with the supplied name. If
 this is used on a http(s) server, the PUT command will be used.
@@ -335,6 +360,9 @@ ask for it interactively.
 .IP "-U/--proxy-user <user:password>"
 Specify user and password to use for Proxy authentication. If no
 password is specified, curl will ask for it interactively.
+.IP "--url <URL>"
+Set the URL to fetch. This option is mostly handy when you wanna specify URL
+in a config file.
 .IP "-v/--verbose"
 Makes the fetching more verbose/talkative. Mostly usable for
 debugging. Lines starting with '>' means data sent by curl, '<'
@@ -393,11 +421,17 @@ The total amount of bytes that were downloaded.
 .B size_upload
 The total amount of bytes that were uploaded.
 .TP
+.B size_header
+The total amount of bytes of the downloaded headers.
+.TP
+.B size_request
+The total amount of bytes that were sent in the HTTP request.
+.TP
 .B speed_download
 The average download speed that curl measured for the complete download.
 .TP
 .B speed_upload
-The average upload speed that curl measured for the complete download.
+The average upload speed that curl measured for the complete upload.
 .RE
 .IP "-x/--proxy <proxyhost[:port]>"
 Use specified proxy. If the port number is not specified, it is assumed at
@@ -502,7 +536,7 @@ FTP weird USER reply. Curl couldn't parse the reply sent to the USER request.
 .IP 13
 FTP weird PASV reply, Curl couldn't parse the reply sent to the PASV request.
 .IP 14
-FTP weird 227 formay. Curl couldn't parse the 227-line the server sent.
+FTP weird 227 format. Curl couldn't parse the 227-line the server sent.
 .IP 15
 FTP can't get host. Couldn't resolve the host IP we got in the 227-line.
 .IP 16
@@ -560,12 +594,23 @@ LDAP search failed.
 Library not found. The LDAP library was not found.
 .IP 41
 Function not found. A required LDAP function was not found.
+.IP 42
+Aborted by callback. An application told curl to abort the operation.
+.IP 43
+Internal error. A function was called with a bad parameter.
+.IP 44
+Internal error. A function was called in a bad order.
+.IP 45
+Interface error. A specified outgoing interface could not be used.
+.IP 46
+Bad password entered. An error was signalled when the password was entered.
+.IP 47
+Too many redirects. When following redirects, curl hit the maximum amount.
 .IP XX
 There will appear more error codes here in future releases. The existing ones
 are meant to never change.
 .SH BUGS
-If you do find any (or have other suggestions), mail Daniel Stenberg
-<Daniel.Stenberg@haxx.se>.
+If you do find bugs, mail them to curl-bug@haxx.se.
 .SH AUTHORS / CONTRIBUTORS
  - Daniel Stenberg <Daniel.Stenberg@haxx.se>
  - Rafael Sagula <sagula@inf.ufrgs.br>
@@ -616,6 +661,13 @@ If you do find any (or have other suggestions), mail Daniel Stenberg
  - Fred Noz <FNoz@siac.com>
  - Caolan McNamara <caolan@csn.ul.ie>
  - Albert Chin-A-Young <china@thewrittenword.com>
+ - Stephen Kick <skick@epicrealm.com>
+ - Martin Hedenfalk <mhe@stacken.kth.se>
+ - Richard Prescott
+ - Jason S. Priebe <priebe@wral-tv.com>
+ - T. Bharath <TBharath@responsenetworks.com>
+ - Alexander Kourakos <awk@users.sourceforge.net>
+ - James Griffiths <griffiths_james@yahoo.com>
 
 .SH WWW
 http://curl.haxx.se

docs/curl_easy_getinfo.3

@@ -0,0 +1,92 @@
+.\" You can view this file with:
+.\" nroff -man [file]
+.\" Written by daniel@haxx.se
+.\"
+.TH curl_easy_init 3 "22 November 2000" "Curl 7.5" "libcurl Manual"
+.SH NAME
+curl_easy_getinfo - Extract information from a curl session (added in 7.4)
+.SH SYNOPSIS
+.B #include <curl/easy.h>
+.sp
+.BI "CURLcode curl_easy_getinfo(CURL *curl, CURLINFO info, ... );"
+.ad
+.SH DESCRIPTION
+Request internal information from the curl session with this function.  The
+third argument
+.B MUST
+be a pointer to a long, a pointer to a char * or a pointer to a double (as
+this documentation describes further down).  The data pointed-to will be
+filled in accordingly and can be relied upon only if the function returns
+CURLE_OK.  This function is intended to get used *AFTER* a performed transfer,
+all results from this function are undefined until the transfer is completed.
+.SH AVAILABLE INFORMATION
+These are informations that can be extracted:
+.TP 0.8i
+.B CURLINFO_EFFECTIVE_URL
+Pass a pointer to a 'char *' to receive the last used effective URL.
+.TP
+.B CURLINFO_HTTP_CODE
+Pass a pointer to a long to receive the last received HTTP code.
+.TP
+.B CURLINFO_FILETIME
+Pass a pointer to a long to receive the remote time of the retrieved
+document. If you get 0, it can be because of many reasons (unknown, the server
+hides it or the server doesn't support the command that tells document time
+etc) and the time of the document is unknown. (Added in 7.5)
+.TP
+.B CURLINFO_TOTAL_TIME
+Pass a pointer to a double to receive the total transaction time in seconds
+for the previous transfer.
+.TP
+.B CURLINFO_NAMELOOKUP_TIME
+Pass a pointer to a double to receive the time, in seconds, it took from the
+start until the name resolving was completed.
+.TP
+.B CURLINFO_CONNECT_TIME
+Pass a pointer to a double to receive the time, in seconds, it took from the
+start until the connect to the remote host (or proxy) was completed.
+.TP
+.B CURLINFO_PRETRANSFER_TIME
+Pass a pointer to a double to receive the time, in seconds, it took from the
+start until the file transfer is just about to begin. This includes all
+pre-transfer commands and negotiations that are specific to the particular
+protocol(s) involved.
+.TP
+.B CURLINFO_SIZE_UPLOAD
+Pass a pointer to a double to receive the total amount of bytes that were
+uploaded.
+.TP
+.B CURLINFO_SIZE_DOWNLOAD
+Pass a pointer to a double to receive the total amount of bytes that were
+downloaded.
+.TP
+.B CURLINFO_SPEED_DOWNLOAD
+Pass a pointer to a double to receive the average download speed that curl
+measured for the complete download.
+.TP
+.B CURLINFO_SPEED_UPLOAD
+Pass a pointer to a double to receive the average upload speed that curl
+measured for the complete upload.
+.TP
+.B CURLINFO_HEADER_SIZE
+Pass a pointer to a long to receive the total size of all the headers
+received.
+.TP
+.B CURLINFO_REQUEST_SIZE
+Pass a pointer to a long to receive the total size of the issued
+requests. This is so far only for HTTP requests. Note that this may be more
+than one request if FOLLOWLOCATION is true.
+.TP
+.B CURLINFO_SSL_VERIFYRESULT
+Pass a pointer to a long to receive the result of the certification
+verification that was requested (using the CURLOPT_SSL_VERIFYPEER option to
+curl_easy_setopt). (Added in 7.4.2)
+.PP
+
+.SH RETURN VALUE
+If the operation was successful, CURLE_OK is returned. Otherwise an
+appropriate error code will be returned.
+.SH "SEE ALSO"
+.BR curl_easy_setopt "(3)"
+.SH BUGS
+Surely there are some, you tell me!

docs/curl_easy_init.3

@@ -2,7 +2,7 @@
 .\" nroff -man [file]
 .\" Written by daniel@haxx.se
 .\"
-.TH curl_easy_init 3 "22 May 2000" "Curl 7.0" "libcurl Manual"
+.TH curl_easy_init 3 "26 September 2000" "Curl 7.0" "libcurl Manual"
 .SH NAME
 curl_easy_init - Start a libcurl "easy" session
 .SH SYNOPSIS
@@ -16,6 +16,9 @@ that you shall use as input to the other easy-functions. The init calls
 intializes curl and this call MUST have a corresponding call to
 .I curl_easy_cleanup
 when the operation is complete.
+
+On win32 systems, you need to init the winsock stuff manually, libcurl will
+not do that for you. WSAStartup() and WSACleanup() should be used accordingly.
 .SH RETURN VALUE
 If this function returns NULL, something went wrong and you cannot use the
 other curl functions.

docs/curl_easy_setopt.3

@@ -2,7 +2,7 @@
 .\" nroff -man [file]
 .\" Written by daniel@haxx.se
 .\"
-.TH curl_easy_setopt 3 "24 August 2000" "Curl 7.2" "libcurl Manual"
+.TH curl_easy_setopt 3 "28 November 2000" "Curl 7.5" "libcurl Manual"
 .SH NAME
 curl_easy_setopt - Set curl easy-session options
 .SH SYNOPSIS
@@ -87,6 +87,12 @@ the end of the host name. The proxy string may be prefixed with
 Set this long with this option to set the proxy port to use unless it is
 specified in the proxy string CURLOPT_PROXY.
 .TP
+.B CURLOPT_HTTPPROXYTUNNEL
+Set the parameter to non-zero to get the library to tunnel all non-HTTP
+operations through the given HTTP proxy. Do note that there is a big
+difference to use a proxy and to tunnel through it. If you don't know what
+this means, you probably don't want this tunnel option. (Added in libcurl 7.3)
+.TP
 .B CURLOPT_VERBOSE
 Set the parameter to non-zero to get the library to display a lot of verbose
 information about its operations.
@@ -194,7 +200,8 @@ post operation. See also the CURLOPT_POST.
 If you want to post data to the server without letting libcurl do a strlen()
 to measure the data size, this option must be used. Also, when this option is
 used, you can post fully binary data which otherwise is likely to fail. If
-this size is set to zero, the library will use strlen() to get the data size.
+this size is set to zero, the library will use strlen() to get the data
+size. (Added in libcurl 7.2)
 .TP
 .B CURLOPT_REFERER
 Pass a pointer to a zero terminated string as parameter. It will be used to
@@ -310,7 +317,20 @@ your server supports the command first.
 Pass a FILE * as parameter. This is the stream to use instead of stderr
 internally when reporting errors.
 .TP
+.B CURLOPT_INTERFACE
+Pass a char * as parameter. This set the interface name to use as outgoing
+network interface. The name can be an interface name, an IP address or a host
+name. (Added in libcurl 7.3)
+.TP
+.B CURLOPT_KRB4LEVEL
+Pass a char * as parameter. Set the krb4 security level, this also enables
+krb4 awareness.  This is a string, 'clear', 'safe', 'confidential' or
+'private'.  If the string is set but doesn't match one of these, 'private'
+will be used. Set the string to NULL to disable kerberos4. The kerberos
+support only works for FTP. (Added in libcurl 7.3)
+.TP
 .B CURLOPT_WRITEINFO
+(NOT PRESENT IN 7.4 or later!)
 Pass a pointer to a zero terminated string as parameter. It will be used to
 report information after a successful request. This string may contain
 variables that will be substituted by their contents when output. Described
@@ -332,6 +352,52 @@ Pass a pointer that will be untouched by libcurl and passed as the first
 argument in the progress callback set with
 .I CURLOPT_PROGRESSFUNCTION
  .
+.TP
+.B CURLOPT_SSL_VERIFYPEER
+Pass a long that is set to a non-zero value to make curl verify the peer's
+certificate. The certificate to verify against must be specified with the
+CURLOPT_CAINFO option. (Added in 7.4.2)
+.TP
+.B CURLOPT_CAINFO
+Pass a char * to a zero terminated file naming holding the certificate to
+verify the peer with. This only makes sense when used in combination with the
+CURLOPT_SSL_VERIFYPEER option. (Added in 7.4.2)
+.TP
+.B CURLOPT_PASSWDFUNCTION
+Pass a pointer to a curl_passwd_callback function that will then be called
+instead of the internal one if libcurl requests a password. The function must
+match this prototype:
+.BI "int my_getpass(void *client, char *prompt, char* buffer, int buflen );"
+If set to NULL, it equals to making the function always fail. If the function
+returns a non-zero value, it will abort the operation and an error
+(CURLE_BAD_PASSWORD_ENTERED) will be returned.
+.I client
+is a generic pointer, see CURLOPT_PASSWDDATA.
+.I prompt
+is a zero-terminated string that is text that prefixes the input request.
+.I buffer
+is a pointer to data where the entered password should be stored and
+.I buflen
+is the maximum number of bytes that may be written in the buffer.
+(Added in 7.4.2)
+.TP
+.B CURLOPT_PASSWDDATA
+Pass a void * to whatever data you want. The passed pointer will be the first
+argument sent to the specifed CURLOPT_PASSWDFUNCTION function. (Added in
+7.4.2)
+.TP
+.B CURLOPT_FILETIME
+Pass a long. If it is a non-zero value, libcurl will attempt to get the
+modification date of the remote document in this operation. This requires that
+the remote server sends the time or replies to a time querying command. The
+curl_easy_getinfo() function with the CURLINFO_FILETIME argument can be used
+after a transfer to extract the received time (if any). (Added in 7.5)
+.TP
+.B CURLOPT_MAXREDIRS
+Pass a long. The set number will be the redirection limit. If that many
+redirections have been followed, the next redirect will cause an error. This
+option only makes sense if the CURLOPT_FOLLOWLOCATION is used at the same
+time. (Added in 7.5)
 .PP
 .SH RETURN VALUE
 0 means the option was set properly, non-zero means an error as

docs/curl_formfree.3

@@ -0,0 +1,23 @@
+.\" You can view this file with:
+.\" nroff -man [file]
+.\" Written by daniel@haxx.se
+.\"
+.TH curl_formfree 3 "17 November 2000" "Curl 7.5" "libcurl Manual"
+.SH NAME
+curl_formfree - free a previously build multipart/formdata HTTP POST chain
+.SH SYNOPSIS
+.B #include <curl/curl.h>
+.sp
+.BI "void curl_formfree(struct HttpPost *" form);
+.ad
+.SH DESCRIPTION
+curl_formfree() is used to clean up data previously built/appended with
+curl_formparse(). This must be called when the data has been used, which
+typically means after the curl_easy_perform() has been called.
+.SH RETURN VALUE
+None
+.SH "SEE ALSO"
+.BR curl_formparse "(3) "
+.SH BUGS
+Surely there are some, you tell me!
+

docs/examples/README

@@ -0,0 +1,8 @@
+EXAMPLES
+
+This directory is for tiny libcurl programming examples. They are meant to
+show some simple steps on how you can build your own application to take full
+advantage of libcurl.
+
+If you end up with other small but still useful example sources, please mail
+them for submission in future packages and on the web site.

docs/examples/curlgtk.c

@@ -0,0 +1,87 @@
+/* curlgtk.c */
+/* Copyright (c) 2000 David Odin (aka DindinX) for MandrakeSoft */
+/* an attempt to use the curl library in concert with a gtk-threaded application */
+
+#include <stdio.h>
+#include <gtk/gtk.h>
+
+#include <curl/curl.h>
+#include <curl/types.h> /* new for v7 */
+#include <curl/easy.h> /* new for v7 */
+
+#include <pthread.h>
+
+GtkWidget *Bar;
+
+size_t my_read_func(void *ptr, size_t size, size_t nmemb, FILE *stream)
+{
+  return fread(ptr, size, nmemb, stream);
+}
+
+int my_progress_func(GtkWidget *Bar, int t, int d)
+{
+/*  printf("%d / %d (%g %%)\n", d, t, d*100.0/t);*/
+  gdk_threads_enter();
+  gtk_progress_set_value(GTK_PROGRESS(Bar), d*100.0/t);
+  gdk_threads_leave();
+  return 0;
+}
+
+void *curl_thread(void *ptr)
+{
+  CURL *curl;
+  CURLcode res;
+  FILE *outfile;
+  gchar *url = ptr;
+  
+  curl = curl_easy_init();
+  if(curl)
+  {
+    outfile = fopen("/tmp/test.curl", "w");
+
+    curl_easy_setopt(curl, CURLOPT_URL, url);
+    curl_easy_setopt(curl, CURLOPT_FILE, outfile);
+    curl_easy_setopt(curl, CURLOPT_READFUNCTION, my_read_func);
+    curl_easy_setopt(curl, CURLOPT_PROGRESSFUNCTION, my_progress_func);
+    curl_easy_setopt(curl, CURLOPT_PROGRESSDATA, Bar);
+    
+    res = curl_easy_perform(curl);
+
+    fclose(outfile);
+    /* always cleanup */
+    curl_easy_cleanup(curl);
+  }
+  return NULL;
+}
+
+int main(int argc, char **argv)
+{
+  GtkWidget *Window, *Frame, *Frame2;
+  GtkAdjustment *adj;
+  pthread_t curl_tid;
+
+  /* Init thread */
+  g_thread_init(NULL);
+  
+  gtk_init(&argc, &argv);
+  Window = gtk_window_new(GTK_WINDOW_TOPLEVEL);
+  Frame = gtk_frame_new(NULL);
+  gtk_frame_set_shadow_type(GTK_FRAME(Frame), GTK_SHADOW_OUT);
+  gtk_container_add(GTK_CONTAINER(Window), Frame);
+  Frame2 = gtk_frame_new(NULL);
+  gtk_frame_set_shadow_type(GTK_FRAME(Frame2), GTK_SHADOW_IN);
+  gtk_container_add(GTK_CONTAINER(Frame), Frame2);
+  gtk_container_set_border_width(GTK_CONTAINER(Frame2), 5);
+  adj = (GtkAdjustment*)gtk_adjustment_new(0, 0, 100, 0, 0, 0);
+  Bar = gtk_progress_bar_new_with_adjustment(adj);
+  gtk_container_add(GTK_CONTAINER(Frame2), Bar);
+  gtk_widget_show_all(Window);
+
+  pthread_create(&curl_tid, NULL, curl_thread, argv[1]);
+    
+  gdk_threads_enter();
+  gtk_main();
+  gdk_threads_leave();
+  return 0;
+}
+

docs/examples/sepheaders.c

@@ -0,0 +1,63 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+
+#include <curl/curl.h>
+#include <curl/types.h>
+#include <curl/easy.h>
+
+size_t  write_data(void *ptr, size_t size, size_t nmemb, FILE *stream)
+{
+  written = fwrite(ptr,size,nmemb,outfile);
+  return written;
+}
+
+int main(int argc, char **argv)
+{
+  CURL *curl_handle;
+  char *headerfilename = "head.out";
+  FILE *headerfile;
+  char *bodyfilename = "body.out";
+  FILE *bodyfile;
+
+  /* init the curl session */
+  curl_handle = curl_easy_init();
+
+  /* set URL to get */
+  curl_easy_setopt(curl_handle, CURLOPT_URL, "http://curl.haxx.se");
+
+  /* no progress meter please */
+  curl_easy_setopt(curl_handle, CURLOPT_NOPROGRESS, 1);
+
+  /* shut up completely */
+  curl_easy_setopt(curl_handle, CURLOPT_MUTE, 1);
+
+  /* send all data to this function  */
+  curl_easy_setopt(curl_handle, CURLOPT_WRITEFUNCTION, write_data);
+
+  /* open the files */
+  headerfile = fopen(headerfilename,"w");
+  if (headerfile == NULL) {
+    curl_easy_cleanup(curl_handle);
+    return -1;
+  }
+  bodyfile = fopen(bodyfilename,"w");
+  if (bodyfile == NULL) {
+    curl_easy_cleanup(curl_handle);
+    return -1;
+  }
+
+  /* we want the headers to this file handle */
+  curl_easy_setopt(curl_handle,   CURLOPT_WRITEHEADER ,headerfile);
+
+  /* get it! */
+  curl_easy_perform(curl_handle);
+
+  /* close the header file */
+  fclose(headerfile);
+
+  /* cleanup curl stuff */
+  curl_easy_cleanup(curl_handle);
+
+  return 0;
+}

docs/examples/simple.c

@@ -0,0 +1,26 @@
+#include <stdio.h>
+
+#include <curl/curl.h>
+#include <curl/types.h>
+#include <curl/easy.h>
+
+int main(int argc, char **argv)
+{
+  CURL *curl;
+  CURLcode res;
+  FILE *headerfile;
+
+  headerfile = fopen("dumpit", "w");
+
+  curl = curl_easy_init();
+  if(curl) {
+    /* what call to write: */
+    curl_easy_setopt(curl, CURLOPT_URL, "curl.haxx.se");
+    curl_easy_setopt(curl, CURLOPT_WRITEHEADER, headerfile);
+    res = curl_easy_perform(curl);
+
+    /* always cleanup */
+    curl_easy_cleanup(curl);
+  }
+  return 0;
+}

include/curl/curl.h

@@ -67,6 +67,10 @@
 
 #include <curl/types.h>
 
+#ifdef  __cplusplus
+extern "C" {
+#endif
+
 struct HttpPost {
   struct HttpPost *next; /* next entry in the list */
   char *name;     /* pointer to allocated name */
@@ -95,8 +99,17 @@ typedef size_t (*curl_read_callback)(char *buffer,
                                      size_t nitems,
                                      FILE *instream);
 
-/* All possible error codes from this version of urlget(). Future versions
-   may return other values, stay prepared. */
+typedef int (*curl_passwd_callback)(void *clientp,
+                                    char *prompt,
+                                    char *buffer,
+                                    int buflen);
+
+/* All possible error codes from all sorts of curl functions. Future versions
+   may return other values, stay prepared.
+
+   Always add new return codes last. Never *EVER* remove any. The return
+   codes must remain the same!
+ */
 
 typedef enum {
   CURLE_OK = 0,
@@ -153,10 +166,14 @@ typedef enum {
   CURLE_FUNCTION_NOT_FOUND,
   
   CURLE_ABORTED_BY_CALLBACK,
-
   CURLE_BAD_FUNCTION_ARGUMENT,
   CURLE_BAD_CALLING_ORDER,
 
+  CURLE_HTTP_PORT_FAILED, /* HTTP Interface operation failed */
+
+  CURLE_BAD_PASSWORD_ENTERED, /* when the my_getpass() returns fail */
+  CURLE_TOO_MANY_REDIRECTS , /* catch endless re-direct loops */
+
   CURL_LAST
 } CURLcode;
 
@@ -165,14 +182,17 @@ typedef enum {
 
 #define CURL_ERROR_SIZE 256
 
-/* maximum URL length we deal with */
+/* maximum URL length we deal with in headers */
 #define URL_MAX_LENGTH 4096 
 #define URL_MAX_LENGTH_TXT "4095"
 
 /* name is uppercase CURLOPT_<name>,
    type is one of the defined CURLOPTTYPE_<type>
    number is unique identifier */
-#define T(name,type,number) CURLOPT_ ## name = CURLOPTTYPE_ ## type + number
+#ifdef CINIT
+#undef CINIT
+#endif
+#define CINIT(name,type,number) CURLOPT_ ## name = CURLOPTTYPE_ ## type + number
 
 /* long may be 32 or 64 bits, but we should never depend on anything else
    but 32 */
@@ -181,74 +201,74 @@ typedef enum {
 #define CURLOPTTYPE_FUNCTIONPOINT 20000
 
 typedef enum {
-  T(NOTHING, LONG, 0), /********* the first one is unused ************/
+  CINIT(NOTHING, LONG, 0), /********* the first one is unused ************/
   
   /* This is the FILE * the regular output should be written to. */
-  T(FILE, OBJECTPOINT, 1),
+  CINIT(FILE, OBJECTPOINT, 1),
 
   /* The full URL to get/put */
-  T(URL,  OBJECTPOINT, 2),
+  CINIT(URL,  OBJECTPOINT, 2),
 
   /* Port number to connect to, if other than default. Specify the CONF_PORT
      flag in the CURLOPT_FLAGS to activate this */
-  T(PORT, LONG, 3),
+  CINIT(PORT, LONG, 3),
 
   /* Name of proxy to use. Specify the CONF_PROXY flag in the CURLOPT_FLAGS to
      activate this */
-  T(PROXY, OBJECTPOINT, 4),
+  CINIT(PROXY, OBJECTPOINT, 4),
   
   /* Name and password to use when fetching. Specify the CONF_USERPWD flag in
      the CURLOPT_FLAGS to activate this */
-  T(USERPWD, OBJECTPOINT, 5),
+  CINIT(USERPWD, OBJECTPOINT, 5),
 
   /* Name and password to use with Proxy. Specify the CONF_PROXYUSERPWD 
      flag in the CURLOPT_FLAGS to activate this */
-  T(PROXYUSERPWD, OBJECTPOINT, 6),
+  CINIT(PROXYUSERPWD, OBJECTPOINT, 6),
 
   /* Range to get, specified as an ASCII string. Specify the CONF_RANGE flag
      in the CURLOPT_FLAGS to activate this */
-  T(RANGE, OBJECTPOINT, 7),
+  CINIT(RANGE, OBJECTPOINT, 7),
 
 #if 0
   /* Configuration flags */
-  T(FLAGS, LONG, 8),
+  CINIT(FLAGS, LONG, 8),
 #endif
   /* Specified file stream to upload from (use as input): */
-  T(INFILE, OBJECTPOINT, 9),
+  CINIT(INFILE, OBJECTPOINT, 9),
 
   /* Buffer to receive error messages in, must be at least CURL_ERROR_SIZE
    * bytes big. If this is not used, error messages go to stderr instead: */
-  T(ERRORBUFFER, OBJECTPOINT, 10),
+  CINIT(ERRORBUFFER, OBJECTPOINT, 10),
 
   /* Function that will be called to store the output (instead of fwrite). The
    * parameters will use fwrite() syntax, make sure to follow them. */
-  T(WRITEFUNCTION, FUNCTIONPOINT, 11),
+  CINIT(WRITEFUNCTION, FUNCTIONPOINT, 11),
 
   /* Function that will be called to read the input (instead of fread). The
    * parameters will use fread() syntax, make sure to follow them. */
-  T(READFUNCTION, FUNCTIONPOINT, 12),
+  CINIT(READFUNCTION, FUNCTIONPOINT, 12),
 
   /* Time-out the read operation after this amount of seconds */
-  T(TIMEOUT, LONG, 13),
+  CINIT(TIMEOUT, LONG, 13),
 
   /* If the CURLOPT_INFILE is used, this can be used to inform libcurl about
    * how large the file being sent really is. That allows better error
    * checking and better verifies that the upload was succcessful. -1 means
    * unknown size. */
-  T(INFILESIZE, LONG, 14),
+  CINIT(INFILESIZE, LONG, 14),
 
   /* POST input fields. */
-  T(POSTFIELDS, OBJECTPOINT, 15),
+  CINIT(POSTFIELDS, OBJECTPOINT, 15),
 
   /* Set the referer page (needed by some CGIs) */
-  T(REFERER, OBJECTPOINT, 16),
+  CINIT(REFERER, OBJECTPOINT, 16),
 
   /* Set the FTP PORT string (interface name, named or numerical IP address)
      Use i.e '-' to use default address. */
-  T(FTPPORT, OBJECTPOINT, 17),
+  CINIT(FTPPORT, OBJECTPOINT, 17),
 
   /* Set the User-Agent string (examined by some CGIs) */
-  T(USERAGENT, OBJECTPOINT, 18),
+  CINIT(USERAGENT, OBJECTPOINT, 18),
 
   /* If the download receives less than "low speed limit" bytes/second
    * during "low speed time" seconds, the operations is aborted.
@@ -257,129 +277,158 @@ typedef enum {
    */
 
   /* Set the "low speed limit" */
-  T(LOW_SPEED_LIMIT, LONG , 19),
+  CINIT(LOW_SPEED_LIMIT, LONG , 19),
 
   /* Set the "low speed time" */
-  T(LOW_SPEED_TIME, LONG, 20),
+  CINIT(LOW_SPEED_TIME, LONG, 20),
 
   /* Set the continuation offset */
-  T(RESUME_FROM, LONG, 21),
+  CINIT(RESUME_FROM, LONG, 21),
 
   /* Set cookie in request: */
-  T(COOKIE, OBJECTPOINT, 22),
+  CINIT(COOKIE, OBJECTPOINT, 22),
 
   /* This points to a linked list of headers, struct HttpHeader kind */
-  T(HTTPHEADER, OBJECTPOINT, 23),
+  CINIT(HTTPHEADER, OBJECTPOINT, 23),
 
   /* This points to a linked list of post entries, struct HttpPost */
-  T(HTTPPOST, OBJECTPOINT, 24),
+  CINIT(HTTPPOST, OBJECTPOINT, 24),
 
   /* name of the file keeping your private SSL-certificate */
-  T(SSLCERT, OBJECTPOINT, 25),
+  CINIT(SSLCERT, OBJECTPOINT, 25),
 
   /* password for the SSL-certificate */
-  T(SSLCERTPASSWD, OBJECTPOINT, 26),
+  CINIT(SSLCERTPASSWD, OBJECTPOINT, 26),
   
   /* send TYPE parameter? */
-  T(CRLF, LONG, 27),
+  CINIT(CRLF, LONG, 27),
 
   /* send linked-list of QUOTE commands */
-  T(QUOTE, OBJECTPOINT, 28),
+  CINIT(QUOTE, OBJECTPOINT, 28),
 
   /* send FILE * to store headers to */
-  T(WRITEHEADER, OBJECTPOINT, 29),
+  CINIT(WRITEHEADER, OBJECTPOINT, 29),
 
 #ifdef MULTIDOC
   /* send linked list of MoreDoc structs */
-  T(MOREDOCS, OBJECTPOINT, 30),
+  CINIT(MOREDOCS, OBJECTPOINT, 30),
 #endif
 
   /* point to a file to read the initial cookies from, also enables
      "cookie awareness" */
-  T(COOKIEFILE, OBJECTPOINT, 31),
+  CINIT(COOKIEFILE, OBJECTPOINT, 31),
 
   /* What version to specifly try to use.
      3 = SSLv3, 2 = SSLv2, all else makes it try v3 first then v2 */
-  T(SSLVERSION, LONG, 32),
+  CINIT(SSLVERSION, LONG, 32),
 
   /* What kind of HTTP time condition to use, see defines */
-  T(TIMECONDITION, LONG, 33),
+  CINIT(TIMECONDITION, LONG, 33),
 
   /* Time to use with the above condition. Specified in number of seconds
      since 1 Jan 1970 */
-  T(TIMEVALUE, LONG, 34),
+  CINIT(TIMEVALUE, LONG, 34),
 
   /* HTTP request, for odd commands like DELETE, TRACE and others */
   /* OBSOLETE DEFINE, left for tradition only */
-  T(HTTPREQUEST, OBJECTPOINT, 35),
+  CINIT(HTTPREQUEST, OBJECTPOINT, 35),
 
   /* Custom request, for customizing the get command like
      HTTP: DELETE, TRACE and others
      FTP: to use a different list command
      */
-  T(CUSTOMREQUEST, OBJECTPOINT, 36),
+  CINIT(CUSTOMREQUEST, OBJECTPOINT, 36),
 
   /* HTTP request, for odd commands like DELETE, TRACE and others */
-  T(STDERR, OBJECTPOINT, 37),
+  CINIT(STDERR, OBJECTPOINT, 37),
 
 #if 0
   /* Progress mode set alternative progress mode displays. Alternative
      ones should now be made by the client, not the lib! */     
-  T(PROGRESSMODE, LONG, 38),
+  CINIT(PROGRESSMODE, LONG, 38),
 #endif
   /* send linked-list of post-transfer QUOTE commands */
-  T(POSTQUOTE, OBJECTPOINT, 39),
+  CINIT(POSTQUOTE, OBJECTPOINT, 39),
 
   /* Pass a pointer to string of the output using full variable-replacement
      as described elsewhere. */
-  T(WRITEINFO, OBJECTPOINT, 40),
+  CINIT(WRITEINFO, OBJECTPOINT, 40),
 
   /* Previous FLAG bits */
-  T(VERBOSE, LONG, 41),      /* talk a lot */
-  T(HEADER, LONG, 42),       /* throw the header out too */
-  T(NOPROGRESS, LONG, 43),   /* shut off the progress meter */
-  T(NOBODY, LONG, 44),       /* use HEAD to get http document */
-  T(FAILONERROR, LONG, 45),  /* no output on http error codes >= 300 */
-  T(UPLOAD, LONG, 46),       /* this is an upload */
-  T(POST, LONG, 47),         /* HTTP POST method */
-  T(FTPLISTONLY, LONG, 48),  /* Use NLST when listing ftp dir */
+  CINIT(VERBOSE, LONG, 41),      /* talk a lot */
+  CINIT(HEADER, LONG, 42),       /* throw the header out too */
+  CINIT(NOPROGRESS, LONG, 43),   /* shut off the progress meter */
+  CINIT(NOBODY, LONG, 44),       /* use HEAD to get http document */
+  CINIT(FAILONERROR, LONG, 45),  /* no output on http error codes >= 300 */
+  CINIT(UPLOAD, LONG, 46),       /* this is an upload */
+  CINIT(POST, LONG, 47),         /* HTTP POST method */
+  CINIT(FTPLISTONLY, LONG, 48),  /* Use NLST when listing ftp dir */
 
-  T(FTPAPPEND, LONG, 50),    /* Append instead of overwrite on upload! */
-  T(NETRC, LONG, 51),        /* read user+password from .netrc */
-  T(FOLLOWLOCATION, LONG, 52),  /* use Location: Luke! */
+  CINIT(FTPAPPEND, LONG, 50),    /* Append instead of overwrite on upload! */
+  CINIT(NETRC, LONG, 51),        /* read user+password from .netrc */
+  CINIT(FOLLOWLOCATION, LONG, 52),  /* use Location: Luke! */
 
   /* This FTPASCII name is now obsolete, to be removed, use the TRANSFERTEXT
      instead. It goes for more protocols than just ftp... */
-  T(FTPASCII, LONG, 53),     /* use TYPE A for transfer */
+  CINIT(FTPASCII, LONG, 53),     /* use TYPE A for transfer */
 
-  T(TRANSFERTEXT, LONG, 53), /* transfer data in text/ASCII format */
-  T(PUT, LONG, 54),          /* PUT the input file */
-  T(MUTE, LONG, 55),         /* force NOPROGRESS */
+  CINIT(TRANSFERTEXT, LONG, 53), /* transfer data in text/ASCII format */
+  CINIT(PUT, LONG, 54),          /* PUT the input file */
+  CINIT(MUTE, LONG, 55),         /* force NOPROGRESS */
 
   /* Function that will be called instead of the internal progress display
    * function. This function should be defined as the curl_progress_callback
    * prototype defines. */
-  T(PROGRESSFUNCTION, FUNCTIONPOINT, 56),
+  CINIT(PROGRESSFUNCTION, FUNCTIONPOINT, 56),
 
   /* Data passed to the progress callback */
-  T(PROGRESSDATA, OBJECTPOINT, 57),
+  CINIT(PROGRESSDATA, OBJECTPOINT, 57),
 
   /* We want the referer field set automatically when following locations */
-  T(AUTOREFERER, LONG, 58),
+  CINIT(AUTOREFERER, LONG, 58),
 
   /* Port of the proxy, can be set in the proxy string as well with:
      "[host]:[port]" */
-  T(PROXYPORT, LONG, 59),
+  CINIT(PROXYPORT, LONG, 59),
 
   /* size of the POST input data, if strlen() is not good to use */
-  T(POSTFIELDSIZE, LONG, 60),
+  CINIT(POSTFIELDSIZE, LONG, 60),
+
+  /* tunnel non-http operations through a HTTP proxy */
+  CINIT(HTTPPROXYTUNNEL, LONG, 61),
+
+  /* Set the interface string to use as outgoing network interface */
+  CINIT(INTERFACE, OBJECTPOINT, 62),
+
+  /* Set the krb4 security level, this also enables krb4 awareness.  This is a
+   * string, 'clear', 'safe', 'confidential' or 'private'.  If the string is
+   * set but doesn't match one of these, 'private' will be used.  */
+  CINIT(KRB4LEVEL, OBJECTPOINT, 63),
+
+  /* Set if we should verify the peer in ssl handshake, set 1 to verify. */
+  CINIT(SSL_VERIFYPEER, LONG, 64),
+  
+  /* The CApath or CAfile used to validate the peer certificate
+     this option is used only if SSL_VERIFYPEER is true */
+  CINIT(CAINFO, OBJECTPOINT, 65),
+
+  /* Function pointer to replace the internal password prompt */
+  CINIT(PASSWDFUNCTION, FUNCTIONPOINT, 66),
+
+  /* Custom pointer that gets passed as first argument to the password
+     function */
+  CINIT(PASSWDDATA, OBJECTPOINT, 67),
+  
+  /* Maximum number of http redirects to follow */
+  CINIT(MAXREDIRS, LONG, 68),
+
+  /* Pass a pointer to a time_t to get a possible date of the requested
+     document! Pass a NULL to shut it off. */
+  CINIT(FILETIME, OBJECTPOINT, 69),
 
   CURLOPT_LASTENTRY /* the last unusued */
 } CURLoption;
 
-#define CURL_PROGRESS_STATS 0 /* default progress display */
-#define CURL_PROGRESS_BAR   1
-
 typedef enum {
   TIMECOND_NONE,
 
@@ -392,10 +441,6 @@ typedef enum {
 
 #ifdef __BEOS__
 #include <support/SupportDefs.h>
-#else
-#ifndef __cplusplus        /* (rabe) */
-typedef char bool;
-#endif                     /* (rabe) */
 #endif
 
 
@@ -414,18 +459,21 @@ int curl_formparse(char *string,
                    struct HttpPost **httppost,
                    struct HttpPost **last_post);
 
+/* cleanup a form: */
+void curl_formfree(struct HttpPost *form);
+
 /* Unix and Win32 getenv function call, this returns a malloc()'ed string that
    MUST be free()ed after usage is complete. */
 char *curl_getenv(char *variable);
 
-/* returns ascii string of the libcurl version */
+/* Returns a static ascii string of the libcurl version. */
 char *curl_version(void);
 
 /* This is the version number */
-#define LIBCURL_VERSION "7.2"
-#define LIBCURL_VERSION_NUM 0x070200
+#define LIBCURL_VERSION "7.5"
+#define LIBCURL_VERSION_NUM 0x070500
 
-/* linked-list structure for the CURLOPT_QUOTE option */
+/* linked-list structure for the CURLOPT_QUOTE option (and other) */
 struct curl_slist {
 	char			*data;
 	struct curl_slist	*next;
@@ -622,4 +670,49 @@ CURLcode curl_disconnect(CURLconnect *connect);
  */
 time_t curl_getdate(const char *p, const time_t *now);
 
+
+#define CURLINFO_STRING   0x100000
+#define CURLINFO_LONG     0x200000
+#define CURLINFO_DOUBLE   0x300000
+#define CURLINFO_MASK     0x0fffff
+#define CURLINFO_TYPEMASK 0xf00000
+
+typedef enum {
+  CURLINFO_NONE, /* first, never use this */
+  CURLINFO_EFFECTIVE_URL    = CURLINFO_STRING + 1,
+  CURLINFO_HTTP_CODE        = CURLINFO_LONG   + 2,
+  CURLINFO_TOTAL_TIME       = CURLINFO_DOUBLE + 3,
+  CURLINFO_NAMELOOKUP_TIME  = CURLINFO_DOUBLE + 4,
+  CURLINFO_CONNECT_TIME     = CURLINFO_DOUBLE + 5,
+  CURLINFO_PRETRANSFER_TIME = CURLINFO_DOUBLE + 6,
+  CURLINFO_SIZE_UPLOAD      = CURLINFO_DOUBLE + 7,
+  CURLINFO_SIZE_DOWNLOAD    = CURLINFO_DOUBLE + 8,
+  CURLINFO_SPEED_DOWNLOAD   = CURLINFO_DOUBLE + 9,
+  CURLINFO_SPEED_UPLOAD     = CURLINFO_DOUBLE + 10,
+  CURLINFO_HEADER_SIZE      = CURLINFO_LONG   + 11,
+  CURLINFO_REQUEST_SIZE     = CURLINFO_LONG   + 12,
+  CURLINFO_SSL_VERIFYRESULT = CURLINFO_LONG   + 13,
+  CURLINFO_FILETIME         = CURLINFO_LONG   + 14,
+
+  CURLINFO_LASTONE          = 15
+} CURLINFO;
+
+/*
+ * NAME curl_getinfo()
+ *
+ * DESCRIPTION
+ *
+ * Request internal information from the curl session with this function.
+ * The third argument MUST be a pointer to a long or a pointer to a char *.
+ * The data pointed to will be filled in accordingly and can be relied upon
+ * only if the function returns CURLE_OK.
+ * This function is intended to get used *AFTER* a performed transfer, all
+ * results are undefined before the transfer is completed.
+ */
+CURLcode curl_getinfo(CURL *curl, CURLINFO info, ...);
+
+#ifdef  __cplusplus
+}
+#endif
+
 #endif /* __CURL_CURL_H */

include/curl/easy.h

@@ -39,8 +39,32 @@
  *
  * ------------------------------------------------------------
  ****************************************************************************/
+#ifdef  __cplusplus
+extern "C" {
+#endif
+
 CURL *curl_easy_init(void);
 CURLcode curl_easy_setopt(CURL *curl, CURLoption option, ...);
 CURLcode curl_easy_perform(CURL *curl);
 void curl_easy_cleanup(CURL *curl);
+
+/*
+ * NAME curl_easy_getinfo()
+ *
+ * DESCRIPTION
+ *
+ * Request internal information from the curl session with this function.  The
+ * third argument MUST be a pointer to a long, a pointer to a char * or a
+ * pointer to a double (as the documentation describes elsewhere).  The data
+ * pointed to will be filled in accordingly and can be relied upon only if the
+ * function returns CURLE_OK.  This function is intended to get used *AFTER* a
+ * performed transfer, all results from this function are undefined until the
+ * transfer is completed.
+ */
+CURLcode curl_easy_getinfo(CURL *curl, CURLINFO info, ...);
+
+#ifdef  __cplusplus
+}
+#endif
+
 #endif

lib/Makefile.am

@@ -2,15 +2,45 @@
 # $Id$
 #
 
-AUTOMAKE_OPTIONS = foreign no-dependencies
+AUTOMAKE_OPTIONS = foreign
 
 lib_LTLIBRARIES = libcurl.la
 
 # Some flags needed when trying to cause warnings ;-)
-# CFLAGS = -g -Wall #-pedantic
+# CFLAGS = -DMALLOCDEBUG -g # -Wall #-pedantic
 
 INCLUDES = -I$(top_srcdir)/include
 
+
+libcurl_la_LDFLAGS = -version-info 1:0:0
+# This flag accepts an argument of the form current[:revision[:age]]. So,
+# passing -version-info 3:12:1 sets current to 3, revision to 12, and age to
+# 1.
+#
+# If either revision or age are omitted, they default to 0. Also note that age
+# must be less than or equal to the current interface number.
+#
+# Here are a set of rules to help you update your library version information: 
+#
+# 1.Start with version information of 0:0:0 for each libtool library.
+#
+# 2.Update the version information only immediately before a public release of
+# your software. More frequent updates are unnecessary, and only guarantee
+# that the current interface number gets larger faster.
+#
+# 3.If the library source code has changed at all since the last update, then
+# increment revision (c:r:a becomes c:r+1:a).
+#
+# 4.If any interfaces have been added, removed, or changed since the last
+# update, increment current, and set revision to 0.
+#
+# 5.If any interfaces have been added since the last public release, then
+# increment age.
+#
+# 6.If any interfaces have been removed since the last public release, then
+# set age to 0.
+#
+
 libcurl_la_SOURCES = \
 arpa_telnet.h  file.c         getpass.h      netrc.h        timeval.c \
 base64.c       file.h         hostip.c       progress.c     timeval.h \
@@ -23,7 +53,8 @@ download.c     getdate.h      ldap.c         ssluse.c       version.c \
 download.h     getenv.c       ldap.h         ssluse.h       \
 escape.c       getenv.h       mprintf.c      telnet.c       \
 escape.h       getpass.c      netrc.c        telnet.h       \
-writeout.c writeout.h highlevel.c strequal.c strequal.h easy.c
+getinfo.c highlevel.c strequal.c strequal.h easy.c \
+security.h security.c krb4.c memdebug.c memdebug.h
 
 # Say $(srcdir), so GNU make does not report an ambiguity with the .y.c rule.
 $(srcdir)/getdate.c: getdate.y

lib/Makefile.in

@@ -77,16 +77,45 @@ RANLIB = @RANLIB@
 VERSION = @VERSION@
 YACC = @YACC@
 
-AUTOMAKE_OPTIONS = foreign no-dependencies
+AUTOMAKE_OPTIONS = foreign
 
 lib_LTLIBRARIES = libcurl.la
 
 # Some flags needed when trying to cause warnings ;-)
-# CFLAGS = -g -Wall #-pedantic
+# CFLAGS = -DMALLOCDEBUG -g # -Wall #-pedantic
 
 INCLUDES = -I$(top_srcdir)/include
 
-libcurl_la_SOURCES =  arpa_telnet.h  file.c         getpass.h      netrc.h        timeval.c base64.c       file.h         hostip.c       progress.c     timeval.h base64.h       formdata.c     hostip.h       progress.h     cookie.c       formdata.h     http.c         sendf.c        cookie.h       ftp.c          http.h         sendf.h        url.c dict.c         ftp.h          if2ip.c        speedcheck.c   url.h dict.h         getdate.c      if2ip.h        speedcheck.h   urldata.h download.c     getdate.h      ldap.c         ssluse.c       version.c download.h     getenv.c       ldap.h         ssluse.h       escape.c       getenv.h       mprintf.c      telnet.c       escape.h       getpass.c      netrc.c        telnet.h       writeout.c writeout.h highlevel.c strequal.c strequal.h easy.c
+libcurl_la_LDFLAGS = -version-info 1:0:0
+# This flag accepts an argument of the form current[:revision[:age]]. So,
+# passing -version-info 3:12:1 sets current to 3, revision to 12, and age to
+# 1.
+#
+# If either revision or age are omitted, they default to 0. Also note that age
+# must be less than or equal to the current interface number.
+#
+# Here are a set of rules to help you update your library version information: 
+#
+# 1.Start with version information of 0:0:0 for each libtool library.
+#
+# 2.Update the version information only immediately before a public release of
+# your software. More frequent updates are unnecessary, and only guarantee
+# that the current interface number gets larger faster.
+#
+# 3.If the library source code has changed at all since the last update, then
+# increment revision (c:r:a becomes c:r+1:a).
+#
+# 4.If any interfaces have been added, removed, or changed since the last
+# update, increment current, and set revision to 0.
+#
+# 5.If any interfaces have been added since the last public release, then
+# increment age.
+#
+# 6.If any interfaces have been removed since the last public release, then
+# set age to 0.
+#
+
+libcurl_la_SOURCES =  arpa_telnet.h  file.c         getpass.h      netrc.h        timeval.c base64.c       file.h         hostip.c       progress.c     timeval.h base64.h       formdata.c     hostip.h       progress.h     cookie.c       formdata.h     http.c         sendf.c        cookie.h       ftp.c          http.h         sendf.h        url.c dict.c         ftp.h          if2ip.c        speedcheck.c   url.h dict.h         getdate.c      if2ip.h        speedcheck.h   urldata.h download.c     getdate.h      ldap.c         ssluse.c       version.c download.h     getenv.c       ldap.h         ssluse.h       escape.c       getenv.h       mprintf.c      telnet.c       escape.h       getpass.c      netrc.c        telnet.h       getinfo.c highlevel.c strequal.c strequal.h easy.c security.h security.c krb4.c memdebug.c memdebug.h
 
 mkinstalldirs = $(SHELL) $(top_srcdir)/mkinstalldirs
 CONFIG_HEADER = ../config.h ../src/config.h
@@ -98,13 +127,12 @@ DEFS = @DEFS@ -I. -I$(srcdir) -I.. -I../src
 CPPFLAGS = @CPPFLAGS@
 LDFLAGS = @LDFLAGS@
 LIBS = @LIBS@
-libcurl_la_LDFLAGS = 
 libcurl_la_LIBADD = 
 libcurl_la_OBJECTS =  file.lo timeval.lo base64.lo hostip.lo progress.lo \
 formdata.lo cookie.lo http.lo sendf.lo ftp.lo url.lo dict.lo if2ip.lo \
 speedcheck.lo getdate.lo download.lo ldap.lo ssluse.lo version.lo \
-getenv.lo escape.lo mprintf.lo telnet.lo getpass.lo netrc.lo \
-writeout.lo highlevel.lo strequal.lo easy.lo
+getenv.lo escape.lo mprintf.lo telnet.lo getpass.lo netrc.lo getinfo.lo \
+highlevel.lo strequal.lo easy.lo security.lo krb4.lo memdebug.lo
 CFLAGS = @CFLAGS@
 COMPILE = $(CC) $(DEFS) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS)
 LTCOMPILE = $(LIBTOOL) --mode=compile $(CC) $(DEFS) $(INCLUDES) $(AM_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS)

lib/Makefile.m32

@@ -1,6 +1,6 @@
 #############################################################
 ## Makefile for building libcurl.a with MingW32 (GCC-2.95) and
-## optionally OpenSSL (0.9.4)
+## optionally OpenSSL (0.9.6)
 ## Use: make -f Makefile.m32
 ##
 ## Comments to: Troy Engel <tengel@sonic.net> or
@@ -9,48 +9,59 @@
 CC = gcc
 AR = ar
 RANLIB = ranlib
-OPENSSL_PATH = ../../openssl-0.9.5a
+STRIP = strip -g
+OPENSSL_PATH = ../../openssl-0.9.6
 
 ########################################################
 ## Nothing more to do below this line!
 
-INCLUDES = -I. -I.. -I../include
+INCLUDES = -I. -I.. -I../include -I../src
 CFLAGS = -g -O2 -DMINGW32
 ifdef SSL
   INCLUDES += -I"$(OPENSSL_PATH)/outinc" -I"$(OPENSSL_PATH)/outinc/openssl"
   CFLAGS += -DUSE_SSLEAY
+  DLL_LIBS = -leay32 -lssl32 -lRSAglue
 endif
 COMPILE = $(CC) $(INCLUDES) $(CFLAGS)
 
 libcurl_a_LIBRARIES = libcurl.a
 
-libcurl_a_SOURCES =  base64.c getenv.c if2ip.h progress.h  \
-base64.h getenv.h mprintf.c setup.h url.c download.c getpass.c \
-mprintf.h ssluse.c url.h download.h hostip.c netrc.c ssluse.h \
-urldata.h formdata.c hostip.h netrc.h stdcheaders.h formdata.h \
-if2ip.c progress.c sendf.c sendf.h speedcheck.c speedcheck.h \
-ftp.c ftp.h getpass.h version.c timeval.c  timeval.h cookie.c \
-cookie.h escape.c escape.h getdate.c getdate.h dict.h dict.c http.c \
-http.h telnet.c telnet.h file.c file.h ldap.c ldap.h writeout.c writeout.h \
-highlevel.c strequal.c strequal.h easy.c
+libcurl_a_SOURCES = arpa_telnet.h file.c getpass.h netrc.h timeval.c base64.c \
+	file.h hostip.c progress.c timeval.h base64.h formdata.c hostip.h progress.h	\
+	cookie.c formdata.h http.c sendf.c cookie.h ftp.c http.h sendf.h url.c dict.c	\
+	ftp.h if2ip.c speedcheck.c url.h dict.h getdate.c if2ip.h speedcheck.h		\
+	urldata.h download.c getdate.h ldap.c ssluse.c version.c download.h getenv.c	\
+	ldap.h ssluse.h escape.c getenv.h mprintf.c telnet.c escape.h getpass.c netrc.c	\
+	telnet.h getinfo.c highlevel.c strequal.c strequal.h easy.c security.h		\
+	security.c krb4.c
 
-libcurl_a_OBJECTS =  base64.o getenv.o mprintf.o url.o download.o \
-getpass.o ssluse.o hostip.o netrc.o formdata.o if2ip.o progress.o \
-sendf.o speedcheck.o ftp.o version.o timeval.o \
-cookie.o escape.o getdate.o dict.o http.o telnet.o file.o ldap.o writeout.o \
-highlevel.o strequal.o easy.o
+libcurl_a_OBJECTS =  file.o timeval.o base64.o hostip.o progress.o \
+	formdata.o cookie.o http.o sendf.o ftp.o url.o dict.o if2ip.o \
+	speedcheck.o getdate.o download.o ldap.o ssluse.o version.o \
+	getenv.o escape.o mprintf.o telnet.o getpass.o netrc.o getinfo.o \
+	highlevel.o strequal.o easy.o security.o krb4.o
 
 LIBRARIES =  $(libcurl_a_LIBRARIES)
 SOURCES = $(libcurl_a_SOURCES)
 OBJECTS = $(libcurl_a_OBJECTS)
 
 
-all: libcurl.a
+all: libcurl.a libcurl.dll libcurldll.a
 
 libcurl.a: $(libcurl_a_OBJECTS) $(libcurl_a_DEPENDENCIES)
 	-@erase libcurl.a
 	$(AR) cru libcurl.a $(libcurl_a_OBJECTS)
 	$(RANLIB) libcurl.a
+	$(STRIP) $@
+
+# remove the last line above to keep debug info
+
+libcurl.dll libcurldll.a: libcurl.a libcurl.def dllinit.o
+	-@erase $@
+	dllwrap --dllname $@ --output-lib libcurldll.a --export-all --def libcurl.def $(libcurl_a_LIBRARIES) dllinit.o -L$(OPENSSL_PATH)/out $(DLL_LIBS) -lwsock32
+	$(STRIP) $@
+
+# remove the last line above to keep debug info
 
 .c.o:
 	$(COMPILE) -c $<

lib/Makefile.vc6

@@ -4,28 +4,30 @@
 ##      (default is release)
 ##
 ## Comments to: Troy Engel <tengel@sonic.net>
+## Updated by: Craig Davison <cd@securityfocus.com>
 
 PROGRAM_NAME = libcurl.lib
-OPENSSL_PATH = ../../openssl-0.9.3a
+PROGRAM_NAME_DEBUG = libcurld.lib
+OPENSSL_PATH = ../../openssl-0.9.6
 
 ########################################################
 ## Nothing more to do below this line!
 
 ## Release
-CCR = cl.exe /ML /O2 /D "NDEBUG"
-LINKR = link.exe -lib
+CCR = cl.exe /MD /O2 /D "NDEBUG"
+LINKR = link.exe -lib /out:$(PROGRAM_NAME)
 
 ## Debug
-CCD = cl.exe /MLd /Gm /ZI /Od /D "_DEBUG" /GZ
-LINKD = link.exe -lib
+CCD = cl.exe /MDd /Gm /ZI /Od /D "_DEBUG" /GZ
+LINKD = link.exe -lib /out:$(PROGRAM_NAME_DEBUG)
 
 ## SSL Release
-CCRS = cl.exe /ML /O2 /D "NDEBUG" /D "USE_SSLEAY" /I "$(OPENSSL_PATH)/inc32" /I "$(OPENSSL_PATH)/inc32/openssl"
-LINKRS = link.exe -lib /LIBPATH:$(OPENSSL_PATH)/out32dll
+CCRS = cl.exe /MD /O2 /D "NDEBUG" /D "USE_SSLEAY" /I "$(OPENSSL_PATH)/include" /I "$(OPENSSL_PATH)/include/openssl"
+LINKRS = link.exe -lib /out:$(PROGRAM_NAME) /LIBPATH:$(OPENSSL_PATH)/out32dll
 
 CFLAGS = /I "../include" /nologo /W3 /GX /D "WIN32" /D "VC6" /D "_MBCS" /D "_LIB" /YX /FD /c /D "MSDOS"
-LFLAGS = /nologo /out:$(PROGRAM_NAME)
-LINKLIBS = kernel32.lib wsock32.lib
+LFLAGS = /nologo
+LINKLIBS = wsock32.lib
 LINKSLIBS = libeay32.lib ssleay32.lib RSAglue.lib
 
 RELEASE_OBJS= \
@@ -53,7 +55,7 @@ RELEASE_OBJS= \
 	timevalr.obj \
 	urlr.obj \
 	filer.obj \
-    writeoutr.obj \
+	getinfor.obj \
 	versionr.obj \
 	easyr.obj \
 	highlevelr.obj \
@@ -84,7 +86,7 @@ DEBUG_OBJS= \
 	timevald.obj \
 	urld.obj \
 	filed.obj \
-    writeoutd.obj \
+	getinfod.obj \
 	versiond.obj  \
 	easyd.obj \
 	highleveld.obj \
@@ -115,7 +117,7 @@ RELEASE_SSL_OBJS= \
 	timevalrs.obj \
 	urlrs.obj \
 	filers.obj \
-    writeouts.obj \
+	getinfors.obj \
 	versionrs.obj \
 	easyrs.obj \
 	highlevelrs.obj \
@@ -146,7 +148,7 @@ LINK_OBJS= \
 	timeval.obj \
 	url.obj \
 	file.obj \
-    writeout.obj \
+	getinfo.obj \
 	version.obj \
 	easy.obj \
 	highlevel.obj \
@@ -163,7 +165,6 @@ debug: $(DEBUG_OBJS)
 release-ssl: $(RELEASE_SSL_OBJS)
 	$(LINKRS) $(LFLAGS) $(LINKLIBS) $(LINKSLIBS) $(LINK_OBJS)
 
-
 ## Release
 base64r.obj: base64.c
 	$(CCR) $(CFLAGS) base64.c
@@ -213,8 +214,8 @@ urlr.obj: url.c
 	$(CCR) $(CFLAGS) url.c
 filer.obj: file.c
 	$(CCR) $(CFLAGS) file.c
-writeoutr.obj: writeout.c
-        $(CCR) $(CFLAGS) writeout.c
+getinfor.obj: getinfo.c
+	$(CCR) $(CFLAGS) getinfo.c
 versionr.obj: version.c
 	$(CCR) $(CFLAGS) version.c
 easyr.obj: easy.c
@@ -240,7 +241,7 @@ ftpd.obj: ftp.c
 httpd.obj: http.c
 	$(CCD) $(CFLAGS) http.c
 ldapd.obj: ldap.c
-	$(CCR) $(CFLAGS) ldap.c
+	$(CCD) $(CFLAGS) ldap.c
 dictd.obj: dict.c
 	$(CCD) $(CFLAGS) dict.c
 telnetd.obj: telnet.c
@@ -273,16 +274,16 @@ urld.obj: url.c
 	$(CCD) $(CFLAGS) url.c
 filed.obj: file.c
 	$(CCD) $(CFLAGS) file.c
-writeoutd.obj: writeout.c
-        $(CCR) $(CFLAGS) writeout.c
+getinfod.obj: getinfo.c
+	$(CCD) $(CFLAGS) getinfo.c
 versiond.obj: version.c
 	$(CCD) $(CFLAGS) version.c
 easyd.obj: easy.c
-	$(CCR) $(CFLAGS) easy.c
+	$(CCD) $(CFLAGS) easy.c
 highleveld.obj: highlevel.c
-	$(CCR) $(CFLAGS) highlevel.c
+	$(CCD) $(CFLAGS) highlevel.c
 strequald.obj: strequal.c
-	$(CCR) $(CFLAGS) strequal.c
+	$(CCD) $(CFLAGS) strequal.c
 
 
 ## Release SSL
@@ -301,7 +302,7 @@ ftprs.obj: ftp.c
 httprs.obj: http.c
 	$(CCRS) $(CFLAGS) http.c
 ldaprs.obj: ldap.c
-	$(CCR) $(CFLAGS) ldap.c
+	$(CCRS) $(CFLAGS) ldap.c
 dictrs.obj: dict.c
 	$(CCRS) $(CFLAGS) dict.c
 telnetrs.obj: telnet.c
@@ -334,16 +335,17 @@ urlrs.obj: url.c
 	$(CCRS) $(CFLAGS) url.c
 filers.obj: file.c
 	$(CCRS) $(CFLAGS) file.c
-writeoutrs.obj: writeout.c
-        $(CCR) $(CFLAGS) writeout.c
+getinfors.obj: getinfo.c
+	$(CCRS) $(CFLAGS) getinfo.c
 versionrs.obj: version.c
 	$(CCRS) $(CFLAGS) version.c
 easyrs.obj: easy.c
-	$(CCR) $(CFLAGS) easy.c
+	$(CCRS) $(CFLAGS) easy.c
 highlevelrs.obj: highlevel.c
-	$(CCR) $(CFLAGS) highlevel.c
+	$(CCRS) $(CFLAGS) highlevel.c
 strequalrs.obj: strequal.c
-	$(CCR) $(CFLAGS) strequal.c
+	$(CCRS) $(CFLAGS) strequal.c
+
 
 clean:
 	-@erase *.obj

lib/base64.c

@@ -1,94 +1,152 @@
-/*****************************************************************************
- *                                  _   _ ____  _     
- *  Project                     ___| | | |  _ \| |    
- *                             / __| | | | |_) | |    
- *                            | (__| |_| |  _ <| |___ 
- *                             \___|\___/|_| \_\_____|
+/*
+ * Copyright (c) 1995 - 1999 Kungliga Tekniska H<>gskolan
+ * (Royal Institute of Technology, Stockholm, Sweden).
+ * All rights reserved.
  * 
- *  The contents of this file are subject to the Mozilla Public License
- *  Version 1.0 (the "License"); you may not use this file except in
- *  compliance with the License. You may obtain a copy of the License at
- *  http://www.mozilla.org/MPL/
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
  * 
- *  Software distributed under the License is distributed on an "AS IS"
- *  basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
- *  License for the specific language governing rights and limitations
- *  under the License.
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
  * 
- *  The Original Code is Curl.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
  * 
- *  The Initial Developer of the Original Code is Daniel Stenberg.
+ * 3. Neither the name of the Institute nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
  * 
- *  Portions created by the Initial Developer are Copyright (C) 1998.
- *  All Rights Reserved.
- *
- * ------------------------------------------------------------
- * Main author:
- * - Daniel Stenberg <daniel@haxx.se>
- *
- * 	http://curl.haxx.se
- *
- * $Source$
- * $Revision$
- * $Date$
- * $Author$
- * $State$
- * $Locker$
- *
- * ------------------------------------------------------------
- ****************************************************************************/
+ * THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
 
-#include <stdio.h>
+#ifdef HAVE_CONFIG_H
+#include <config.h>
+#endif
+#include <stdlib.h>
+#include <string.h>
+#include "base64.h"
 
-/* ---- Base64 Encoding --- */
-static char table64[]=
-  "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
 
-void base64Encode(char *intext, char *output)
+static char base64[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
+
+static int pos(char c)
 {
-  unsigned char ibuf[3];
-  unsigned char obuf[4];
+  char *p;
+  for(p = base64; *p; p++)
+    if(*p == c)
+      return p - base64;
+  return -1;
+}
+
+#if 1
+int base64_encode(const void *data, int size, char **str)
+{
+  char *s, *p;
   int i;
-  int inputparts;
+  int c;
+  const unsigned char *q;
 
-  while(*intext) {
-    for (i = inputparts = 0; i < 3; i++) { 
-      if(*intext) {
-        inputparts++;
-        ibuf[i] = *intext;
-        intext++;
+  p = s = (char*)malloc(size*4/3+4);
+  if (p == NULL)
+      return -1;
+  q = (const unsigned char*)data;
+  i=0;
+  for(i = 0; i < size;){
+    c=q[i++];
+    c*=256;
+    if(i < size)
+      c+=q[i];
+    i++;
+    c*=256;
+    if(i < size)
+      c+=q[i];
+    i++;
+    p[0]=base64[(c&0x00fc0000) >> 18];
+    p[1]=base64[(c&0x0003f000) >> 12];
+    p[2]=base64[(c&0x00000fc0) >> 6];
+    p[3]=base64[(c&0x0000003f) >> 0];
+    if(i > size)
+      p[3]='=';
+    if(i > size+1)
+      p[2]='=';
+    p+=4;
   }
+  *p=0;
+  *str = s;
+  return strlen(s);
+}
+#endif
+
+int base64_decode(const char *str, void *data)
+{
+  const char *p;
+  unsigned char *q;
+  int c;
+  int x;
+  int done = 0;
+  q=(unsigned char*)data;
+  for(p=str; *p && !done; p+=4){
+    x = pos(p[0]);
+    if(x >= 0)
+      c = x;
+    else{
+      done = 3;
+      break;
+    }
+    c*=64;
+    
+    x = pos(p[1]);
+    if(x >= 0)
+      c += x;
     else
-        ibuf[i] = 0;
-    }
+      return -1;
+    c*=64;
     
-    obuf [0] = (ibuf [0] & 0xFC) >> 2;
-    obuf [1] = ((ibuf [0] & 0x03) << 4) | ((ibuf [1] & 0xF0) >> 4);
-    obuf [2] = ((ibuf [1] & 0x0F) << 2) | ((ibuf [2] & 0xC0) >> 6);
-    obuf [3] = ibuf [2] & 0x3F;
+    if(p[2] == '=')
+      done++;
+    else{
+      x = pos(p[2]);
+      if(x >= 0)
+	c += x;
+      else
+	return -1;
+    }
+    c*=64;
     
-    switch(inputparts) {
-    case 1: /* only one byte read */
-      sprintf(output, "%c%c==", 
-              table64[obuf[0]],
-              table64[obuf[1]]);
-      break;
-    case 2: /* two bytes read */
-      sprintf(output, "%c%c%c=", 
-              table64[obuf[0]],
-              table64[obuf[1]],
-              table64[obuf[2]]);
-      break;
-    default:
-      sprintf(output, "%c%c%c%c", 
-              table64[obuf[0]],
-              table64[obuf[1]],
-              table64[obuf[2]],
-              table64[obuf[3]] );
-      break;
+    if(p[3] == '=')
+      done++;
+    else{
+      if(done)
+	return -1;
+      x = pos(p[3]);
+      if(x >= 0)
+	c += x;
+      else
+	return -1;
     }
-    output += 4;
+    if(done < 3)
+      *q++=(c&0x00ff0000)>>16;
+      
+    if(done < 2)
+      *q++=(c&0x0000ff00)>>8;
+    if(done < 1)
+      *q++=(c&0x000000ff)>>0;
   }
-  *output=0;
+  return q - (unsigned char*)data;
 }
-/* ---- End of Base64 Encoding ---- */

lib/base64.h

@@ -1,44 +1,39 @@
+/*
+ * Copyright (c) 1995, 1996, 1997 Kungliga Tekniska H<>gskolan
+ * (Royal Institute of Technology, Stockholm, Sweden).
+ * All rights reserved.
+ * 
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 
+ * 3. Neither the name of the Institute nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ * 
+ * THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
 #ifndef __BASE64_H
 #define __BASE64_H
-/*****************************************************************************
- *                                  _   _ ____  _     
- *  Project                     ___| | | |  _ \| |    
- *                             / __| | | | |_) | |    
- *                            | (__| |_| |  _ <| |___ 
- *                             \___|\___/|_| \_\_____|
- *
- *  The contents of this file are subject to the Mozilla Public License
- *  Version 1.0 (the "License"); you may not use this file except in
- *  compliance with the License. You may obtain a copy of the License at
- *  http://www.mozilla.org/MPL/
- *
- *  Software distributed under the License is distributed on an "AS IS"
- *  basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
- *  License for the specific language governing rights and limitations
- *  under the License.
- *
- *  The Original Code is Curl.
- *
- *  The Initial Developer of the Original Code is Daniel Stenberg.
- *
- *  Portions created by the Initial Developer are Copyright (C) 1998.
- *  All Rights Reserved.
- *
- * ------------------------------------------------------------
- * Main author:
- * - Daniel Stenberg <daniel@haxx.se>
- *
- * 	http://curl.haxx.se
- *
- * $Source$
- * $Revision$
- * $Date$
- * $Author$
- * $State$
- * $Locker$
- *
- * ------------------------------------------------------------
- ****************************************************************************/
 
-void base64Encode(char *intext, char *output);
+int base64_encode(const void *data, int size, char **str);
+
 #endif

lib/cookie.c

@@ -65,6 +65,11 @@ Example set of cookies:
 #include "getdate.h"
 #include "strequal.h"
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 /****************************************************************************
  *
  * cookie_add()
@@ -99,7 +104,8 @@ struct Cookie *cookie_add(struct CookieInfo *c,
 
     semiptr=strchr(lineptr, ';'); /* first, find a semicolon */
     ptr = lineptr;
-    while(semiptr) {
+    do {
+      if(semiptr)
         *semiptr='\0'; /* zero terminate for a while */
       /* we have a <what>=<this> pair or a 'secure' word here */
       if(strchr(ptr, '=')) {
@@ -156,12 +162,15 @@ struct Cookie *cookie_add(struct CookieInfo *c,
             ; /* unsupported keyword without assign! */
         }
       }
+      if(!semiptr)
+        continue; /* we already know there are no more cookies */
+
       *semiptr=';'; /* put the semicolon back */
       ptr=semiptr+1;
       while(ptr && *ptr && isspace((int)*ptr))
         ptr++;
       semiptr=strchr(ptr, ';'); /* now, find the next semicolon */
-    }
+    } while(semiptr);
   }
   else {
     /* This line is NOT a HTTP header style line, we do offer support for
@@ -400,7 +409,7 @@ struct Cookie *cookie_getlist(struct CookieInfo *c,
 	 /* now check if the domain is correct */
 	 domlen=co->domain?strlen(co->domain):0;
 	 if(!co->domain ||
-	    ((domlen<hostlen) &&
+	    ((domlen<=hostlen) &&
 	     strequal(host+(hostlen-domlen), co->domain)) ) {
 	    /* the right part of the host matches the domain stuff in the
 	       cookie data */
@@ -492,6 +501,7 @@ void cookie_cleanup(struct CookieInfo *c)
 	 free(co);
 	 co = next;
       }
+      free(c); /* free the base struct as well */
    }
 }
 

lib/dict.c

@@ -233,7 +233,7 @@ CURLcode dict(struct connectdata *conn)
       int i;
 	
       ppath++;
-      for (i = 0; (i < URL_MAX_LENGTH) && (ppath[i]); i++) {
+      for (i = 0; ppath[i]; i++) {
         if (ppath[i] == ':')
           ppath[i] = ' ';
       }

lib/dllinit.c

@@ -0,0 +1,82 @@
+/* dllinit.c -- Portable DLL initialization. 
+   Copyright (C) 1998, 1999 Free Software Foundation, Inc.
+   Contributed by Mumit Khan (khan@xraylith.wisc.edu).
+
+   I've used DllMain as the DLL "main" since that's the most common 
+   usage. MSVC and Mingw32 both default to DllMain as the standard
+   callback from the linker entry point. Cygwin, as of b20.1, also
+   uses DllMain as the default callback from the entry point.
+
+   The real entry point is typically always defined by the runtime 
+   library, and usually never overridden by (casual) user. What you can 
+   override however is the callback routine that the entry point calls, 
+   and this file provides such a callback function, DllMain.
+
+   Mingw32: The default entry point for mingw32 is DllMainCRTStartup
+   which is defined in libmingw32.a This in turn calls DllMain which is
+   defined here. If not defined, there is a stub in libmingw32.a which
+   does nothing.
+
+   Cygwin: The default entry point for Cygwin b20.1 or newer is
+   __cygwin_dll_entry which is defined in libcygwin.a. This in turn
+   calls the routine DllMain. If not defined, there is a stub in
+   libcygwin.a which does nothing. 
+
+   MSVC: MSVC runtime calls DllMain, just like Mingw32.
+
+   Summary: If you need to do anything special in DllMain, just add it
+   here. Otherwise, the default setup should be just fine for 99%+ of
+   the time. I strongly suggest that you *not* change the entry point,
+   but rather change DllMain as appropriate.
+
+ */
+
+
+#define WIN32_LEAN_AND_MEAN
+#include <windows.h>
+#undef WIN32_LEAN_AND_MEAN
+#include <stdio.h>
+
+BOOL APIENTRY DllMain (HINSTANCE hInst, DWORD reason, 
+                       LPVOID reserved /* Not used. */ );
+
+/*
+ *----------------------------------------------------------------------
+ *
+ * DllMain --
+ *
+ *	This routine is called by the Mingw32, Cygwin32 or VC++ C run 
+ *	time library init code, or the Borland DllEntryPoint routine. It 
+ *	is responsible for initializing various dynamically loaded 
+ *	libraries.
+ *
+ * Results:
+ *      TRUE on sucess, FALSE on failure.
+ *
+ * Side effects:
+ *
+ *----------------------------------------------------------------------
+ */
+BOOL APIENTRY
+DllMain (
+	 HINSTANCE hInst /* Library instance handle. */ ,
+	 DWORD reason /* Reason this function is being called. */ ,
+	 LPVOID reserved /* Not used. */ )
+{
+
+  switch (reason)
+    {
+    case DLL_PROCESS_ATTACH:
+      break;
+
+    case DLL_PROCESS_DETACH:
+      break;
+
+    case DLL_THREAD_ATTACH:
+      break;
+
+    case DLL_THREAD_DETACH:
+      break;
+    }
+  return TRUE;
+}

lib/easy.c

@@ -109,6 +109,9 @@ CURL *curl_easy_init(void)
     return NULL;
 
   data->interf = CURLI_EASY; /* mark it as an easy one */
+  /* SAC */
+  data->device = NULL;
+
   return data;
 }
 
@@ -159,3 +162,13 @@ void curl_easy_cleanup(CURL *curl)
   curl_close(curl);
   curl_free();
 }
+
+CURLcode curl_easy_getinfo(CURL *curl, CURLINFO info, ...)
+{
+  va_list arg;
+  void *paramp;
+  va_start(arg, info);
+  paramp = va_arg(arg, void *);
+
+  return curl_getinfo(curl, info, paramp);
+}

lib/escape.c

@@ -41,10 +41,18 @@
 /* Escape and unescape URL encoding in strings. The functions return a new
  * allocated string or NULL if an error occurred.  */
 
+#include "setup.h"
+#include <curl/curl.h>
+
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 char *curl_escape(char *string)
 {
    int alloc=strlen(string)+1;
@@ -88,16 +96,24 @@ char *curl_unescape(char *string, int length)
    unsigned char in;
    int index=0;
    int hex;
+   char querypart=FALSE; /* everything to the right of a '?' letter is
+                            the "query part" where '+' should become ' '.
+                            RFC 2316, section 3.10 */
   
-   while(--alloc) {
+   while(--alloc > 0) {
       in = *string;
-      if('+' == in)
+      if(querypart && ('+' == in))
          in = ' ';
+      else if(!querypart && ('?' == in)) {
+        /* we have "walked in" to the query part */
+        querypart=TRUE;
+      }
       else if('%' == in) {
         /* encoded part */
         if(sscanf(string+1, "%02X", &hex)) {
           in = hex;
           string+=2;
+          alloc-=2;
         }
       }
 

lib/file.c

@@ -103,6 +103,10 @@
 #define _MPRINTF_REPLACE /* use our functions only */
 #include <curl/mprintf.h>
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
 
 CURLcode file(struct connectdata *conn)
 {
@@ -151,9 +155,6 @@ CURLcode file(struct connectdata *conn)
      this is both more efficient than the former call to download() and
      it avoids problems with select() and recv() on file descriptors
      in Winsock */
-#if 0
-  ProgressInit (data, expected_size);
-#endif
   if(expected_size != -1)
     pgrsSetDownloadSize(data, expected_size);
 
@@ -170,10 +171,11 @@ CURLcode file(struct connectdata *conn)
        Windows systems if the target is stdout. Use -O or -o parameters
        to prevent CR/LF translation (this then goes to a binary mode
        file descriptor). */
-    if(nread != data->fwrite (buf, 1, nread, data->out)) {
-      failf (data, "Failed writing output");
-      return CURLE_WRITE_ERROR;
-    }
+
+    res = client_write(data, CLIENTWRITE_BODY, buf, nread);
+    if(res)
+      return res;
+
     now = tvnow();
     if(pgrsUpdate(data))
       res = CURLE_ABORTED_BY_CALLBACK;
@@ -184,7 +186,5 @@ CURLcode file(struct connectdata *conn)
 
   close(fd);
 
-  free(actual_path);
-
   return res;
 }

lib/formdata.c

@@ -63,6 +63,11 @@
 
 #include "strequal.h"
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 /* Length of the random boundary string. The risk of this being used
    in binary data is very close to zero, 64^32 makes
    6277101735386680763835789423207666416102355444464034512896
@@ -378,7 +383,7 @@ char *MakeFormBoundary(void)
   return retstring;
 }
 
-
+/* Used from http.c */ 
 void FormFree(struct FormData *form)
 {
   struct FormData *next;
@@ -390,6 +395,28 @@ void FormFree(struct FormData *form)
   } while((form=next)); /* continue */
 }
 
+/* external function to free up a whole form post chain */
+void curl_formfree(struct HttpPost *form)
+{
+  struct HttpPost *next;
+  do {
+    next=form->next;  /* the following form line */
+
+    /* recurse to sub-contents */
+    if(form->more)
+      curl_formfree(form->more);
+
+    if(form->name)
+      free(form->name); /* free the name */
+    if(form->contents)
+      free(form->contents); /* free the contents */
+    if(form->contenttype)
+      free(form->contenttype); /* free the content type */
+    free(form);       /* free the struct */
+
+  } while((form=next)); /* continue */
+}
+
 struct FormData *getFormData(struct HttpPost *post,
 			     int *sizep)
 {
@@ -458,12 +485,20 @@ struct FormData *getFormData(struct HttpPost *post,
 			     "\r\nContent-Type: %s",
 			     file->contenttype);
       }
+
+#if 0
+      /* The header Content-Transfer-Encoding: seems to confuse some receivers
+       * (like the built-in PHP engine). While I can't see any reason why it
+       * should, I can just as well skip this to the benefit of the users who
+       * are using such confused receivers.
+       */
+      
       if(file->contenttype &&
 	 !strnequal("text/", file->contenttype, 5)) {
 	/* this is not a text content, mention our binary encoding */
 	size += AddFormData(&form, "\r\nContent-Transfer-Encoding: binary", 0);
       }
-
+#endif
 
       size += AddFormData(&form, "\r\n\r\n", 0);
 

lib/ftp.c

@@ -85,6 +85,15 @@
 #include "progress.h"
 #include "download.h"
 #include "escape.h"
+#include "http.h" /* for HTTP proxy tunnel stuff */
+
+#ifdef KRB4
+#include "security.h"
+#endif
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
 
 /* returns last node in linked list */
 static struct curl_slist *slist_get_last(struct curl_slist *list)
@@ -120,7 +129,7 @@ struct curl_slist *curl_slist_append(struct curl_slist *list, char *data)
 	}
 	else {
 		fprintf(stderr, "Cannot allocate memory for QUOTE list.\n");
-		exit(-1);
+		return NULL;
 	}
 
 	if (list) {
@@ -209,7 +218,8 @@ static CURLcode AllowServerConnect(struct UrlData *data,
 			isdigit((int)line[2]) && (' ' == line[3]))
 
 int GetLastResponse(int sockfd, char *buf,
-                    struct connectdata *conn)
+                    struct connectdata *conn,
+                    int *ftpcode)
 {
   int nread;
   int keepon=TRUE;
@@ -225,6 +235,8 @@ int GetLastResponse(int sockfd, char *buf,
 #define SELECT_TIMEOUT 2
   int error = SELECT_OK;
 
+  *ftpcode=0; /* 0 for errors */
+
   if(data->timeout) {
     /* if timeout is requested, find out how much remaining time we have */
     timeout = data->timeout - /* timeout time */
@@ -265,8 +277,8 @@ int GetLastResponse(int sockfd, char *buf,
         break;
       default:
 #ifdef USE_SSLEAY
-        if (data->use_ssl) {
-          keepon = SSL_read(data->ssl, ptr, 1);
+        if (data->ssl.use) {
+          keepon = SSL_read(data->ssl.handle, ptr, 1);
         }
         else {
 #endif
@@ -285,6 +297,19 @@ int GetLastResponse(int sockfd, char *buf,
     }
     *ptr=0; /* zero terminate */
 
+#if KRB4
+    { /* handle the security-oriented responses 6xx ***/
+      /* FIXME: some errorchecking perhaps... ***/
+      if(strncmp(buf, "631", 3) == 0)
+        sec_read_msg(conn, buf, prot_safe);
+      else if(strncmp(buf, "632", 3) == 0)
+        sec_read_msg(conn, buf, prot_private);
+      else if(strncmp(buf, "633", 3) == 0)
+        sec_read_msg(conn, buf, prot_confidential);
+      nread = strlen(buf);
+    }
+#endif
+
     if(data->bits.verbose && buf[0]) {
       fputs("< ", data->err);
       fwrite(buf, 1, nread, data->err);
@@ -296,6 +321,8 @@ int GetLastResponse(int sockfd, char *buf,
   if(error)
     return -error;
 
+  *ftpcode=atoi(buf); /* return the initial number like this */
+
   return nread;
 }
 
@@ -316,53 +343,6 @@ char *getmyhost(char *buf, int buf_size)
   return buf;
 }
 
-#if 0
-/*
- * URLfix()
- *
- * This function returns a string converted FROM the input URL format to a
- * format that is more likely usable for the remote server. That is, all
- * special characters (found as %XX-codes) will be eascaped with \<letter>.
- */
-
-static char *URLfix(char *string)
-{
-  /* The length of the new string can't be longer than twice the original
-     string, if all letters are '+'... */
-  int alloc = strlen(string)*2;
-  char *ns = malloc(alloc);
-  unsigned char in;
-  int index=0;
-  int hex;
-   
-  while(*string) {
-    in = *string;
-    switch(in) {
-    case '+':
-      ns[index++] = '\\';
-      ns[index++] = ' ';
-      string++;
-      continue;
-
-    case '%':
-      /* encoded part */
-      if(sscanf(string+1, "%02X", &hex)) {
-        ns[index++] = '\\';
-        ns[index++] = hex;
-        string+=3;
-        continue;
-      }
-      /* FALLTHROUGH */
-    default:
-      ns[index++] = in;
-      string++;
-    }
-  }
-  ns[index]=0; /* terminate it */
-  return ns;
-}
-#endif
-
 /* ftp_connect() should do everything that is to be considered a part
    of the connection phase. */
 CURLcode ftp_connect(struct connectdata *conn)
@@ -372,6 +352,8 @@ CURLcode ftp_connect(struct connectdata *conn)
   struct UrlData *data=conn->data;
   char *buf = data->buffer; /* this is our buffer */
   struct FTP *ftp;
+  CURLcode result;
+  int ftpcode;
 
   myalarm(0); /* switch off the alarm stuff */
 
@@ -387,44 +369,73 @@ CURLcode ftp_connect(struct connectdata *conn)
   ftp->user = data->user;
   ftp->passwd = data->passwd;
 
+  if (data->bits.tunnel_thru_httpproxy) {
+    /* We want "seamless" FTP operations through HTTP proxy tunnel */
+    result = GetHTTPProxyTunnel(data, data->firstsocket,
+                                data->hostname, data->remote_port);
+    if(CURLE_OK != result)
+      return result;
+  }
+
   /* The first thing we do is wait for the "220*" line: */
-  nread = GetLastResponse(data->firstsocket, buf, conn);
+  nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
   if(nread < 0)
     return CURLE_OPERATION_TIMEOUTED;
-  if(strncmp(buf, "220", 3)) {
+
+  if(ftpcode != 220) {
     failf(data, "This doesn't seem like a nice ftp-server response");
     return CURLE_FTP_WEIRD_SERVER_REPLY;
   }
 
+#ifdef KRB4
+  /* if not anonymous login, try a secure login */
+  if(data->bits.krb4) {
+
+    /* request data protection level (default is 'clear') */
+    sec_request_prot(conn, "private");
+
+    /* We set private first as default, in case the line below fails to
+       set a valid level */
+    sec_request_prot(conn, data->krb4_level);
+
+    data->cmdchannel = fdopen(data->firstsocket, "w");
+
+    if(sec_login(conn) != 0)
+      infof(data, "Logging in with password in cleartext!\n");
+    else
+      infof(data, "Authentication successful\n");
+  }
+#endif
+  
   /* send USER */
-  sendf(data->firstsocket, data, "USER %s\r\n", ftp->user);
+  ftpsendf(data->firstsocket, conn, "USER %s", ftp->user);
 
   /* wait for feedback */
-  nread = GetLastResponse(data->firstsocket, buf, conn);
+  nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
   if(nread < 0)
     return CURLE_OPERATION_TIMEOUTED;
 
-  if(!strncmp(buf, "530", 3)) {
+  if(ftpcode == 530) {
     /* 530 User ... access denied
        (the server denies to log the specified user) */
     failf(data, "Access denied: %s", &buf[4]);
     return CURLE_FTP_ACCESS_DENIED;
   }
-  else if(!strncmp(buf, "331", 3)) {
+  else if(ftpcode == 331) {
     /* 331 Password required for ...
        (the server requires to send the user's password too) */
-    sendf(data->firstsocket, data, "PASS %s\r\n", ftp->passwd);
-    nread = GetLastResponse(data->firstsocket, buf, conn);
+    ftpsendf(data->firstsocket, conn, "PASS %s", ftp->passwd);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
-    if(!strncmp(buf, "530", 3)) {
+    if(ftpcode == 530) {
       /* 530 Login incorrect.
          (the username and/or the password are incorrect) */
       failf(data, "the username and/or the password are incorrect");
       return CURLE_FTP_USER_PASSWORD_INCORRECT;
     }
-    else if(!strncmp(buf, "230", 3)) {
+    else if(ftpcode == 230) {
       /* 230 User ... logged in.
          (user successfully logged in) */
         
@@ -435,10 +446,23 @@ CURLcode ftp_connect(struct connectdata *conn)
       return CURLE_FTP_WEIRD_PASS_REPLY;
     }
   }
-  else if(! strncmp(buf, "230", 3)) {
+  else if(buf[0] == '2') {
     /* 230 User ... logged in.
        (the user logged in without password) */
     infof(data, "We have successfully logged in\n");
+#ifdef KRB4
+	/* we are logged in (with Kerberos)
+	 * now set the requested protection level
+	 */
+    if(conn->sec_complete)
+      sec_set_protection_level(conn);
+
+    /* we may need to issue a KAUTH here to have access to the files
+     * do it if user supplied a password
+     */
+    if(conn->data->passwd && *conn->data->passwd)
+      krb_kauth(conn);
+#endif
   }
   else {
     failf(data, "Odd return code after USER");
@@ -457,6 +481,7 @@ CURLcode ftp_done(struct connectdata *conn)
   size_t nread;
   char *buf = data->buffer; /* this is our buffer */
   struct curl_slist *qitem; /* QUOTE item */
+  int ftpcode;
 
   if(data->bits.upload) {
     if((-1 != data->infilesize) && (data->infilesize != *ftp->bytecountp)) {
@@ -471,26 +496,31 @@ CURLcode ftp_done(struct connectdata *conn)
       failf(data, "Received only partial file");
       return CURLE_PARTIAL_FILE;
     }
-    else if(0 == *ftp->bytecountp) {
+    else if(!data->bits.no_body && (0 == *ftp->bytecountp)) {
       failf(data, "No data was received!");
       return CURLE_FTP_COULDNT_RETR_FILE;
     }
   }
+#ifdef KRB4
+  sec_fflush_fd(conn, data->secondarysocket);
+#endif
   /* shut down the socket to inform the server we're done */
   sclose(data->secondarysocket);
   data->secondarysocket = -1;
 
+  if(!data->bits.no_body) {  
     /* now let's see what the server says about the transfer we
        just performed: */
-  nread = GetLastResponse(data->firstsocket, buf, conn);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
     /* 226 Transfer complete, 250 Requested file action okay, completed. */
-  if(!strncmp(buf, "226", 3) && !strncmp(buf, "250", 3)) {
+    if((ftpcode != 226) && (ftpcode != 250)) {
       failf(data, "%s", buf+4);
       return CURLE_FTP_WRITE_ERROR;
     }
+  }
 
   /* Send any post-transfer QUOTE strings? */
   if(data->postquote) {
@@ -499,9 +529,9 @@ CURLcode ftp_done(struct connectdata *conn)
     while (qitem) {
       /* Send string */
       if (qitem->data) {
-        sendf(data->firstsocket, data, "%s\r\n", qitem->data);
+        ftpsendf(data->firstsocket, conn, "%s", qitem->data);
 
-        nread = GetLastResponse(data->firstsocket, buf, conn);
+        nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
         if(nread < 0)
           return CURLE_OPERATION_TIMEOUTED;
 
@@ -515,12 +545,8 @@ CURLcode ftp_done(struct connectdata *conn)
     }
   }
 
-  if(ftp->file)
-    free(ftp->file);
-  if(ftp->dir)
-    free(ftp->dir);
-
-  /* TBD: the ftp struct is still allocated here */
+  free(ftp);
+  data->proto.ftp=NULL; /* it is gone */
 
   return CURLE_OK;
 }
@@ -548,6 +574,7 @@ CURLcode _ftp(struct connectdata *conn)
   struct FTP *ftp = data->proto.ftp;
 
   long *bytecountp = ftp->bytecountp;
+  int ftpcode; /* for ftp status */
 
   /* Send any QUOTE strings? */
   if(data->quote) {
@@ -556,9 +583,9 @@ CURLcode _ftp(struct connectdata *conn)
     while (qitem) {
       /* Send string */
       if (qitem->data) {
-        sendf(data->firstsocket, data, "%s\r\n", qitem->data);
+        ftpsendf(data->firstsocket, conn, "%s", qitem->data);
 
-        nread = GetLastResponse(data->firstsocket, buf, conn);
+        nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
         if(nread < 0)
           return CURLE_OPERATION_TIMEOUTED;
 
@@ -572,6 +599,48 @@ CURLcode _ftp(struct connectdata *conn)
     }
   }
 
+  /* change directory first! */
+  if(ftp->dir && ftp->dir[0]) {
+    ftpsendf(data->firstsocket, conn, "CWD %s", ftp->dir);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
+    if(nread < 0)
+      return CURLE_OPERATION_TIMEOUTED;
+
+    if(ftpcode != 250) {
+      failf(data, "Couldn't change to directory %s", ftp->dir);
+      return CURLE_FTP_ACCESS_DENIED;
+    }
+  }
+
+  if(data->bits.get_filetime && ftp->file) {
+    /* we have requested to get the modified-time of the file, this is yet
+       again a grey area as the MDTM is not kosher RFC959 */
+    ftpsendf(data->firstsocket, conn, "MDTM %s", ftp->file);
+
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
+    if(nread < 0)
+      return CURLE_OPERATION_TIMEOUTED;
+
+    if(ftpcode == 213) {
+      /* we got a time. Format should be: "YYYYMMDDHHMMSS[.sss]" where the
+         last .sss part is optional and means fractions of a second */
+      int year, month, day, hour, minute, second;
+      if(6 == sscanf(buf+4, "%04d%02d%02d%02d%02d%02d",
+                     &year, &month, &day, &hour, &minute, &second)) {
+        /* we have a time, reformat it */
+        time_t secs=time(NULL);
+        sprintf(buf, "%04d%02d%02d %02d:%02d:%02d",
+                year, month, day, hour, minute, second);
+        /* now, convert this into a time() value: */
+        data->progress.filetime = curl_getdate(buf, &secs);
+      }
+      else {
+        infof(data, "unsupported MDTM reply format\n");
+      }
+    }
+
+  }
+
   /* If we have selected NOBODY, it means that we only want file information.
      Which in FTP can't be much more than the file size! */
   if(data->bits.no_body) {
@@ -579,33 +648,59 @@ CURLcode _ftp(struct connectdata *conn)
        may not support it! It is however the only way we have to get a file's
        size! */
     int filesize;
-    sendf(data->firstsocket, data, "SIZE %s\r\n", ftp->file);
 
-    nread = GetLastResponse(data->firstsocket, buf, conn);
+    /* Some servers return different sizes for different modes, and thus we
+       must set the proper type before we check the size */
+    ftpsendf(data->firstsocket, conn, "TYPE %s",
+             (data->bits.ftp_ascii)?"A":"I");
+
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
-    if(strncmp(buf, "213", 3)) {
+    if(ftpcode != 200) {
+      failf(data, "Couldn't set %s mode",
+            (data->bits.ftp_ascii)?"ASCII":"binary");
+      return (data->bits.ftp_ascii)? CURLE_FTP_COULDNT_SET_ASCII:
+        CURLE_FTP_COULDNT_SET_BINARY;
+    }
+
+    ftpsendf(data->firstsocket, conn, "SIZE %s", ftp->file);
+
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
+    if(nread < 0)
+      return CURLE_OPERATION_TIMEOUTED;
+
+    if(ftpcode != 213) {
       failf(data, "Couldn't get file size: %s", buf+4);
       return CURLE_FTP_COULDNT_GET_SIZE;
     }
     /* get the size from the ascii string: */
     filesize = atoi(buf+4);
 
-    sprintf(buf, "Content-Length: %d\n", filesize);
+    sprintf(buf, "Content-Length: %d\r\n", filesize);
+    result = client_write(data, CLIENTWRITE_BOTH, buf, 0);
+    if(result)
+      return result;
 
-    if(strlen(buf) != data->fwrite(buf, 1, strlen(buf), data->out)) {
-      failf (data, "Failed writing output");
-      return CURLE_WRITE_ERROR;
-    }
-    if(data->writeheader) {
-      /* the header is requested to be written to this file */
-      if(strlen(buf) != data->fwrite (buf, 1, strlen(buf),
-                                      data->writeheader)) {
-        failf (data, "Failed writing output");
-        return CURLE_WRITE_ERROR;
-      }
+#ifdef HAVE_STRFTIME
+    if(data->bits.get_filetime && data->progress.filetime) {
+      struct tm *tm;
+#ifdef HAVE_LOCALTIME_R
+      struct tm buffer;
+      tm = (struct tm *)localtime_r(&data->progress.filetime, &buffer);
+#else
+      tm = localtime(&data->progress.filetime);
+#endif
+      /* format: "Tue, 15 Nov 1994 12:45:26 GMT" */
+      strftime(buf, BUFSIZE-1, "Last-Modified: %a, %d %b %Y %H:%M:%S %Z\r\n",
+               tm);
+      result = client_write(data, CLIENTWRITE_BOTH, buf, 0);
+      if(result)
+        return result;
     }
+#endif
+
     return CURLE_OK;
   }
 
@@ -613,23 +708,24 @@ CURLcode _ftp(struct connectdata *conn)
   if(data->bits.ftp_use_port) {
     struct sockaddr_in sa;
     struct hostent *h=NULL;
+    char *hostdataptr=NULL;
     size_t size;
     unsigned short porttouse;
     char myhost[256] = "";
 
     if(data->ftpport) {
       if(if2ip(data->ftpport, myhost, sizeof(myhost))) {
-        h = GetHost(data, myhost, hostent_buf, sizeof(hostent_buf));
+        h = GetHost(data, myhost, &hostdataptr);
       }
       else {
         if(strlen(data->ftpport)>1)
-          h = GetHost(data, data->ftpport, hostent_buf, sizeof(hostent_buf));
+          h = GetHost(data, data->ftpport, &hostdataptr);
         if(h)
-          strcpy(myhost,data->ftpport);
+          strcpy(myhost, data->ftpport); /* buffer overflow risk */
       }
     }
     if(! *myhost) {
-      h=GetHost(data, getmyhost(myhost,sizeof(myhost)), hostent_buf, sizeof(hostent_buf));
+      h=GetHost(data, getmyhost(myhost, sizeof(myhost)), &hostdataptr);
     }
     infof(data, "We connect from %s\n", myhost);
 
@@ -658,18 +754,24 @@ CURLcode _ftp(struct connectdata *conn)
 
           if ( listen(portsock, 1) < 0 ) {
             failf(data, "listen(2) failed on socket");
+            free(hostdataptr);
             return CURLE_FTP_PORT_FAILED;
           }
         }
         else {
           failf(data, "bind(2) failed on socket");
+          free(hostdataptr);
           return CURLE_FTP_PORT_FAILED;
         }
       }
       else {
         failf(data, "socket(2) failed (%s)");
+        free(hostdataptr);
         return CURLE_FTP_PORT_FAILED;
       }
+      if(hostdataptr)
+        /* free the memory used for name lookup */
+        free(hostdataptr);
     }
     else {
       failf(data, "could't find my own IP address (%s)", myhost);
@@ -689,40 +791,42 @@ CURLcode _ftp(struct connectdata *conn)
       sscanf( inet_ntoa(in), "%hu.%hu.%hu.%hu",
               &ip[0], &ip[1], &ip[2], &ip[3]);
 #endif
-      sendf(data->firstsocket, data, "PORT %d,%d,%d,%d,%d,%d\r\n",
+      ftpsendf(data->firstsocket, conn, "PORT %d,%d,%d,%d,%d,%d",
             ip[0], ip[1], ip[2], ip[3],
             porttouse >> 8,
             porttouse & 255);
     }
 
-    nread = GetLastResponse(data->firstsocket, buf, conn);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
-    if(strncmp(buf, "200", 3)) {
+    if(ftpcode != 200) {
       failf(data, "Server does not grok PORT, try without it!");
       return CURLE_FTP_PORT_FAILED;
     }     
   }
   else { /* we use the PASV command */
 
-    sendf(data->firstsocket, data, "PASV\r\n");
+    ftpsendf(data->firstsocket, conn, "PASV");
 
-    nread = GetLastResponse(data->firstsocket, buf, conn);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
-    if(strncmp(buf, "227", 3)) {
+    if(ftpcode != 227) {
       failf(data, "Odd return code after PASV");
       return CURLE_FTP_WEIRD_PASV_REPLY;
     }
     else {
       int ip[4];
       int port[2];
-      unsigned short newport;
+      unsigned short newport; /* remote port, not necessary the local one */
+      unsigned short connectport; /* the local port connect() should use! */
       char newhost[32];
       struct hostent *he;
       char *str=buf,*ip_addr;
+      char *hostdataptr=NULL;
 
       /*
        * New 227-parser June 3rd 1999.
@@ -746,21 +850,36 @@ CURLcode _ftp(struct connectdata *conn)
 	 failf(data, "Couldn't interpret this 227-reply: %s", buf);
 	 return CURLE_FTP_WEIRD_227_FORMAT;
       }
+
       sprintf(newhost, "%d.%d.%d.%d", ip[0], ip[1], ip[2], ip[3]);
-      he = GetHost(data, newhost, hostent_buf, sizeof(hostent_buf));
+      newport = (port[0]<<8) + port[1];
+      if(data->bits.httpproxy) {
+        /*
+         * This is a tunnel through a http proxy and we need to connect to the
+         * proxy again here. We already have the name info for it since the
+         * previous lookup.
+         */
+        he = conn->hp;
+        connectport =
+          (unsigned short)data->port; /* we connect to the proxy's port */
+      }
+      else {
+        /* normal, direct, ftp connection */
+        he = GetHost(data, newhost, &hostdataptr);
         if(!he) {
           failf(data, "Can't resolve new host %s", newhost);
           return CURLE_FTP_CANT_GET_HOST;
         }
+        connectport = newport; /* we connect to the remote port */
+      }
 	
-	
-      newport = (port[0]<<8) + port[1];
       data->secondarysocket = socket(AF_INET, SOCK_STREAM, 0);
 
       memset((char *) &serv_addr, '\0', sizeof(serv_addr));
       memcpy((char *)&(serv_addr.sin_addr), he->h_addr, he->h_length);
       serv_addr.sin_family = he->h_addrtype;
-      serv_addr.sin_port = htons(newport);
+
+      serv_addr.sin_port = htons(connectport);
 
       if(data->bits.verbose) {
         struct in_addr in;
@@ -822,9 +941,12 @@ CURLcode _ftp(struct connectdata *conn)
 #else
               ip_addr = inet_ntoa(in),
 #endif
-              newport);
+              connectport);
       }
 	
+      if(hostdataptr)
+        free(hostdataptr);
+
       if (connect(data->secondarysocket, (struct sockaddr *) &serv_addr,
                   sizeof(serv_addr)) < 0) {
         switch(errno) {
@@ -845,37 +967,30 @@ CURLcode _ftp(struct connectdata *conn)
         }
         return CURLE_FTP_CANT_RECONNECT;
       }
-    }
 
+      if (data->bits.tunnel_thru_httpproxy) {
+        /* We want "seamless" FTP operations through HTTP proxy tunnel */
+        result = GetHTTPProxyTunnel(data, data->secondarysocket,
+                                    newhost, newport);
+        if(CURLE_OK != result)
+          return result;
+      }
+    }
   }
   /* we have the (new) data connection ready */
-  infof(data, "Connected!\n");
-
-  /* change directory first */
-
-  if(ftp->dir && ftp->dir[0]) {
-    sendf(data->firstsocket, data, "CWD %s\r\n", ftp->dir);
-    nread = GetLastResponse(data->firstsocket, buf, conn);
-    if(nread < 0)
-      return CURLE_OPERATION_TIMEOUTED;
-
-    if(strncmp(buf, "250", 3)) {
-      failf(data, "Couldn't change to directory %s", ftp->dir);
-      return CURLE_FTP_ACCESS_DENIED;
-    }
-  }
+  infof(data, "Connected the data stream!\n");
 
   if(data->bits.upload) {
 
     /* Set type to binary (unless specified ASCII) */
-    sendf(data->firstsocket, data, "TYPE %s\r\n",
+    ftpsendf(data->firstsocket, conn, "TYPE %s",
           (data->bits.ftp_ascii)?"A":"I");
 
-    nread = GetLastResponse(data->firstsocket, buf, conn);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
-    if(strncmp(buf, "200", 3)) {
+    if(ftpcode != 200) {
       failf(data, "Couldn't set %s mode",
             (data->bits.ftp_ascii)?"ASCII":"binary");
       return (data->bits.ftp_ascii)? CURLE_FTP_COULDNT_SET_ASCII:
@@ -900,13 +1015,13 @@ CURLcode _ftp(struct connectdata *conn)
         /* we could've got a specified offset from the command line,
            but now we know we didn't */
 
-        sendf(data->firstsocket, data, "SIZE %s\r\n", ftp->file);
+        ftpsendf(data->firstsocket, conn, "SIZE %s", ftp->file);
 
-        nread = GetLastResponse(data->firstsocket, buf, conn);
+        nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
         if(nread < 0)
           return CURLE_OPERATION_TIMEOUTED;
 
-        if(strncmp(buf, "213", 3)) {
+        if(ftpcode != 213) {
           failf(data, "Couldn't get file size: %s", buf+4);
           return CURLE_FTP_COULDNT_GET_SIZE;
         }
@@ -918,25 +1033,9 @@ CURLcode _ftp(struct connectdata *conn)
       if(data->resume_from) {
         /* do we still game? */
         int passed=0;
-#if 0
-        /* Set resume file transfer offset */
-        infof(data, "Instructs server to resume from offset %d\n",
-              data->resume_from);
-
-        sendf(data->firstsocket, data, "REST %d\r\n", data->resume_from);
-
-        nread = GetLastResponse(data->firstsocket, buf, conn);
-        if(nread < 0)
-          return CURLE_OPERATION_TIMEOUTED;
-
-        if(strncmp(buf, "350", 3)) {
-          failf(data, "Couldn't use REST: %s", buf+4);
-          return CURLE_FTP_COULDNT_USE_REST;
-        }
-#else
         /* enable append instead */
         data->bits.ftp_append = 1;
-#endif
+
         /* Now, let's read off the proper amount of bytes from the
            input. If we knew it was a proper file we could've just
            fseek()ed but we only have a stream here */
@@ -964,8 +1063,8 @@ CURLcode _ftp(struct connectdata *conn)
           data->infilesize -= data->resume_from;
 
           if(data->infilesize <= 0) {
-            infof(data, "File already completely uploaded\n");
-            return CURLE_OK;
+            failf(data, "File already completely uploaded\n");
+            return CURLE_FTP_COULDNT_STOR_FILE;
           }
         }
         /* we've passed, proceed as normal */
@@ -975,15 +1074,15 @@ CURLcode _ftp(struct connectdata *conn)
     /* Send everything on data->in to the socket */
     if(data->bits.ftp_append)
       /* we append onto the file instead of rewriting it */
-      sendf(data->firstsocket, data, "APPE %s\r\n", ftp->file);
+      ftpsendf(data->firstsocket, conn, "APPE %s", ftp->file);
     else
-      sendf(data->firstsocket, data, "STOR %s\r\n", ftp->file);
+      ftpsendf(data->firstsocket, conn, "STOR %s", ftp->file);
 
-    nread = GetLastResponse(data->firstsocket, buf, conn);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
-    if(atoi(buf)>=400) {
+    if(ftpcode>=400) {
       failf(data, "Failed FTP upload:%s", buf+3);
       /* oops, we never close the sockets! */
       return CURLE_FTP_COULDNT_STOR_FILE;
@@ -1001,9 +1100,7 @@ CURLcode _ftp(struct connectdata *conn)
        size prior to the actual upload. */
 
     pgrsSetUploadSize(data, data->infilesize);
-#if 0
-    ProgressInit(data, data->infilesize);
-#endif
+
     result = Transfer(conn, -1, -1, FALSE, NULL, /* no download */
                       data->secondarysocket, bytecountp);
     if(result)
@@ -1051,11 +1148,7 @@ CURLcode _ftp(struct connectdata *conn)
       infof(data, "range-download from %d to %d, totally %d bytes\n",
             from, to, totalsize);
     }
-#if 0
-    if(!ppath[0])
-      /* make sure this becomes a valid name */
-      ppath="./";
-#endif
+
     if((data->bits.ftp_list_only) || !ftp->file) {
       /* The specified path ends with a slash, and therefore we think this
          is a directory that is requested, use LIST. But before that we
@@ -1063,13 +1156,13 @@ CURLcode _ftp(struct connectdata *conn)
       dirlist = TRUE;
 
       /* Set type to ASCII */
-      sendf(data->firstsocket, data, "TYPE A\r\n");
+      ftpsendf(data->firstsocket, conn, "TYPE A");
 	
-      nread = GetLastResponse(data->firstsocket, buf, conn);
+      nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
       if(nread < 0)
         return CURLE_OPERATION_TIMEOUTED;
 	
-      if(strncmp(buf, "200", 3)) {
+      if(ftpcode != 200) {
         failf(data, "Couldn't set ascii mode");
         return CURLE_FTP_COULDNT_SET_ASCII;
       }
@@ -1078,20 +1171,20 @@ CURLcode _ftp(struct connectdata *conn)
          better used since the LIST command output is not specified or
          standard in any way */
 
-      sendf(data->firstsocket, data, "%s\r\n",
+      ftpsendf(data->firstsocket, conn, "%s",
             data->customrequest?data->customrequest:
             (data->bits.ftp_list_only?"NLST":"LIST"));
     }
     else {
       /* Set type to binary (unless specified ASCII) */
-      sendf(data->firstsocket, data, "TYPE %s\r\n",
-            (data->bits.ftp_list_only)?"A":"I");
+      ftpsendf(data->firstsocket, conn, "TYPE %s",
+               (data->bits.ftp_ascii)?"A":"I");
 
-      nread = GetLastResponse(data->firstsocket, buf, conn);
+      nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
       if(nread < 0)
         return CURLE_OPERATION_TIMEOUTED;
 
-      if(strncmp(buf, "200", 3)) {
+      if(ftpcode != 200) {
         failf(data, "Couldn't set %s mode",
               (data->bits.ftp_ascii)?"ASCII":"binary");
         return (data->bits.ftp_ascii)? CURLE_FTP_COULDNT_SET_ASCII:
@@ -1106,13 +1199,13 @@ CURLcode _ftp(struct connectdata *conn)
          * of the file we're gonna get. If we can get the size, this is by far
          * the best way to know if we're trying to resume beyond the EOF.  */
 
-        sendf(data->firstsocket, data, "SIZE %s\r\n", ftp->file);
+        ftpsendf(data->firstsocket, conn, "SIZE %s", ftp->file);
 
-        nread = GetLastResponse(data->firstsocket, buf, conn);
+        nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
         if(nread < 0)
           return CURLE_OPERATION_TIMEOUTED;
 
-        if(strncmp(buf, "213", 3)) {
+        if(ftpcode != 213) {
           infof(data, "server doesn't support SIZE: %s", buf+4);
           /* We couldn't get the size and therefore we can't know if there
              really is a part of the file left to get, although the server
@@ -1150,26 +1243,26 @@ CURLcode _ftp(struct connectdata *conn)
         infof(data, "Instructs server to resume from offset %d\n",
               data->resume_from);
 
-        sendf(data->firstsocket, data, "REST %d\r\n", data->resume_from);
+        ftpsendf(data->firstsocket, conn, "REST %d", data->resume_from);
 
-        nread = GetLastResponse(data->firstsocket, buf, conn);
+        nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
         if(nread < 0)
           return CURLE_OPERATION_TIMEOUTED;
 
-        if(strncmp(buf, "350", 3)) {
+        if(ftpcode != 350) {
           failf(data, "Couldn't use REST: %s", buf+4);
           return CURLE_FTP_COULDNT_USE_REST;
         }
       }
 
-      sendf(data->firstsocket, data, "RETR %s\r\n", ftp->file);
+      ftpsendf(data->firstsocket, conn, "RETR %s", ftp->file);
     }
 
-    nread = GetLastResponse(data->firstsocket, buf, conn);
+    nread = GetLastResponse(data->firstsocket, buf, conn, &ftpcode);
     if(nread < 0)
       return CURLE_OPERATION_TIMEOUTED;
 
-    if(!strncmp(buf, "150", 3) || !strncmp(buf, "125", 3)) {
+    if((ftpcode == 150) || (ftpcode == 125)) {
 
       /*
         A;
@@ -1224,25 +1317,10 @@ CURLcode _ftp(struct connectdata *conn)
           }
             
         }
-#if 0
-        if(2 != sscanf(buf, "%*[^(](%d bytes%c", &size, &paren))
-          size=-1;
-#endif
       }
       else if(downloadsize > -1)
         size = downloadsize;
 
-#if 0
-      if((size > -1) && (data->resume_from>0)) {
-        size -= data->resume_from;
-        if(size <= 0) {
-          failf(data, "Offset (%d) was beyond file size (%d)",
-                data->resume_from, data->resume_from+size);
-          return CURLE_PARTIAL_FILE;
-        }
-      }
-#endif
-
       if(data->bits.ftp_use_port) {
         result = AllowServerConnect(data, portsock);
         if( result )
@@ -1287,13 +1365,15 @@ CURLcode ftp(struct connectdata *conn)
      it */
   ftp->file = strrchr(conn->ppath, '/');
   if(ftp->file) {
+    if(ftp->file != conn->ppath)
+      dirlength=ftp->file-conn->ppath; /* don't count the traling slash */
+
     ftp->file++; /* point to the first letter in the file name part or
                     remain NULL */
   }
   else {
     ftp->file = conn->ppath; /* there's only a file part */
   }
-  dirlength=ftp->file-conn->ppath;
 
   if(*ftp->file) {
     ftp->file = curl_unescape(ftp->file, 0);
@@ -1321,6 +1401,14 @@ CURLcode ftp(struct connectdata *conn)
 
   retcode = _ftp(conn);
 
+  /* clean up here, success or error doesn't matter */
+  if(ftp->file)
+    free(ftp->file);
+  if(ftp->dir)
+    free(ftp->dir);
+
+  ftp->file = ftp->dir = NULL; /* zero */
+
   return retcode;
 }
 

lib/getdate.c

@@ -390,7 +390,7 @@ static const short yycheck[] = {     0,
     56
 };
 /* -*-C-*-  Note some compilers choke on comments on `#line' lines.  */
-#line 3 "/opt/TWWfsw/bison/share/bison.simple"
+#line 3 "/usr/local/share/bison.simple"
 /* This file comes from bison-1.28.  */
 
 /* Skeleton output parser for bison,
@@ -604,7 +604,7 @@ __yy_memcpy (char *to, char *from, unsigned int count)
 #endif
 #endif
 
-#line 217 "/opt/TWWfsw/bison/share/bison.simple"
+#line 217 "/usr/local/share/bison.simple"
 
 /* The user can define YYPARSE_PARAM as the name of an argument to be passed
    into yyparse.  The argument should have type void *.
@@ -1295,7 +1295,7 @@ case 50:
     break;}
 }
    /* the action file gets copied in in place of this dollarsign */
-#line 543 "/opt/TWWfsw/bison/share/bison.simple"
+#line 543 "/usr/local/share/bison.simple"
 
   yyvsp -= yylen;
   yyssp -= yylen;
@@ -1981,7 +1981,7 @@ curl_getdate (const char *p, const time_t *now)
   yyInput = p;
   Start = now ? *now : time ((time_t *) NULL);
 #ifdef HAVE_LOCALTIME_R
-  tmp = localtime_r(&Start, &keeptime);
+  tmp = (struct tm *)localtime_r(&Start, &keeptime);
 #else
   tmp = localtime (&Start);
 #endif

lib/getdate.y

@@ -934,7 +934,7 @@ curl_getdate (const char *p, const time_t *now)
   yyInput = p;
   Start = now ? *now : time ((time_t *) NULL);
 #ifdef HAVE_LOCALTIME_R
-  tmp = localtime_r(&Start, &keeptime);
+  tmp = (struct tm *)localtime_r(&Start, &keeptime);
 #else
   tmp = localtime (&Start);
 #endif

lib/getenv.c

@@ -45,6 +45,10 @@
 #include <windows.h>
 #endif
 
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 char *GetEnv(char *variable)
 {
 #ifdef WIN32

lib/getinfo.c

@@ -0,0 +1,127 @@
+/*****************************************************************************
+ *                                  _   _ ____  _     
+ *  Project                     ___| | | |  _ \| |    
+ *                             / __| | | | |_) | |    
+ *                            | (__| |_| |  _ <| |___ 
+ *                             \___|\___/|_| \_\_____|
+ *
+ *  The contents of this file are subject to the Mozilla Public License
+ *  Version 1.0 (the "License"); you may not use this file except in
+ *  compliance with the License. You may obtain a copy of the License at
+ *  http://www.mozilla.org/MPL/
+ *
+ *  Software distributed under the License is distributed on an "AS IS"
+ *  basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
+ *  License for the specific language governing rights and limitations
+ *  under the License.
+ *
+ *  The Original Code is Curl.
+ *
+ *  The Initial Developer of the Original Code is Daniel Stenberg.
+ *
+ *  Portions created by the Initial Developer are Copyright (C) 1999.
+ *  All Rights Reserved.
+ *
+ * ------------------------------------------------------------
+ * Main author:
+ * - Daniel Stenberg <daniel@haxx.se>
+ *
+ * 	http://curl.haxx.se
+ *
+ * $Source$
+ * $Revision$
+ * $Date$
+ * $Author$
+ * $State$
+ * $Locker$
+ *
+ * ------------------------------------------------------------
+ ****************************************************************************/
+
+#include "setup.h"
+
+#include <curl/curl.h>
+
+#include "urldata.h"
+
+#include <stdio.h>
+#include <string.h>
+#include <stdarg.h>
+
+CURLcode curl_getinfo(CURL *curl, CURLINFO info, ...)
+{
+  va_list arg;
+  long *param_longp;
+  double *param_doublep;
+  char **param_charp;
+  struct UrlData *data = (struct UrlData *)curl;
+  va_start(arg, info);
+
+  switch(info&CURLINFO_TYPEMASK) {
+  default:
+    return CURLE_BAD_FUNCTION_ARGUMENT;
+  case CURLINFO_STRING:
+    param_charp = va_arg(arg, char **);  
+    if(NULL == param_charp)
+      return CURLE_BAD_FUNCTION_ARGUMENT;
+    break;
+  case CURLINFO_LONG:
+    param_longp = va_arg(arg, long *);
+    if(NULL == param_longp)
+      return CURLE_BAD_FUNCTION_ARGUMENT;
+    break;
+  case CURLINFO_DOUBLE:
+    param_doublep = va_arg(arg, double *);
+    if(NULL == param_doublep)
+      return CURLE_BAD_FUNCTION_ARGUMENT;
+    break;
+  }
+  
+  switch(info) {
+  case CURLINFO_EFFECTIVE_URL:
+    *param_charp = data->url?data->url:"";
+    break;
+  case CURLINFO_HTTP_CODE:
+    *param_longp = data->progress.httpcode;
+    break;
+  case CURLINFO_FILETIME:
+    *param_longp = data->progress.filetime;
+    break;
+  case CURLINFO_HEADER_SIZE:
+    *param_longp = data->header_size;
+    break;
+  case CURLINFO_REQUEST_SIZE:
+    *param_longp = data->request_size;
+    break;
+  case CURLINFO_TOTAL_TIME:
+    *param_doublep = data->progress.timespent;
+    break;
+  case CURLINFO_NAMELOOKUP_TIME:
+    *param_doublep = data->progress.t_nslookup;
+    break;
+  case CURLINFO_CONNECT_TIME:
+    *param_doublep = data->progress.t_connect;
+    break;
+  case CURLINFO_PRETRANSFER_TIME:
+    *param_doublep =  data->progress.t_pretransfer;
+    break;
+  case CURLINFO_SIZE_UPLOAD:
+    *param_doublep =  data->progress.uploaded;
+    break;
+  case CURLINFO_SIZE_DOWNLOAD:
+    *param_doublep = data->progress.downloaded;
+    break;
+  case CURLINFO_SPEED_DOWNLOAD:
+    *param_doublep =  data->progress.dlspeed;
+    break;
+  case CURLINFO_SPEED_UPLOAD:
+    *param_doublep = data->progress.ulspeed;
+    break;
+  case CURLINFO_SSL_VERIFYRESULT:
+    *param_longp = data->ssl.certverifyresult;
+    break;
+  default:
+    return CURLE_BAD_FUNCTION_ARGUMENT;
+  }
+  return CURLE_OK;
+}

lib/getpass.c

@@ -4,10 +4,11 @@
  * Redistribution and use are freely permitted provided that:
  *
  *   1) This header remain in tact.
- *   2) The prototype for getpass is not changed from:
+ *   2) The prototypes for getpass and getpass_r are not changed from:
  *         char *getpass(const char *prompt)
+ *         char *getpass_r(const char *prompt, char* buffer, int buflen)
  *   3) This source code is not used outside of this(getpass.c) file.
- *   3) Any changes to this(getpass.c) source code are made publicly available.
+ *   4) Any changes to this(getpass.c) source code are made publicly available.
  *
  * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED WARRANTIES,
  * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
@@ -34,19 +35,19 @@
  *   Daniel Stenberg <daniel@haxx.se>
  */
 
-#ifndef WIN32
 #ifdef HAVE_CONFIG_H
 #  include <config.h>
 #endif
 
+#ifndef HAVE_GETPASS_R
+
+#ifndef WIN32
 #ifdef HAVE_TERMIOS_H
 #  if !defined(HAVE_TCGETATTR) && !defined(HAVE_TCSETATTR) 
 #    undef HAVE_TERMIOS_H
 #  endif
 #endif
 
-#define INPUT_BUFFER 128
-
 #ifndef RETSIGTYPE
 #  define RETSIGTYPE void
 #endif
@@ -70,11 +71,10 @@
 #  define perror(x) fprintf(stderr, "Error in: %s\n", x)
 #endif
 
-char *getpass(const char *prompt)
+char *getpass_r(const char *prompt, char *buffer, int buflen)
 {
   FILE *infp;
   FILE *outfp;
-  static char buf[INPUT_BUFFER];
   RETSIGTYPE (*sigint)();
 #ifndef __EMX__
   RETSIGTYPE (*sigtstp)();
@@ -115,25 +115,25 @@ char *getpass(const char *prompt)
 #ifdef HAVE_TERMIOS_H
   if(tcgetattr(outfd, &orig) != 0)
   {
-    perror("tcgetattr");
+    ; /*perror("tcgetattr");*/
   }
   noecho = orig;
   noecho.c_lflag &= ~ECHO;
   if(tcsetattr(outfd, TCSANOW, &noecho) != 0)
   {
-    perror("tcgetattr");
+    ; /*perror("tcgetattr");*/
   }
 #else
 #  ifdef HAVE_TERMIO_H
   if(ioctl(outfd, TCGETA, &orig) != 0)
   {
-    perror("ioctl");
+    ; /*perror("ioctl");*/
   }
   noecho = orig;
   noecho.c_lflag &= ~ECHO;
   if(ioctl(outfd, TCSETA, &noecho) != 0)
   {
-    perror("ioctl");
+    ; /*perror("ioctl");*/
   }
 #  else
 #  endif
@@ -142,8 +142,8 @@ char *getpass(const char *prompt)
   fputs(prompt, outfp);
   fflush(outfp);
 
-  bytes_read=read(infd, buf, INPUT_BUFFER);
-  buf[bytes_read > 0 ? (bytes_read -1) : 0] = '\0';
+  bytes_read=read(infd, buffer, buflen);
+  buffer[bytes_read > 0 ? (bytes_read -1) : 0] = '\0';
 
   /* print a new line if needed */
 #ifdef HAVE_TERMIOS_H
@@ -157,18 +157,18 @@ char *getpass(const char *prompt)
 
   /*
    * reset term charectaristics, use TCSAFLUSH incase the
-   * user types more than INPUT_BUFFER
+   * user types more than buflen
    */
 #ifdef HAVE_TERMIOS_H
   if(tcsetattr(outfd, TCSAFLUSH, &orig) != 0)
   {
-    perror("tcgetattr");
+    ; /*perror("tcgetattr");*/
   }
 #else
 #  ifdef HAVE_TERMIO_H
   if(ioctl(outfd, TCSETA, &orig) != 0)
   {
-    perror("ioctl");
+    ; /*perror("ioctl");*/
   }
 #  else
 #  endif
@@ -179,15 +179,38 @@ char *getpass(const char *prompt)
   signal(SIGTSTP, sigtstp);
 #endif
 
-  return(buf);
+  return buffer; /* we always return success */
 }
-#else
+#else /* WIN32 */
 #include <stdio.h>
+#include <conio.h>
+char *getpass_r(const char *prompt, char *buffer, int buflen)
+{
+  int i;
+  printf("%s", prompt);
+ 
+  for(i=0; i<buflen; i++) {
+    buffer[i] = getch();
+    if ( buffer[i] == '\r' ) {
+      buffer[i] = 0;
+      break;
+    }
+  }
+  /* if user didn't hit ENTER, terminate buffer */
+  if (i==buflen)
+    buffer[buflen-1]=0;
+
+  return buffer; /* we always return success */
+}
+#endif
+
+#endif /* ifndef HAVE_GETPASS_R */
+
+#if 0
+/* for consistensy, here's the old-style function: */
 char *getpass(const char *prompt)
 {
-	static char password[80];
-	printf(prompt);
-	gets(password);
-	return password;
+  static char buf[256];
+  return getpass_r(prompt, buf, sizeof(buf));
 }
-#endif /* don't do anything if WIN32 */
+#endif

lib/getpass.h

@@ -1 +1,8 @@
-char *getpass(const char *prompt);
+#ifndef __GETPASS_H
+#define __GETPASS_H
+/*
+ * Returning NULL will abort the continued operation!
+ */
+char* getpass_r(char *prompt, char* buffer, int buflen );
+
+#endif

lib/highlevel.c

@@ -107,11 +107,19 @@
 #include "getpass.h"
 #include "progress.h"
 #include "getdate.h"
-#include "writeout.h"
 
 #define _MPRINTF_REPLACE /* use our functions only */
 #include <curl/mprintf.h>
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
+#ifndef min
+#define min(a, b)   ((a) < (b) ? (a) : (b))
+#endif
+
 CURLcode 
 _Transfer(struct connectdata *c_conn)
 {
@@ -141,6 +149,7 @@ _Transfer(struct connectdata *c_conn)
   long bodywrites=0;
 
   char newurl[URL_MAX_LENGTH];		/* buffer for Location: URL */
+  int writetype;
 
   /* the highest fd we use + 1 */
   struct UrlData *data;
@@ -166,13 +175,16 @@ _Transfer(struct connectdata *c_conn)
 #define KEEP_WRITE 2
 
   pgrsTime(data, TIMER_PRETRANSFER);
+  speedinit(data);
 
   if (!conn->getheader) {
     header = FALSE;
     if(conn->size > 0)
       pgrsSetDownloadSize(data, conn->size);
   }
-  {
+  /* we want header and/or body, if neither then don't do this! */
+  if(conn->getheader ||
+     !data->bits.no_body) {
     fd_set readfd;
     fd_set writefd;
     fd_set rkeepfd;
@@ -325,24 +337,16 @@ _Transfer(struct connectdata *c_conn)
 
                 /* now, only output this if the header AND body are requested:
                  */
-                if (data->bits.http_include_header) {
-                  if((p - data->headerbuff) !=
-                     data->fwrite (data->headerbuff, 1,
-                                   p - data->headerbuff, data->out)) {
-                    failf (data, "Failed writing output");
-                    return CURLE_WRITE_ERROR;
-                  }
-                }
-                if(data->writeheader) {
-                  /* obviously, the header is requested to be written to
-                     this file: */
-                  if((p - data->headerbuff) !=
-                     data->fwrite (data->headerbuff, 1, p - data->headerbuff,
-                                   data->writeheader)) {
-                    failf (data, "Failed writing output");
-                    return CURLE_WRITE_ERROR;
-                  }
-                }
+                writetype = CLIENTWRITE_HEADER;
+                if (data->bits.http_include_header)
+                  writetype |= CLIENTWRITE_BODY;
+
+                urg = client_write(data, writetype, data->headerbuff,
+                                   p - data->headerbuff);
+                if(urg)
+                  return urg;
+
+                data->header_size += p - data->headerbuff;
                 break;		/* exit header line loop */
               }
               
@@ -391,9 +395,11 @@ _Transfer(struct connectdata *c_conn)
               }
               else if(strnequal("Last-Modified:", p,
                                 strlen("Last-Modified:")) &&
-                      data->timecondition) {
+                      (data->timecondition || data->bits.get_filetime) ) {
                 time_t secs=time(NULL);
                 timeofdoc = curl_getdate(p+strlen("Last-Modified:"), &secs);
+                if(data->bits.get_filetime)
+                  data->progress.filetime = timeofdoc;
               }
               else if ((code >= 300 && code < 400) &&
                        (data->bits.http_follow_location) &&
@@ -405,20 +411,15 @@ _Transfer(struct connectdata *c_conn)
                 data->newurl = strdup (newurl);
               }
 
-              if (data->bits.http_include_header) {
-                if(hbuflen != data->fwrite (p, 1, hbuflen, data->out)) {
-                  failf (data, "Failed writing output");
-                  return CURLE_WRITE_ERROR;
-                }
-              }
-              if(data->writeheader) {
-                /* the header is requested to be written to this file */
-                if(hbuflen != data->fwrite (p, 1, hbuflen,
-                                            data->writeheader)) {
-                  failf (data, "Failed writing output");
-                  return CURLE_WRITE_ERROR;
-                }
-              }
+              writetype = CLIENTWRITE_HEADER;
+              if (data->bits.http_include_header)
+                writetype |= CLIENTWRITE_BODY;
+
+              urg = client_write(data, writetype, p, hbuflen);
+              if(urg)
+                return urg;
+
+              data->header_size += hbuflen;
               
               /* reset hbufp pointer && hbuflen */
               hbufp = data->headerbuff;
@@ -502,10 +503,9 @@ _Transfer(struct connectdata *c_conn)
 
             pgrsSetDownloadCounter(data, (double)bytecount);
             
-            if (nread != data->fwrite (str, 1, nread, data->out)) {
-              failf (data, "Failed writing output");
-              return CURLE_WRITE_ERROR;
-            }
+            urg = client_write(data, CLIENTWRITE_BODY, str, nread);
+            if(urg)
+              return urg;
 
           } /* if (! header and data to read ) */
         } /* if( read from socket ) */
@@ -520,7 +520,7 @@ _Transfer(struct connectdata *c_conn)
           if(data->crlf)
             buf = data->buffer; /* put it back on the buffer */
 
-          nread = data->fread(buf, 1, BUFSIZE, data->in);
+          nread = data->fread(buf, 1, conn->upload_bufsize, data->in);
 
           /* the signed int typecase of nread of for systems that has
              unsigned size_t */
@@ -568,6 +568,15 @@ _Transfer(struct connectdata *c_conn)
       if (urg)
 	return urg;
 
+      if(data->progress.ulspeed > conn->upload_bufsize) {
+        /* If we're transfering more data per second than fits in our buffer,
+           we increase the buffer size to adjust to the current
+           speed. However, we must not set it larger than BUFSIZE. We don't
+           adjust it downwards again since we don't see any point in that!
+        */
+        conn->upload_bufsize=(long)min(data->progress.ulspeed, BUFSIZE);
+      }
+
       if (data->timeout && (tvdiff (now, start) > data->timeout)) {
 	failf (data, "Operation timed out with %d out of %d bytes received",
 	       bytecount, conn->size);
@@ -598,11 +607,12 @@ CURLcode curl_transfer(CURL *curl)
 {
   CURLcode res;
   struct UrlData *data = curl;
-  struct connectdata *c_connect;
+  struct connectdata *c_connect=NULL;
 
   pgrsStartNow(data);
 
   do {
+    pgrsTime(data, TIMER_STARTSINGLE);
     res = curl_connect(curl, (CURLconnect **)&c_connect);
     if(res == CURLE_OK) {
       res = curl_do(c_connect);
@@ -613,13 +623,24 @@ CURLcode curl_transfer(CURL *curl)
       }
 
       if((res == CURLE_OK) && data->newurl) {
-        /* Location: redirect */
+        /* Location: redirect
+ 
+           This is assumed to happen for HTTP(S) only!
+        */
         char prot[16];
         char path[URL_MAX_LENGTH];
+	if (data->maxredirs && (data->followlocation >= data->maxredirs)) {
+	  failf(data,"Maximum (%d) redirects followed", data->maxredirs);
+          curl_disconnect(c_connect);
+          res=CURLE_TOO_MANY_REDIRECTS;
+	  break;
+	}
 
         /* mark the next request as a followed location: */
         data->bits.this_is_a_follow = TRUE;
 
+        data->followlocation++; /* count location-followers */
+
         if(data->bits.http_auto_referer) {
           /* We are asked to automatically set the previous URL as the
              referer when we get the next URL. We pick the ->url field,
@@ -698,9 +719,13 @@ CURLcode curl_transfer(CURL *curl)
           data->port = 0;
         }
 
+        if(data->bits.urlstringalloc)
+          free(data->url);
+      
         /* TBD: set the URL with curl_setopt() */
         data->url = data->newurl;
         data->newurl = NULL; /* don't show! */
+        data->bits.urlstringalloc = TRUE; /* the URL is allocated */
 
         /* Disable both types of POSTs, since doing a second POST when
            following isn't what anyone would want! */
@@ -722,10 +747,6 @@ CURLcode curl_transfer(CURL *curl)
   if(data->newurl)
     free(data->newurl);
 
-  if((CURLE_OK == res) && data->writeinfo) {
-    /* Time to output some info to stdout */
-    WriteOut(data);
-  }
   return res;
 }
 

lib/hostip.c

@@ -41,7 +41,6 @@
 #include "setup.h"
 
 #include <string.h>
-#include <malloc.h>
 #include <errno.h>
 
 #define _REENTRANT
@@ -73,6 +72,11 @@
 #include "inet_ntoa_r.h"
 #endif
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 /* --- resolve name or IP-number --- */
 
 char *MakeIP(unsigned long num,char *addr, int addr_len)
@@ -103,12 +107,22 @@ char *MakeIP(unsigned long num,char *addr, int addr_len)
 #endif
 struct hostent *GetHost(struct UrlData *data,
                         char *hostname,
-                        char *buf,
-                        int buf_size )
+                        char **bufp)
 {
   struct hostent *h = NULL;
   unsigned long in;
-  int ret;
+  int ret; /* this variable is unused on several platforms but used on some */
+
+#define CURL_NAMELOOKUP_SIZE 9000
+
+  /* Allocate enough memory to hold the full name information structs and
+   * everything. OSF1 is known to require at least 8872 bytes. The buffer
+   * required for storing all possible aliases and IP numbers is according to
+   * Stevens' Unix Network Programming 2nd editor, p. 304: 8192 bytes! */
+  char *buf = (char *)malloc(CURL_NAMELOOKUP_SIZE);
+  if(!buf)
+    return NULL; /* major failure */
+  *bufp = buf;
 
   if ( (in=inet_addr(hostname)) != INADDR_NONE ) {
     struct in_addr *addrentry;
@@ -123,18 +137,20 @@ struct hostent *GetHost(struct UrlData *data,
     h->h_length = sizeof(*addrentry);
     h->h_name = *(h->h_addr_list) + h->h_length;
     /* bad one h->h_name = (char*)(h->h_addr_list + h->h_length); */
-    MakeIP(ntohl(in),h->h_name,buf_size - (long)(h->h_name) + (long)buf);
+    MakeIP(ntohl(in),h->h_name, CURL_NAMELOOKUP_SIZE - (long)(h->h_name) + (long)buf);
   }
 #if defined(HAVE_GETHOSTBYNAME_R)
   else {
     int h_errnop;
-    memset(buf,0,buf_size);	/* workaround for gethostbyname_r bug in qnx nto */
+     /* Workaround for gethostbyname_r bug in qnx nto. It is also _required_
+        for some of these functions. */
+    memset(buf, 0, CURL_NAMELOOKUP_SIZE);
 #ifdef HAVE_GETHOSTBYNAME_R_5
     /* Solaris, IRIX and more */
     if ((h = gethostbyname_r(hostname,
                              (struct hostent *)buf,
                              buf + sizeof(struct hostent),
-                             buf_size - sizeof(struct hostent),
+                             CURL_NAMELOOKUP_SIZE - sizeof(struct hostent),
                              &h_errnop)) == NULL )
 #endif
 #ifdef HAVE_GETHOSTBYNAME_R_6
@@ -142,21 +158,25 @@ struct hostent *GetHost(struct UrlData *data,
     if( gethostbyname_r(hostname,
                         (struct hostent *)buf,
                         buf + sizeof(struct hostent),
-                        buf_size - sizeof(struct hostent),
+                        CURL_NAMELOOKUP_SIZE - sizeof(struct hostent),
                         &h, /* DIFFERENCE */
                         &h_errnop))
 #endif
 #ifdef HAVE_GETHOSTBYNAME_R_3
     /* AIX, Digital Unix, HPUX 10, more? */
 
-    /* August 4th, 2000. I don't have any such system around so I write this
-       blindly in hope it might work or that someone else will help me fix
-       this. August 22nd, 2000: Albert Chin-A-Young brought an updated version
-       that should work! */
+    if(CURL_NAMELOOKUP_SIZE >=
+       (sizeof(struct hostent)+sizeof(struct hostent_data)))
+
+      /* August 22nd, 2000: Albert Chin-A-Young brought an updated version
+       * that should work! September 20: Richard Prescott worked on the buffer
+       * size dilemma. */
 
       ret = gethostbyname_r(hostname,
                           (struct hostent *)buf,
                           (struct hostent_data *)(buf + sizeof(struct hostent)));
+    else
+      ret = -1; /* failure, too smallish buffer size */
     
     /* result expected in h */
     h = (struct hostent*)buf;
@@ -165,11 +185,16 @@ struct hostent *GetHost(struct UrlData *data,
 #endif
       {
       infof(data, "gethostbyname_r(2) failed for %s\n", hostname);
+      h = NULL; /* set return code to NULL */
+      free(buf);
+      *bufp=NULL;
     }
 #else
   else {
     if ((h = gethostbyname(hostname)) == NULL ) {
       infof(data, "gethostbyname(2) failed for %s\n", hostname);
+      free(buf);
+      *bufp=NULL;
     }
 #endif
   }

lib/hostip.h

@@ -40,7 +40,6 @@
  * ------------------------------------------------------------
  ****************************************************************************/
 
-extern struct hostent *GetHost(struct UrlData *data, char *hostname, char *buf, int buf_size );
-extern char *MakeIP(unsigned long num,char *addr, int addr_len);
+struct hostent *GetHost(struct UrlData *data, char *hostname, char **bufp );
 
 #endif

lib/http.c

@@ -117,6 +117,11 @@
 #define _MPRINTF_REPLACE /* use our functions only */
 #include <curl/mprintf.h>
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 /*
  * This function checks the linked list of custom HTTP headers for a particular
  * header (prefix).
@@ -134,38 +139,33 @@ bool static checkheaders(struct UrlData *data, char *thisheader)
   return FALSE;
 }
 
-CURLcode http_connect(struct connectdata *conn)
-{
-  struct UrlData *data;
-
-  data=conn->data;
-
-  /* If we are not using a proxy and we want a secure connection,
-   * perform SSL initialization & connection now.
-   * If using a proxy with https, then we must tell the proxy to CONNECT
-   * us to the host we want to talk to.  Only after the connect
-   * has occured, can we start talking SSL
+/*
+ * GetHTTPProxyTunnel() requires that we're connected to a HTTP proxy. This
+ * function will issue the necessary commands to get a seamless tunnel through
+ * this proxy. After that, the socket can be used just as a normal socket.
  */
-   if (conn->protocol & PROT_HTTPS) {
-     if (data->bits.httpproxy) {
+
+CURLcode GetHTTPProxyTunnel(struct UrlData *data, int tunnelsocket,
+                            char *hostname, int remote_port)
+{
+  int httperror=0;
+  int subversion=0;
+
+  infof(data, "Establish HTTP proxy tunnel to %s:%d\n", hostname, remote_port);
 
   /* OK, now send the connect statment */
-        sendf(data->firstsocket, data,
+  sendf(tunnelsocket, data,
         "CONNECT %s:%d HTTP/1.0\015\012"
         "%s"
         "%s"
         "\r\n",
-              data->hostname, data->remote_port,
+        hostname, remote_port,
         (data->bits.proxy_user_passwd)?data->ptr_proxyuserpwd:"",
         (data->useragent?data->ptr_uagent:"")
         );
 
   /* wait for the proxy to send us a HTTP/1.0 200 OK header */
-	/* Daniel rewrote this part Nov 5 1998 to make it more obvious */
-	{
-	  int httperror=0;
-	  int subversion=0;
-	  while(GetLine(data->firstsocket, data->buffer, data)) {
+  while(GetLine(tunnelsocket, data->buffer, data)) {
     if('\r' == data->buffer[0])
       break; /* end of headers */
     if(2 == sscanf(data->buffer, "HTTP/1.%d %d",
@@ -182,8 +182,31 @@ CURLcode http_connect(struct connectdata *conn)
       failf(data, "Received error code %d from proxy", httperror);
     return CURLE_READ_ERROR;
   }
+
+  infof (data, "Proxy replied to CONNECT request\n");
+  return CURLE_OK;
 }
-        infof (data, "Proxy has replied to CONNECT request\n");
+
+CURLcode http_connect(struct connectdata *conn)
+{
+  struct UrlData *data;
+  CURLcode result;
+
+  data=conn->data;
+
+  /* If we are not using a proxy and we want a secure connection,
+   * perform SSL initialization & connection now.
+   * If using a proxy with https, then we must tell the proxy to CONNECT
+   * us to the host we want to talk to.  Only after the connect
+   * has occured, can we start talking SSL
+   */
+  if (conn->protocol & PROT_HTTPS) {
+    if (data->bits.httpproxy) {
+      /* HTTPS through a proxy can only be done with a tunnel */
+      result = GetHTTPProxyTunnel(data, data->firstsocket,
+                                  data->hostname, data->remote_port);
+      if(CURLE_OK != result)
+        return result;
     }
 
     /* now, perform the SSL initialization for this socket */
@@ -231,7 +254,8 @@ CURLcode http_done(struct connectdata *conn)
     *bytecount = http->readbytecount + http->writebytecount;
   }
 
-  /* TBD: the HTTP struct remains allocated here */
+  free(http);
+  data->proto.http=NULL; /* it is gone */
 
   return CURLE_OK;
 }
@@ -269,7 +293,7 @@ CURLcode http(struct connectdata *conn)
   }
 
   if((data->bits.user_passwd) && !checkheaders(data, "Authorization:")) {
-    char authorization[512];
+    char *authorization;
 
     /* To prevent the user+password to get sent to other than the original
        host due to a location-follow, we do some weirdo checks here */
@@ -277,9 +301,12 @@ CURLcode http(struct connectdata *conn)
        !data->auth_host ||
        strequal(data->auth_host, data->hostname)) {
       sprintf(data->buffer, "%s:%s", data->user, data->passwd);
-      base64Encode(data->buffer, authorization);
+      if(base64_encode(data->buffer, strlen(data->buffer),
+                      &authorization) >= 0) {
         data->ptr_userpwd = maprintf( "Authorization: Basic %s\015\012",
                                       authorization);
+        free(authorization);
+      }
     }
   }
   if((data->bits.set_range) && !checkheaders(data, "Range:")) {
@@ -300,7 +327,7 @@ CURLcode http(struct connectdata *conn)
   }
   if ((data->bits.httpproxy) && !(conn->protocol&PROT_HTTPS))  {
     /* The path sent to the proxy is in fact the entire URL */
-    strncpy(ppath, data->url, URL_MAX_LENGTH-1);
+    ppath = data->url;
   }
   if(data->bits.http_formpost) {
     /* we must build the whole darned post sequence first, so that we have
@@ -309,6 +336,12 @@ CURLcode http(struct connectdata *conn)
   }
 
   if(!checkheaders(data, "Host:")) {
+    if(((conn->protocol&PROT_HTTPS) && (data->remote_port == PORT_HTTPS)) ||
+       (!(conn->protocol&PROT_HTTPS) && (data->remote_port == PORT_HTTP)) )
+      /* If (HTTPS on port 443) OR (non-HTTPS on port 80) then don't include
+         the port number in the host string */
+      data->ptr_host = maprintf("Host: %s\r\n", host);
+    else
       data->ptr_host = maprintf("Host: %s:%d\r\n", host, data->remote_port);
   }
 
@@ -319,8 +352,14 @@ CURLcode http(struct connectdata *conn)
     http->p_accept = "Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, */*\r\n";
 
   do {
+    send_buffer *req_buffer;
     struct curl_slist *headers=data->headers;
-    sendf(data->firstsocket, data,
+
+    /* initialize a dynamic send-buffer */
+    req_buffer = add_buffer_init();
+
+    /* add the main request stuff */
+    add_bufferf(req_buffer,
                 "%s " /* GET/HEAD/POST/PUT */
                 "%s HTTP/1.0\r\n" /* path */
                 "%s" /* proxyuserpwd */
@@ -351,25 +390,23 @@ CURLcode http(struct connectdata *conn)
 
     if(co) {
       int count=0;
+      struct Cookie *store=co;
       /* now loop through all cookies that matched */
       while(co) {
         if(co->value && strlen(co->value)) {
           if(0 == count) {
-            sendf(data->firstsocket, data,
-                  "Cookie:");
+            add_bufferf(req_buffer, "Cookie: ");
           }
-          sendf(data->firstsocket, data,
-                "%s%s=%s", count?"; ":"", co->name,
-                co->value);
+          add_bufferf(req_buffer,
+                      "%s%s=%s", count?"; ":"", co->name, co->value);
           count++;
         }
         co = co->next; /* next cookie please */
       }
       if(count) {
-        sendf(data->firstsocket, data,
-              "\r\n");
+        add_buffer(req_buffer, "\r\n", 2);
       }
-      cookie_freelist(co); /* free the cookie list */
+      cookie_freelist(store); /* free the cookie list */
       co=NULL;
     }
 
@@ -398,15 +435,15 @@ CURLcode http(struct connectdata *conn)
       switch(data->timecondition) {
       case TIMECOND_IFMODSINCE:
       default:
-        sendf(data->firstsocket, data,
+        add_bufferf(req_buffer,
                     "If-Modified-Since: %s\r\n", buf);
         break;
       case TIMECOND_IFUNMODSINCE:
-        sendf(data->firstsocket, data,
+        add_bufferf(req_buffer,
                     "If-Unmodified-Since: %s\r\n", buf);
         break;
       case TIMECOND_LASTMOD:
-        sendf(data->firstsocket, data,
+        add_bufferf(req_buffer,
                     "Last-Modified: %s\r\n", buf);
         break;
       }
@@ -418,15 +455,13 @@ CURLcode http(struct connectdata *conn)
         /* we require a colon for this to be a true header */
 
         ptr++; /* pass the colon */
-        while(*ptr && isspace(*ptr))
+        while(*ptr && isspace((int)*ptr))
           ptr++;
 
         if(*ptr) {
           /* only send this if the contents was non-blank */
 
-          sendf(data->firstsocket, data,
-                "%s\015\012",
-                headers->data);
+          add_bufferf(req_buffer, "%s\r\n", headers->data);
         }
       }
       headers = headers->next;
@@ -447,12 +482,14 @@ CURLcode http(struct connectdata *conn)
                        generated form data */
       data->in = (FILE *)&http->form;
 
-      sendf(data->firstsocket, data,
-            "Content-Length: %d\r\n",
-            http->postsize-2);
+      add_bufferf(req_buffer,
+                  "Content-Length: %d\r\n", http->postsize-2);
 
+      /* set upload size to the progress meter */
       pgrsSetUploadSize(data, http->postsize);
 
+      data->request_size = 
+        add_buffer_send(data->firstsocket, conn, req_buffer);
       result = Transfer(conn, data->firstsocket, -1, TRUE,
                         &http->readbytecount,
                           data->firstsocket,
@@ -466,16 +503,21 @@ CURLcode http(struct connectdata *conn)
       /* Let's PUT the data to the server! */
 
       if(data->infilesize>0) {
-        sendf(data->firstsocket, data,
+        add_bufferf(req_buffer,
                     "Content-Length: %d\r\n\r\n", /* file size */
                     data->infilesize );
       }
       else
-        sendf(data->firstsocket, data,
-              "\015\012");
+        add_bufferf(req_buffer, "\015\012");
 
+      /* set the upload size to the progress meter */
       pgrsSetUploadSize(data, data->infilesize);
 
+      /* this sends the buffer and frees all the buffer resources */
+      data->request_size = 
+        add_buffer_send(data->firstsocket, conn, req_buffer);
+
+      /* prepare for transfer */
       result = Transfer(conn, data->firstsocket, -1, TRUE,
                         &http->readbytecount,
                         data->firstsocket,
@@ -491,28 +533,35 @@ CURLcode http(struct connectdata *conn)
         if(!checkheaders(data, "Content-Length:"))
           /* we allow replacing this header, although it isn't very wise to
              actually set your own */
-          sendf(data->firstsocket, data,
+          add_bufferf(req_buffer,
                       "Content-Length: %d\r\n",
                       (data->postfieldsize?data->postfieldsize:
                        strlen(data->postfields)) );
 
         if(!checkheaders(data, "Content-Type:"))
-          sendf(data->firstsocket, data,
+          add_bufferf(req_buffer,
                       "Content-Type: application/x-www-form-urlencoded\r\n");
 
         /* and here comes the actual data */
         if(data->postfieldsize) {
-          ssend(data->firstsocket, data, "\r\n", 2);
-          ssend(data->firstsocket, data, data->postfields, data->postfieldsize);
-          ssend(data->firstsocket, data, "\r\n", 2);
+          add_buffer(req_buffer, "\r\n", 2);
+          add_buffer(req_buffer, data->postfields,
+                     data->postfieldsize);
+          add_buffer(req_buffer, "\r\n", 2);
         }
-        sendf(data->firstsocket, data,
+        else {
+          add_bufferf(req_buffer,
                       "\r\n"
                       "%s\r\n",
                       data->postfields );
         }
+      }
       else
-        sendf(data->firstsocket, data, "\r\n");
+        add_buffer(req_buffer, "\r\n", 2);
+
+      /* issue the request */
+      data->request_size = 
+        add_buffer_send(data->firstsocket, conn, req_buffer);
 
       /* HTTP GET/HEAD download: */
       result = Transfer(conn, data->firstsocket, -1, TRUE, bytecount,

lib/http.h

@@ -41,8 +41,11 @@
  * ------------------------------------------------------------
  ****************************************************************************/
 
-/* protocol-specific functions set up to be called by the main engine */
+/* ftp can use this as well */
+CURLcode GetHTTPProxyTunnel(struct UrlData *data, int tunnelsocket,
+                            char *hostname, int remote_port);
 
+/* protocol-specific functions set up to be called by the main engine */
 CURLcode http(struct connectdata *conn);
 CURLcode http_done(struct connectdata *conn);
 CURLcode http_connect(struct connectdata *conn);

lib/krb4.c

@@ -0,0 +1,497 @@
+/* modified by Martin Hedenfalk <mhe@stacken.kth.se> for use in Curl
+ * last modified 2000-09-18
+ */
+
+/*
+ * Copyright (c) 1995, 1996, 1997, 1998, 1999 Kungliga Tekniska H<>gskolan
+ * (Royal Institute of Technology, Stockholm, Sweden).
+ * All rights reserved.
+ * 
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 
+ * 3. Neither the name of the Institute nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ * 
+ * THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include "setup.h"
+
+#ifdef KRB4
+
+#include "security.h"
+#include "base64.h"
+#include <stdlib.h>
+#include <netdb.h>
+#include <syslog.h>
+#include <string.h>
+#include <krb.h>
+
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
+#ifdef FTP_SERVER
+#define LOCAL_ADDR ctrl_addr
+#define REMOTE_ADDR his_addr
+#else
+/*#define LOCAL_ADDR myctladdr***/
+/*#define REMOTE_ADDR hisctladdr***/
+#endif
+
+/*extern struct sockaddr *LOCAL_ADDR, *REMOTE_ADDR;***/
+
+#define LOCAL_ADDR (&local_addr)
+#define REMOTE_ADDR (&conn->serv_addr)
+#define myctladdr LOCAL_ADDR
+#define hisctladdr REMOTE_ADDR
+
+static struct sockaddr_in local_addr;
+
+struct krb4_data {
+    des_cblock key;
+    des_key_schedule schedule;
+    char name[ANAME_SZ];
+    char instance[INST_SZ];
+    char realm[REALM_SZ];
+};
+
+#ifndef HAVE_STRLCPY
+
+size_t
+strlcpy (char *dst, const char *src, size_t dst_sz)
+{
+    size_t n;
+    char *p;
+
+    for (p = dst, n = 0;
+	 n + 1 < dst_sz && *src != '\0';
+	 ++p, ++src, ++n)
+	*p = *src;
+    *p = '\0';
+    if (*src == '\0')
+	return n;
+    else
+	return n + strlen (src);
+}
+
+#endif
+
+static int
+krb4_check_prot(void *app_data, int level)
+{
+    if(level == prot_confidential)
+	return -1;
+    return 0;
+}
+
+static int
+krb4_decode(void *app_data, void *buf, int len, int level,
+	    struct connectdata *conn)
+{
+    MSG_DAT m;
+    int e;
+    struct krb4_data *d = app_data;
+    
+    if(level == prot_safe)
+	e = krb_rd_safe(buf, len, &d->key,
+			(struct sockaddr_in *)REMOTE_ADDR,
+			(struct sockaddr_in *)LOCAL_ADDR, &m);
+    else
+	e = krb_rd_priv(buf, len, d->schedule, &d->key, 
+			(struct sockaddr_in *)REMOTE_ADDR,
+			(struct sockaddr_in *)LOCAL_ADDR, &m);
+    if(e){
+	syslog(LOG_ERR, "krb4_decode: %s", krb_get_err_text(e));
+	return -1;
+    }
+    memmove(buf, m.app_data, m.app_length);
+    return m.app_length;
+}
+
+static int
+krb4_overhead(void *app_data, int level, int len)
+{
+    return 31;
+}
+
+static int
+krb4_encode(void *app_data, void *from, int length, int level, void **to,
+	    struct connectdata *conn)
+{
+    struct krb4_data *d = app_data;
+    *to = malloc(length + 31);
+    if(level == prot_safe)
+	return krb_mk_safe(from, *to, length, &d->key, 
+			   (struct sockaddr_in *)LOCAL_ADDR,
+			   (struct sockaddr_in *)REMOTE_ADDR);
+    else if(level == prot_private)
+	return krb_mk_priv(from, *to, length, d->schedule, &d->key, 
+			   (struct sockaddr_in *)LOCAL_ADDR,
+			   (struct sockaddr_in *)REMOTE_ADDR);
+    else
+	return -1;
+}
+
+#ifdef FTP_SERVER
+
+static int
+krb4_adat(void *app_data, void *buf, size_t len)
+{
+    KTEXT_ST tkt;
+    AUTH_DAT auth_dat;
+    char *p;
+    int kerror;
+    u_int32_t cs;
+    char msg[35]; /* size of encrypted block */
+    int tmp_len;
+    struct krb4_data *d = app_data;
+    char inst[INST_SZ];
+    struct sockaddr_in *his_addr_sin = (struct sockaddr_in *)his_addr;
+
+    memcpy(tkt.dat, buf, len);
+    tkt.length = len;
+
+    k_getsockinst(0, inst, sizeof(inst));
+    kerror = krb_rd_req(&tkt, "ftp", inst, 
+			his_addr_sin->sin_addr.s_addr, &auth_dat, "");
+    if(kerror == RD_AP_UNDEC){
+	k_getsockinst(0, inst, sizeof(inst));
+	kerror = krb_rd_req(&tkt, "rcmd", inst, 
+			    his_addr_sin->sin_addr.s_addr, &auth_dat, "");
+    }
+
+    if(kerror){
+	reply(535, "Error reading request: %s.", krb_get_err_text(kerror));
+	return -1;
+    }
+    
+    memcpy(d->key, auth_dat.session, sizeof(d->key));
+    des_set_key(&d->key, d->schedule);
+
+    strlcpy(d->name, auth_dat.pname, sizeof(d->name));
+    strlcpy(d->instance, auth_dat.pinst, sizeof(d->instance));
+    strlcpy(d->realm, auth_dat.prealm, sizeof(d->instance));
+
+    cs = auth_dat.checksum + 1;
+    {
+	unsigned char tmp[4];
+	KRB_PUT_INT(cs, tmp, 4, sizeof(tmp));
+	tmp_len = krb_mk_safe(tmp, msg, 4, &d->key,
+			      (struct sockaddr_in *)LOCAL_ADDR,
+			      (struct sockaddr_in *)REMOTE_ADDR);
+    }
+    if(tmp_len < 0){
+	reply(535, "Error creating reply: %s.", strerror(errno));
+	return -1;
+    }
+    len = tmp_len;
+    if(base64_encode(msg, len, &p) < 0) {
+	reply(535, "Out of memory base64-encoding.");
+	return -1;
+    }
+    reply(235, "ADAT=%s", p);
+    sec_complete = 1;
+    free(p);
+    return 0;
+}
+
+static int
+krb4_userok(void *app_data, char *user)
+{
+    struct krb4_data *d = app_data;
+    return krb_kuserok(d->name, d->instance, d->realm, user);
+}
+
+struct sec_server_mech krb4_server_mech = {
+    "KERBEROS_V4",
+    sizeof(struct krb4_data),
+    NULL, /* init */
+    NULL, /* end */
+    krb4_check_prot,
+    krb4_overhead,
+    krb4_encode,
+    krb4_decode,
+    /* */
+    NULL,
+    krb4_adat,
+    NULL, /* pbsz */
+    NULL, /* ccc */
+    krb4_userok
+};
+
+#else /* FTP_SERVER */
+
+static int
+mk_auth(struct krb4_data *d, KTEXT adat, 
+	char *service, char *host, int checksum)
+{
+    int ret;
+    CREDENTIALS cred;
+    char sname[SNAME_SZ], inst[INST_SZ], realm[REALM_SZ];
+
+    strlcpy(sname, service, sizeof(sname));
+    strlcpy(inst, krb_get_phost(host), sizeof(inst));
+    strlcpy(realm, krb_realmofhost(host), sizeof(realm));
+    ret = krb_mk_req(adat, sname, inst, realm, checksum);
+    if(ret)
+	return ret;
+    strlcpy(sname, service, sizeof(sname));
+    strlcpy(inst, krb_get_phost(host), sizeof(inst));
+    strlcpy(realm, krb_realmofhost(host), sizeof(realm));
+    ret = krb_get_cred(sname, inst, realm, &cred);
+    memmove(&d->key, &cred.session, sizeof(des_cblock));
+    des_key_sched(&d->key, d->schedule);
+    memset(&cred, 0, sizeof(cred));
+    return ret;
+}
+
+static int
+krb4_auth(void *app_data, struct connectdata *conn)
+{
+    int ret;
+    char *p;
+    int len;
+    KTEXT_ST adat;
+    MSG_DAT msg_data;
+    int checksum;
+    u_int32_t cs;
+    struct krb4_data *d = app_data;
+    struct sockaddr_in *localaddr  = (struct sockaddr_in *)LOCAL_ADDR;
+    struct sockaddr_in *remoteaddr = (struct sockaddr_in *)REMOTE_ADDR;
+    char *host = conn->hp->h_name;
+    size_t nread;
+    int l = sizeof(local_addr);
+
+    if(getsockname(conn->data->firstsocket, LOCAL_ADDR, &l) < 0)
+	perror("getsockname()");
+
+    checksum = getpid();
+    ret = mk_auth(d, &adat, "ftp", host, checksum);
+    if(ret == KDC_PR_UNKNOWN)
+	ret = mk_auth(d, &adat, "rcmd", host, checksum);
+    if(ret){
+	printf("%s\n", krb_get_err_text(ret));
+	return AUTH_CONTINUE;
+    }
+
+#ifdef HAVE_KRB_GET_OUR_IP_FOR_REALM
+    if (krb_get_config_bool("nat_in_use")) {
+      struct in_addr natAddr;
+
+      if (krb_get_our_ip_for_realm(krb_realmofhost(host),
+				   &natAddr) != KSUCCESS
+	  && krb_get_our_ip_for_realm(NULL, &natAddr) != KSUCCESS)
+	printf("Can't get address for realm %s\n",
+	       krb_realmofhost(host));
+      else {
+	if (natAddr.s_addr != localaddr->sin_addr.s_addr) {
+	  printf("Using NAT IP address (%s) for kerberos 4\n",
+		 inet_ntoa(natAddr));
+	  localaddr->sin_addr = natAddr;
+	  
+	  /*
+	   * This not the best place to do this, but it
+	   * is here we know that (probably) NAT is in
+	   * use!
+	   */
+
+	  /*passivemode = 1;***/
+	  /*printf("Setting: Passive mode on.\n");***/
+	}
+      }
+    }
+#endif
+
+    /*printf("Local address is %s\n", inet_ntoa(localaddr->sin_addr));***/
+    /*printf("Remote address is %s\n", inet_ntoa(remoteaddr->sin_addr));***/
+
+    if(base64_encode(adat.dat, adat.length, &p) < 0) {
+	printf("Out of memory base64-encoding.\n");
+	return AUTH_CONTINUE;
+    }
+    /*ret = command("ADAT %s", p)*/
+    ftpsendf(conn->data->firstsocket, conn, "ADAT %s", p);
+    /* wait for feedback */
+    nread = GetLastResponse(conn->data->firstsocket,
+			    conn->data->buffer, conn);
+    if(nread < 0)
+	return /*CURLE_OPERATION_TIMEOUTED*/-1;
+    free(p);
+
+    if(/*ret != COMPLETE*/conn->data->buffer[0] != '2'){
+	printf("Server didn't accept auth data.\n");
+	return AUTH_ERROR;
+    }
+
+    p = strstr(/*reply_string*/conn->data->buffer, "ADAT=");
+    if(!p){
+	printf("Remote host didn't send adat reply.\n");
+	return AUTH_ERROR;
+    }
+    p += 5;
+    len = base64_decode(p, adat.dat);
+    if(len < 0){
+	printf("Failed to decode base64 from server.\n");
+	return AUTH_ERROR;
+    }
+    adat.length = len;
+    ret = krb_rd_safe(adat.dat, adat.length, &d->key, 
+		      (struct sockaddr_in *)hisctladdr, 
+		      (struct sockaddr_in *)myctladdr, &msg_data);
+    if(ret){
+	printf("Error reading reply from server: %s.\n", 
+	       krb_get_err_text(ret));
+	return AUTH_ERROR;
+    }
+    krb_get_int(msg_data.app_data, &cs, 4, 0);
+    if(cs - checksum != 1){
+	printf("Bad checksum returned from server.\n");
+	return AUTH_ERROR;
+    }
+    return AUTH_OK;
+}
+
+struct sec_client_mech krb4_client_mech = {
+    "KERBEROS_V4",
+    sizeof(struct krb4_data),
+    NULL, /* init */
+    krb4_auth,
+    NULL, /* end */
+    krb4_check_prot,
+    krb4_overhead,
+    krb4_encode,
+    krb4_decode
+};
+
+#endif /* FTP_SERVER */
+
+void krb_kauth(struct connectdata *conn)
+{
+    int ret;
+    char buf[1024];
+    des_cblock key;
+    des_key_schedule schedule;
+    KTEXT_ST tkt, tktcopy;
+    char *name;
+    char *p;
+    int overbose;
+    char passwd[100];
+    int tmp;
+    size_t nread;
+	
+    int save;
+
+    save = set_command_prot(conn, prot_private);
+    /*ret = command("SITE KAUTH %s", name);***/
+    ftpsendf(conn->data->firstsocket, conn,
+             "SITE KAUTH %s", conn->data->user);
+    /* wait for feedback */
+    nread = GetLastResponse(conn->data->firstsocket, conn->data->buffer, conn);
+    if(nread < 0)
+	return /*CURLE_OPERATION_TIMEOUTED*/;
+
+    if(/*ret != CONTINUE*/conn->data->buffer[0] != '3'){
+	/*verbose = overbose;***/
+	set_command_prot(conn, save);
+	/*code = -1;***/
+	return;
+    }
+    /*verbose = overbose;***/
+    p = strstr(/*reply_string***/conn->data->buffer, "T=");
+    if(!p){
+	printf("Bad reply from server.\n");
+	set_command_prot(conn, save);
+	/*code = -1;***/
+	return;
+    }
+    p += 2;
+    tmp = base64_decode(p, &tkt.dat);
+    if(tmp < 0){
+	printf("Failed to decode base64 in reply.\n");
+	set_command_prot(conn, save);
+	/*code = -1;***/
+	return;
+    }
+    tkt.length = tmp;
+    tktcopy.length = tkt.length;
+    
+    p = strstr(/*reply_string***/conn->data->buffer, "P=");
+    if(!p){
+	printf("Bad reply from server.\n");
+	/*verbose = overbose;***/
+	set_command_prot(conn, save);
+	/*code = -1;***/
+	return;
+    }
+    name = p + 2;
+    for(; *p && *p != ' ' && *p != '\r' && *p != '\n'; p++);
+    *p = 0;
+
+#if 0
+    snprintf(buf, sizeof(buf), "Password for %s:", name);
+    if (des_read_pw_string (passwd, sizeof(passwd)-1, buf, 0))
+        *passwd = '\0';
+    des_string_to_key (passwd, &key);
+#else
+    des_string_to_key (conn->data->passwd, &key);
+#endif
+
+    des_key_sched(&key, schedule);
+    
+    des_pcbc_encrypt((des_cblock*)tkt.dat, (des_cblock*)tktcopy.dat,
+		     tkt.length,
+		     schedule, &key, DES_DECRYPT);
+    if (strcmp ((char*)tktcopy.dat + 8,
+		KRB_TICKET_GRANTING_TICKET) != 0) {
+        afs_string_to_key (passwd,
+			   krb_realmofhost(/*hostname***/conn->hp->h_name),
+			   &key);
+	des_key_sched (&key, schedule);
+	des_pcbc_encrypt((des_cblock*)tkt.dat, (des_cblock*)tktcopy.dat,
+			 tkt.length,
+			 schedule, &key, DES_DECRYPT);
+    }
+    memset(key, 0, sizeof(key));
+    memset(schedule, 0, sizeof(schedule));
+    memset(passwd, 0, sizeof(passwd));
+    if(base64_encode(tktcopy.dat, tktcopy.length, &p) < 0) {
+      failf(conn->data, "Out of memory base64-encoding.\n");
+      set_command_prot(conn, save);
+      /*code = -1;***/
+      return;
+    }
+    memset (tktcopy.dat, 0, tktcopy.length);
+    /*ret = command("SITE KAUTH %s %s", name, p);***/
+    ftpsendf(conn->data->firstsocket, conn,
+             "SITE KAUTH %s %s", name, p);
+    /* wait for feedback */
+    nread = GetLastResponse(conn->data->firstsocket, conn->data->buffer, conn);
+    if(nread < 0)
+	return /*CURLE_OPERATION_TIMEOUTED*/;
+    free(p);
+    set_command_prot(conn, save);
+}
+
+#endif /* KRB4 */

lib/ldap.c

@@ -134,8 +134,7 @@ static void * DynaGetFunction(char *name)
 static int WriteProc(void *param, char *text, int len)
 {
   struct UrlData *data = (struct UrlData *)param;
-
-  data->fwrite(text, 1, strlen(text), data->out);
+  client_write(data, CLIENTWRITE_BODY, text, 0);
   return 0;
 }
 

lib/libcurl.def

@@ -0,0 +1,42 @@
+;
+; Definition file for the DLL version of the LIBCURL library from curl
+;
+
+LIBRARY		LIBCURL
+
+DESCRIPTION	'curl libcurl - http://curl.haxx.se'
+
+EXPORTS
+	curl_close @ 1 ; 
+	curl_connect @ 2 ; 
+	curl_disconnect @ 3 ; 
+	curl_do @ 4 ; 
+	curl_done @ 5 ; 
+	curl_easy_cleanup @ 6 ; 
+	curl_easy_getinfo @ 7 ; 
+	curl_easy_init @ 8 ; 
+	curl_easy_perform @ 9 ; 
+	curl_easy_setopt @ 10 ; 
+	curl_escape @ 11 ; 
+	curl_formparse @ 12 ; 
+	curl_free @ 13 ; 
+	curl_getdate @ 14 ; 
+	curl_getenv @ 15 ; 
+	curl_init @ 16 ; 
+	curl_open @ 17 ; 
+	curl_read @ 18 ; 
+	curl_setopt @ 19 ; 
+	curl_slist_append @ 20 ; 
+	curl_slist_free_all @ 21 ; 
+	curl_transfer @ 22 ; 
+	curl_unescape @ 23 ; 
+	curl_version @ 24 ; 
+	curl_write @ 25 ; 
+	maprintf @ 26 ; 
+ 	mfprintf @ 27 ;
+ 	mprintf @ 28 ;
+ 	msprintf @ 29 ;
+ 	msnprintf @ 30 ;
+ 	mvfprintf @ 31 ;
+	strequal @ 32 ; 
+	strnequal @ 33 ; 

lib/memdebug.c

@@ -0,0 +1,118 @@
+#ifdef MALLOCDEBUG
+/*****************************************************************************
+ *                                  _   _ ____  _     
+ *  Project                     ___| | | |  _ \| |    
+ *                             / __| | | | |_) | |    
+ *                            | (__| |_| |  _ <| |___ 
+ *                             \___|\___/|_| \_\_____|
+ *
+ *  The contents of this file are subject to the Mozilla Public License
+ *  Version 1.0 (the "License"); you may not use this file except in
+ *  compliance with the License. You may obtain a copy of the License at
+ *  http://www.mozilla.org/MPL/
+ *
+ *  Software distributed under the License is distributed on an "AS IS"
+ *  basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
+ *  License for the specific language governing rights and limitations
+ *  under the License.
+ *
+ *  The Original Code is Curl.
+ *
+ *  The Initial Developer of the Original Code is Daniel Stenberg.
+ *
+ *  Portions created by the Initial Developer are Copyright (C) 1999.
+ *  All Rights Reserved.
+ *
+ * ------------------------------------------------------------
+ * Main author:
+ * - Daniel Stenberg <daniel@haxx.se>
+ *
+ * 	http://curl.haxx.se
+ *
+ * $Source$
+ * $Revision$
+ * $Date$
+ * $Author$
+ * $State$
+ * $Locker$
+ *
+ * ------------------------------------------------------------
+ ****************************************************************************/
+
+#include "setup.h"
+
+#include <curl/curl.h>
+
+#define _MPRINTF_REPLACE
+#include <curl/mprintf.h>
+#include "urldata.h"
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+
+/*
+ * Note that these debug functions are very simple and they are meant to
+ * remain so. For advanced analysis, record a log file and write perl scripts
+ * to analyze them!
+ *
+ * Don't use these with multithreaded test programs!
+ */
+
+FILE *logfile;
+
+/* this sets the log file name */
+void curl_memdebug(char *logname)
+{
+  logfile = fopen(logname, "w");
+}
+
+
+void *curl_domalloc(size_t size, int line, char *source)
+{
+  void *mem=(malloc)(size);
+  fprintf(logfile?logfile:stderr, "MEM %s:%d malloc(%d) = %p\n",
+          source, line, size, mem);
+  return mem;
+}
+
+char *curl_dostrdup(char *str, int line, char *source)
+{
+  char *mem;
+  size_t len;
+  
+  if(NULL ==str) {
+    fprintf(stderr, "ILLEGAL strdup() on NULL at %s:%d\n",
+            source, line);
+    exit(2);
+  }
+
+  mem=(strdup)(str);
+  len=strlen(str)+1;
+  fprintf(logfile?logfile:stderr, "MEM %s:%d strdup(%p) (%d) = %p\n",
+          source, line, str, len, mem);
+  return mem;
+}
+
+void *curl_dorealloc(void *ptr, size_t size, int line, char *source)
+{
+  void *mem=(realloc)(ptr, size);
+  fprintf(logfile?logfile:stderr, "MEM %s:%d realloc(%p, %d) = %p\n",
+          source, line, ptr, size, mem);
+  return mem;
+}
+
+void curl_dofree(void *ptr, int line, char *source)
+{
+  if(NULL == ptr) {
+    fprintf(stderr, "ILLEGAL free() on NULL at %s:%d\n",
+            source, line);
+    exit(2);
+  }
+
+  (free)(ptr);
+
+  fprintf(logfile?logfile:stderr, "MEM %s:%d free(%p)\n",
+          source, line, ptr);
+}
+
+#endif /* MALLOCDEBUG */

lib/memdebug.h

@@ -0,0 +1,13 @@
+#ifdef MALLOCDEBUG
+void *curl_domalloc(size_t size, int line, char *source);
+void *curl_dorealloc(void *ptr, size_t size, int line, char *source);
+void curl_dofree(void *ptr, int line, char *source);
+char *curl_dostrdup(char *str, int line, char *source);
+void curl_memdebug(char *logname);
+
+/* Set this symbol on the command-line, recompile all lib-sources */
+#define strdup(ptr) curl_dostrdup(ptr, __LINE__, __FILE__)
+#define malloc(size) curl_domalloc(size, __LINE__, __FILE__)
+#define realloc(ptr,size) curl_dorealloc(ptr, size, __LINE__, __FILE__)
+#define free(ptr) curl_dofree(ptr, __LINE__, __FILE__)
+#endif

lib/mprintf.c

@@ -98,6 +98,10 @@ static const char rcsid[] = "@(#)$Id$";
 #include <ctype.h>
 #include <string.h>
 
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
 
 #define BUFFSIZE 256 /* buffer for long-to-str and float-to-str calcs */
 #define MAX_PARAMETERS 128 /* lame static limit */
@@ -997,33 +1001,6 @@ static int dprintf_formatf(
   return done;
 }
 
-static int StoreNonPrintable(int output, struct nsprintf *infop)
-{
-  /* If the character isn't printable then we convert it */
-  char work[64], *w;
-  int num = output;
-
-  w = &work[sizeof(work)];
-  *(--w) = (char)0;
-  for(; num > 0; num /= 10) {
-    *(--w) = lower_digits[num % 10];
-  }
-  if (infop->length + strlen(w) + 1 < infop->max)
-    {
-      infop->buffer[0] = '\\';
-      infop->buffer++;
-      infop->length++;
-      for (; *w; w++)
-	{
-	  infop->buffer[0] = *w;
-	  infop->buffer++;
-	  infop->length++;
-	}
-      return output;
-    }
-  return -1;
-}
-
 /* fputc() look-alike */
 static int addbyter(int output, FILE *data)
 {
@@ -1031,16 +1008,9 @@ static int addbyter(int output, FILE *data)
  
   if(infop->length < infop->max) {
     /* only do this if we haven't reached max length yet */
-    if (isprint(output) || isspace(output))
-      {
     infop->buffer[0] = (char)output; /* store */
     infop->buffer++; /* increase pointer */
     infop->length++; /* we are now one byte larger */
-      }
-    else
-      {
-	return StoreNonPrintable(output, infop);
-      }
     return output; /* fputc() returns like this on success */
   }
   return -1;

lib/progress.c

@@ -113,7 +113,7 @@ void pgrsDone(struct UrlData *data)
   if(!(data->progress.flags & PGRS_HIDE)) {
     data->progress.lastshow=0;
     pgrsUpdate(data); /* the final (forced) update */
-    fprintf(stderr, "\n");
+    fprintf(data->err, "\n");
   }
 }
 
@@ -124,14 +124,24 @@ void pgrsTime(struct UrlData *data, timerid timer)
   case TIMER_NONE:
     /* mistake filter */
     break;
+  case TIMER_STARTSINGLE:
+    /* This is set at the start of a single fetch, there may be several
+       fetches within an operation, why we add all other times relative
+       to this one */
+    data->progress.t_startsingle = tvnow();
+    break;
+
   case TIMER_NAMELOOKUP:
-    data->progress.t_nslookup = tvnow();
+    data->progress.t_nslookup += tvdiff(tvnow(),
+                                        data->progress.t_startsingle);
     break;
   case TIMER_CONNECT:
-    data->progress.t_connect = tvnow();
+    data->progress.t_connect += tvdiff(tvnow(),
+                                       data->progress.t_startsingle);
     break;
   case TIMER_PRETRANSFER:
-    data->progress.t_pretransfer = tvnow();
+    data->progress.t_pretransfer += tvdiff(tvnow(),
+                                           data->progress.t_startsingle);
     break;
   case TIMER_POSTRANSFER:
     /* this is the normal end-of-transfer thing */
@@ -183,7 +193,7 @@ int pgrsUpdate(struct UrlData *data)
   struct timeval now;
   int result;
 
-  char max5[6][6];
+  char max5[6][10];
   double dlpercen=0;
   double ulpercen=0;
   double total_percen=0;
@@ -191,12 +201,7 @@ int pgrsUpdate(struct UrlData *data)
   double total_transfer;
   double total_expected_transfer;
 
-#define CURR_TIME 5
-
-  static double speeder[ CURR_TIME ];
-  static int speeder_c=0;
-
-  int nowindex = speeder_c% CURR_TIME;
+  int nowindex = data->progress.speeder_c% CURR_TIME;
   int checkindex;
   int count;
 
@@ -241,15 +246,19 @@ int pgrsUpdate(struct UrlData *data)
   /* Let's do the "current speed" thing, which should use the fastest
          of the dl/ul speeds */
 
-  speeder[ nowindex ] = data->progress.downloaded>data->progress.uploaded?
+  data->progress.speeder[ nowindex ] =
+    data->progress.downloaded>data->progress.uploaded?
     data->progress.downloaded:data->progress.uploaded;
-  speeder_c++; /* increase */
-  count = ((speeder_c>=CURR_TIME)?CURR_TIME:speeder_c) - 1;
-  checkindex = (speeder_c>=CURR_TIME)?speeder_c%CURR_TIME:0;
+  data->progress.speeder_c++; /* increase */
+  count = ((data->progress.speeder_c>=CURR_TIME)?
+           CURR_TIME:data->progress.speeder_c) - 1;
+  checkindex = (data->progress.speeder_c>=CURR_TIME)?
+    data->progress.speeder_c%CURR_TIME:0;
 
   /* find out the average speed the last CURR_TIME seconds */
   data->progress.current_speed =
-    (speeder[nowindex]-speeder[checkindex])/(count?count:1);
+    (data->progress.speeder[nowindex]-
+     data->progress.speeder[checkindex])/(count?count:1);
 
   if(data->progress.flags & PGRS_HIDE)
     return 0;
@@ -284,6 +293,7 @@ int pgrsUpdate(struct UrlData *data)
          total estimate! */
   total_estimate = ulestimate>dlestimate?ulestimate:dlestimate;
 
+
   /* If we have a total estimate, we can display that and the expected
          time left */
   if(total_estimate) {
@@ -312,7 +322,7 @@ int pgrsUpdate(struct UrlData *data)
   if(total_expected_transfer)
     total_percen=(double)(total_transfer/total_expected_transfer)*100;
 
-  fprintf(stderr,
+  fprintf(data->err,
           "\r%3d %s  %3d %s  %3d %s  %s  %s %s %s %s %s",
           (int)total_percen,                            /* total % */
           max5data(total_expected_transfer, max5[2]),   /* total size */

lib/progress.h

@@ -49,6 +49,7 @@ typedef enum {
   TIMER_CONNECT,
   TIMER_PRETRANSFER,
   TIMER_POSTRANSFER,
+  TIMER_STARTSINGLE,
   TIMER_LAST /* must be last */
 } timerid;
   

lib/security.c

@@ -0,0 +1,659 @@
+/* modified by Martin Hedenfalk <mhe@stacken.kth.se> for use in Curl
+ * last modified 2000-09-18
+ * Even more obscurified to merge better into libcurl by Daniel Stenberg.
+ */
+
+/*
+ * Copyright (c) 1998, 1999 Kungliga Tekniska H<>gskolan
+ * (Royal Institute of Technology, Stockholm, Sweden).
+ * All rights reserved.
+ * 
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 
+ * 3. Neither the name of the Institute nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ * 
+ * THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include "setup.h"
+
+#ifdef KRB4
+
+#include <curl/mprintf.h>
+
+#include "security.h"
+#include <stdlib.h>
+#include <string.h>
+#include <netdb.h>
+#include "base64.h"
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
+#define min(a, b)   ((a) < (b) ? (a) : (b))
+
+static struct {
+    enum protection_level level;
+    const char *name;
+} level_names[] = {
+    { prot_clear, "clear" },
+    { prot_safe, "safe" },
+    { prot_confidential, "confidential" },
+    { prot_private, "private" }
+};
+
+static const char *
+level_to_name(enum protection_level level)
+{
+    int i;
+    for(i = 0; i < sizeof(level_names) / sizeof(level_names[0]); i++)
+	if(level_names[i].level == level)
+	    return level_names[i].name;
+    return "unknown";
+}
+
+#ifndef FTP_SERVER /* not used in server */
+static enum protection_level 
+name_to_level(const char *name)
+{
+    int i;
+    for(i = 0; i < sizeof(level_names) / sizeof(level_names[0]); i++)
+	if(!strncasecmp(level_names[i].name, name, strlen(name)))
+	    return level_names[i].level;
+    return (enum protection_level)-1;
+}
+#endif
+
+#ifdef FTP_SERVER
+
+static struct sec_server_mech *mechs[] = {
+#ifdef KRB5
+    &gss_server_mech,
+#endif
+#ifdef KRB4
+    &krb4_server_mech,
+#endif
+    NULL
+};
+
+static struct sec_server_mech *mech;
+
+#else
+
+static struct sec_client_mech *mechs[] = {
+#ifdef KRB5
+    &gss_client_mech,
+#endif
+#ifdef KRB4
+    &krb4_client_mech,
+#endif
+    NULL
+};
+
+static struct sec_client_mech *mech;
+
+#endif
+
+int
+sec_getc(struct connectdata *conn, FILE *F)
+{
+  if(conn->sec_complete && conn->data_prot) {
+    char c;
+    if(sec_read(conn, fileno(F), &c, 1) <= 0)
+      return EOF;
+    return c;
+  } else
+    return getc(F);
+}
+
+static int
+block_read(int fd, void *buf, size_t len)
+{
+    unsigned char *p = buf;
+    int b;
+    while(len) {
+	b = read(fd, p, len);
+	if (b == 0)
+	    return 0;
+	else if (b < 0)
+	    return -1;
+	len -= b;
+	p += b;
+    }
+    return p - (unsigned char*)buf;
+}
+
+static int
+block_write(int fd, void *buf, size_t len)
+{
+    unsigned char *p = buf;
+    int b;
+    while(len) {
+	b = write(fd, p, len);
+	if(b < 0)
+	    return -1;
+	len -= b;
+	p += b;
+    }
+    return p - (unsigned char*)buf;
+}
+
+static int
+sec_get_data(struct connectdata *conn,
+             int fd, struct krb4buffer *buf, int level)
+{
+  int len;
+  int b;
+
+  b = block_read(fd, &len, sizeof(len));
+  if (b == 0)
+    return 0;
+  else if (b < 0)
+    return -1;
+  len = ntohl(len);
+  buf->data = realloc(buf->data, len);
+  b = block_read(fd, buf->data, len);
+  if (b == 0)
+    return 0;
+  else if (b < 0)
+    return -1;
+  buf->size = (*mech->decode)(conn->app_data, buf->data, len,
+                              conn->data_prot, conn);
+  buf->index = 0;
+  return 0;
+}
+
+static size_t
+buffer_read(struct krb4buffer *buf, void *data, size_t len)
+{
+    len = min(len, buf->size - buf->index);
+    memcpy(data, (char*)buf->data + buf->index, len);
+    buf->index += len;
+    return len;
+}
+
+static size_t
+buffer_write(struct krb4buffer *buf, void *data, size_t len)
+{
+    if(buf->index + len > buf->size) {
+	void *tmp;
+	if(buf->data == NULL)
+	    tmp = malloc(1024);
+	else
+	    tmp = realloc(buf->data, buf->index + len);
+	if(tmp == NULL)
+	    return -1;
+	buf->data = tmp;
+	buf->size = buf->index + len;
+    }
+    memcpy((char*)buf->data + buf->index, data, len);
+    buf->index += len;
+    return len;
+}
+
+int
+sec_read(struct connectdata *conn, int fd, void *buffer, int length)
+{
+    size_t len;
+    int rx = 0;
+
+    if(conn->sec_complete == 0 || conn->data_prot == 0)
+      return read(fd, buffer, length);
+
+    if(conn->in_buffer.eof_flag){
+      conn->in_buffer.eof_flag = 0;
+      return 0;
+    }
+    
+    len = buffer_read(&conn->in_buffer, buffer, length);
+    length -= len;
+    rx += len;
+    buffer = (char*)buffer + len;
+    
+    while(length) {
+      if(sec_get_data(conn, fd, &conn->in_buffer, conn->data_prot) < 0)
+        return -1;
+      if(conn->in_buffer.size == 0) {
+        if(rx)
+          conn->in_buffer.eof_flag = 1;
+        return rx;
+      }
+      len = buffer_read(&conn->in_buffer, buffer, length);
+      length -= len;
+      rx += len;
+      buffer = (char*)buffer + len;
+    }
+    return rx;
+}
+
+static int
+sec_send(struct connectdata *conn, int fd, char *from, int length)
+{
+  int bytes;
+  void *buf;
+  bytes = (*mech->encode)(conn->app_data, from, length, conn->data_prot, &buf, conn);
+  bytes = htonl(bytes);
+  block_write(fd, &bytes, sizeof(bytes));
+  block_write(fd, buf, ntohl(bytes));
+  free(buf);
+  return length;
+}
+
+int
+sec_fflush(struct connectdata *conn, FILE *F)
+{
+  if(conn->data_prot != prot_clear) {
+    if(conn->out_buffer.index > 0){
+      sec_write(conn, fileno(F),
+                conn->out_buffer.data, conn->out_buffer.index);
+      conn->out_buffer.index = 0;
+    }
+    sec_send(conn, fileno(F), NULL, 0);
+  }
+  fflush(F);
+  return 0;
+}
+
+int
+sec_fflush_fd(struct connectdata *conn, int fd)
+{
+  if(conn->data_prot != prot_clear) {
+    if(conn->out_buffer.index > 0){
+      sec_write(conn, fd,
+                conn->out_buffer.data, conn->out_buffer.index);
+      conn->out_buffer.index = 0;
+    }
+    sec_send(conn, fd, NULL, 0);
+  }
+  return 0;
+}
+
+int
+sec_write(struct connectdata *conn, int fd, char *buffer, int length)
+{
+  int len = conn->buffer_size;
+  int tx = 0;
+      
+  if(conn->data_prot == prot_clear)
+    return write(fd, buffer, length);
+
+  len -= (*mech->overhead)(conn->app_data, conn->data_prot, len);
+  while(length){
+    if(length < len)
+      len = length;
+    sec_send(conn, fd, buffer, len);
+    length -= len;
+    buffer += len;
+    tx += len;
+  }
+  return tx;
+}
+
+int
+sec_vfprintf2(struct connectdata *conn, FILE *f, const char *fmt, va_list ap)
+{
+  char *buf;
+  int ret;
+  if(conn->data_prot == prot_clear)
+    return vfprintf(f, fmt, ap);
+  else {
+    buf = maprintf(fmt, ap);
+    ret = buffer_write(&conn->out_buffer, buf, strlen(buf));
+    free(buf);
+    return ret;
+  }
+}
+
+int
+sec_fprintf2(struct connectdata *conn, FILE *f, const char *fmt, ...)
+{
+    int ret;
+    va_list ap;
+    va_start(ap, fmt);
+    ret = sec_vfprintf2(conn, f, fmt, ap);
+    va_end(ap);
+    return ret;
+}
+
+int
+sec_putc(struct connectdata *conn, int c, FILE *F)
+{
+  char ch = c;
+  if(conn->data_prot == prot_clear)
+    return putc(c, F);
+    
+  buffer_write(&conn->out_buffer, &ch, 1);
+  if(c == '\n' || conn->out_buffer.index >= 1024 /* XXX */) {
+    sec_write(conn, fileno(F), conn->out_buffer.data, conn->out_buffer.index);
+    conn->out_buffer.index = 0;
+  }
+  return c;
+}
+
+int
+sec_read_msg(struct connectdata *conn, char *s, int level)
+{
+    int len;
+    char *buf;
+    int code;
+    
+    buf = malloc(strlen(s));
+    len = base64_decode(s + 4, buf); /* XXX */
+    
+    len = (*mech->decode)(conn->app_data, buf, len, level, conn);
+    if(len < 0)
+	return -1;
+    
+    buf[len] = '\0';
+
+    if(buf[3] == '-')
+	code = 0;
+    else
+	sscanf(buf, "%d", &code);
+    if(buf[len-1] == '\n')
+	buf[len-1] = '\0';
+    strcpy(s, buf);
+    free(buf);
+    return code;
+}
+
+/* modified to return how many bytes written, or -1 on error ***/
+int
+sec_vfprintf(struct connectdata *conn, FILE *f, const char *fmt, va_list ap)
+{
+    int ret = 0;
+    char *buf;
+    void *enc;
+    int len;
+    if(!conn->sec_complete)
+	return vfprintf(f, fmt, ap);
+    
+    buf = maprintf(fmt, ap);
+    len = (*mech->encode)(conn->app_data, buf, strlen(buf),
+                          conn->command_prot, &enc,
+			  conn);
+    free(buf);
+    if(len < 0) {
+	failf(conn->data, "Failed to encode command.\n");
+	return -1;
+    }
+    if(base64_encode(enc, len, &buf) < 0){
+      failf(conn->data, "Out of memory base64-encoding.\n");
+      return -1;
+    }
+#ifdef FTP_SERVER
+    if(command_prot == prot_safe)
+	fprintf(f, "631 %s\r\n", buf);
+    else if(command_prot == prot_private)
+	fprintf(f, "632 %s\r\n", buf);
+    else if(command_prot == prot_confidential)
+	fprintf(f, "633 %s\r\n", buf);
+#else
+    if(conn->command_prot == prot_safe)
+	ret = fprintf(f, "MIC %s", buf);
+    else if(conn->command_prot == prot_private)
+	ret = fprintf(f, "ENC %s", buf);
+    else if(conn->command_prot == prot_confidential)
+	ret = fprintf(f, "CONF %s", buf);
+#endif
+    free(buf);
+    return ret;
+}
+
+int
+sec_fprintf(struct connectdata *conn, FILE *f, const char *fmt, ...)
+{
+    va_list ap;
+    int ret;
+    va_start(ap, fmt);
+    ret = sec_vfprintf(conn, f, fmt, ap);
+    va_end(ap);
+    return ret;
+}
+
+/* end common stuff */
+
+#ifdef FTP_SERVER
+
+/* snip */
+
+#else /* FTP_SERVER */
+
+#if 0
+void
+sec_status(void)
+{
+    if(conn->sec_complete){
+	printf("Using %s for authentication.\n", mech->name);
+	printf("Using %s command channel.\n", level_to_name(command_prot));
+	printf("Using %s data channel.\n", level_to_name(data_prot));
+	if(buffer_size > 0)
+	    printf("Protection buffer size: %lu.\n", 
+		   (unsigned long)buffer_size);
+    }else{
+	printf("Not using any security mechanism.\n");
+    }
+}
+#endif
+
+static int
+sec_prot_internal(struct connectdata *conn, int level)
+{
+    int ret;
+    char *p;
+    unsigned int s = 1048576;
+    size_t nread;
+
+    if(!conn->sec_complete){
+      infof(conn->data, "No security data exchange has taken place.\n");
+      return -1;
+    }
+
+    if(level){
+      ftpsendf(conn->data->firstsocket, conn,
+               "PBSZ %u", s);
+      /* wait for feedback */
+      nread = GetLastResponse(conn->data->firstsocket,
+                              conn->data->buffer, conn);
+      if(nread < 0)
+        return /*CURLE_OPERATION_TIMEOUTED*/-1;
+      if(/*ret != COMPLETE*/conn->data->buffer[0] != '2'){
+        failf(conn->data, "Failed to set protection buffer size.\n");
+        return -1;
+      }
+      conn->buffer_size = s;
+      p = strstr(/*reply_string*/conn->data->buffer, "PBSZ=");
+      if(p)
+        sscanf(p, "PBSZ=%u", &s);
+      if(s < conn->buffer_size)
+        conn->buffer_size = s;
+    }
+
+    ftpsendf(conn->data->firstsocket, conn,
+             "PROT %c", level["CSEP"]);
+    /* wait for feedback */
+    nread = GetLastResponse(conn->data->firstsocket,
+			    conn->data->buffer, conn);
+    if(nread < 0)
+      return /*CURLE_OPERATION_TIMEOUTED*/-1;
+    if(/*ret != COMPLETE*/conn->data->buffer[0] != '2'){
+      failf(conn->data, "Failed to set protection level.\n");
+      return -1;
+    }
+    
+    conn->data_prot = (enum protection_level)level;
+    return 0;
+}
+
+enum protection_level
+set_command_prot(struct connectdata *conn, enum protection_level level)
+{
+    enum protection_level old = conn->command_prot;
+    conn->command_prot = level;
+    return old;
+}
+
+#if 0
+void
+sec_prot(int argc, char **argv)
+{
+    int level = -1;
+
+    if(argc < 2 || argc > 3)
+	goto usage;
+    if(!sec_complete) {
+	printf("No security data exchange has taken place.\n");
+	code = -1;
+	return;
+    }
+    level = name_to_level(argv[argc - 1]);
+    
+    if(level == -1)
+	goto usage;
+    
+    if((*mech->check_prot)(conn->app_data, level)) {
+	printf("%s does not implement %s protection.\n", 
+	       mech->name, level_to_name(level));
+	code = -1;
+	return;
+    }
+    
+    if(argc == 2 || strncasecmp(argv[1], "data", strlen(argv[1])) == 0) {
+	if(sec_prot_internal(level) < 0){
+	    code = -1;
+	    return;
+	}
+    } else if(strncasecmp(argv[1], "command", strlen(argv[1])) == 0)
+	set_command_prot(level);
+    else
+	goto usage;
+    code = 0;
+    return;
+ usage:
+    printf("usage: %s [command|data] [clear|safe|confidential|private]\n",
+	   argv[0]);
+    code = -1;
+}
+#endif
+
+void
+sec_set_protection_level(struct connectdata *conn)
+{
+  if(conn->sec_complete && conn->data_prot != conn->request_data_prot)
+    sec_prot_internal(conn, conn->request_data_prot);
+}
+
+
+int
+sec_request_prot(struct connectdata *conn, char *level)
+{
+  int l = name_to_level(level);
+  if(l == -1)
+    return -1;
+  conn->request_data_prot = (enum protection_level)l;
+  return 0;
+}
+
+int
+sec_login(struct connectdata *conn)
+{
+    int ret;
+    struct sec_client_mech **m;
+    size_t nread;
+    struct UrlData *data=conn->data;
+
+    for(m = mechs; *m && (*m)->name; m++) {
+	void *tmp;
+
+	tmp = realloc(conn->app_data, (*m)->size);
+	if (tmp == NULL) {
+          failf (data, "realloc %u failed", (*m)->size);
+          return -1;
+	}
+	conn->app_data = tmp;
+	    
+	if((*m)->init && (*(*m)->init)(conn->app_data) != 0) {
+	    infof(data, "Skipping %s...\n", (*m)->name);
+	    continue;
+	}
+	infof(data, "Trying %s...\n", (*m)->name);
+	/*ret = command("AUTH %s", (*m)->name);***/
+	ftpsendf(conn->data->firstsocket, conn,
+                 "AUTH %s", (*m)->name);
+	/* wait for feedback */
+	nread = GetLastResponse(conn->data->firstsocket,
+				conn->data->buffer, conn);
+	if(nread < 0)
+	    return /*CURLE_OPERATION_TIMEOUTED*/-1;
+	if(/*ret != CONTINUE*/conn->data->buffer[0] != '3'){
+	    if(/*code == 504*/strncmp(conn->data->buffer,"504",3) == 0) {
+		infof(data,
+                      "%s is not supported by the server.\n", (*m)->name);
+	    }
+            else if(/*code == 534*/strncmp(conn->data->buffer,"534",3) == 0) {
+              infof(data, "%s rejected as security mechanism.\n", (*m)->name);
+	    }
+            else if(/*ret == ERROR*/conn->data->buffer[0] == '5') {
+              infof(data, "The server doesn't support the FTP "
+                    "security extensions.\n");
+              return -1;
+	    }
+	    continue;
+	}
+
+	ret = (*(*m)->auth)(conn->app_data, /*host***/conn);
+	
+	if(ret == AUTH_CONTINUE)
+          continue;
+	else if(ret != AUTH_OK){
+          /* mechanism is supposed to output error string */
+	    return -1;
+	}
+	mech = *m;
+	conn->sec_complete = 1;
+	conn->command_prot = prot_safe;
+	break;
+    }
+    
+    return *m == NULL;
+}
+
+void
+sec_end(struct connectdata *conn)
+{
+    if (mech != NULL) {
+	if(mech->end)
+	    (*mech->end)(conn->app_data);
+	memset(conn->app_data, 0, mech->size);
+	free(conn->app_data);
+	conn->app_data = NULL;
+    }
+    conn->sec_complete = 0;
+    conn->data_prot = (enum protection_level)0;
+}
+
+#endif /* FTP_SERVER */
+
+#endif /* KRB4 */

lib/security.h

@@ -0,0 +1,134 @@
+/* modified by Martin Hedenfalk <mhe@stacken.kth.se> for use in Curl
+ * last modified 2000-09-18
+ */
+
+/*
+ * Copyright (c) 1998, 1999 Kungliga Tekniska H<>gskolan
+ * (Royal Institute of Technology, Stockholm, Sweden). 
+ * All rights reserved. 
+ *
+ * Redistribution and use in source and binary forms, with or without 
+ * modification, are permitted provided that the following conditions 
+ * are met: 
+ *
+ * 1. Redistributions of source code must retain the above copyright 
+ *    notice, this list of conditions and the following disclaimer. 
+ *
+ * 2. Redistributions in binary form must reproduce the above copyright 
+ *    notice, this list of conditions and the following disclaimer in the 
+ *    documentation and/or other materials provided with the distribution. 
+ *
+ * 3. Neither the name of the Institute nor the names of its contributors 
+ *    may be used to endorse or promote products derived from this software 
+ *    without specific prior written permission. 
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE INSTITUTE AND CONTRIBUTORS ``AS IS'' AND 
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE INSTITUTE OR CONTRIBUTORS BE LIABLE 
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 
+ * SUCH DAMAGE. 
+ */
+
+/* $Id$ */
+
+#ifndef __security_h__
+#define __security_h__
+
+#include <stdarg.h>
+#include "urldata.h"  /* for struct connectdata * */
+
+struct sec_client_mech {
+    char *name;
+    size_t size;
+    int (*init)(void *);
+    int (*auth)(void *, struct connectdata *);
+    void (*end)(void *);
+    int (*check_prot)(void *, int);
+    int (*overhead)(void *, int, int);
+    int (*encode)(void *, void*, int, int, void**, struct connectdata *);
+    int (*decode)(void *, void*, int, int, struct connectdata *);
+};
+
+struct sec_server_mech {
+    char *name;
+    size_t size;
+    int (*init)(void *);
+    void (*end)(void *);
+    int (*check_prot)(void *, int);
+    int (*overhead)(void *, int, int);
+    int (*encode)(void *, void*, int, int, void**);
+    int (*decode)(void *, void*, int, int);
+
+    int (*auth)(void *);
+    int (*adat)(void *, void*, size_t);
+    size_t (*pbsz)(void *, size_t);
+    int (*ccc)(void*);
+    int (*userok)(void*, char*);
+};
+
+#define AUTH_OK		0
+#define AUTH_CONTINUE	1
+#define AUTH_ERROR	2
+
+#ifdef FTP_SERVER
+extern struct sec_server_mech krb4_server_mech, gss_server_mech;
+#else
+extern struct sec_client_mech krb4_client_mech, gss_client_mech;
+#endif
+
+extern int sec_complete;
+
+#ifdef FTP_SERVER
+extern char *ftp_command;
+void new_ftp_command(char*);
+void delete_ftp_command(void);
+#endif
+
+/* ---- */
+
+
+int sec_fflush (struct connectdata *conn, FILE *);
+int sec_fflush_fd(struct connectdata *conn, int fd);
+int sec_fprintf (struct connectdata *, FILE *, const char *, ...);
+int sec_getc (struct connectdata *conn, FILE *);
+int sec_putc (struct connectdata *conn, int, FILE *);
+int sec_read (struct connectdata *conn, int, void *, int);
+int sec_read_msg (struct connectdata *conn, char *, int);
+
+int sec_vfprintf(struct connectdata *, FILE *, const char *, va_list);
+int sec_fprintf2(struct connectdata *conn, FILE *f, const char *fmt, ...);
+int sec_vfprintf2(struct connectdata *conn, FILE *, const char *, va_list);
+int sec_write (struct connectdata *conn, int, char *, int);
+
+#ifdef FTP_SERVER
+void adat (char *);
+void auth (char *);
+void ccc (void);
+void mec (char *, enum protection_level);
+void pbsz (int);
+void prot (char *);
+void delete_ftp_command (void);
+void new_ftp_command (char *);
+int sec_userok (char *);
+int secure_command (void);
+enum protection_level get_command_prot(void);
+#else
+void sec_end (struct connectdata *);
+int sec_login (struct connectdata *);
+void sec_prot (int, char **);
+int sec_request_prot (struct connectdata *conn, char *);
+void sec_set_protection_level(struct connectdata *conn);
+void sec_status (void);
+
+enum protection_level set_command_prot(struct connectdata *,
+                                       enum protection_level);
+
+#endif
+
+#endif /* __security_h__ */  

lib/sendf.c

@@ -53,9 +53,20 @@
 
 #include <curl/curl.h>
 #include "urldata.h"
+#include "sendf.h"
 
+#define _MPRINTF_REPLACE /* use the internal *printf() functions */
 #include <curl/mprintf.h>
 
+#ifdef KRB4
+#include "security.h"
+#include <string.h>
+#endif
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 /* infof() is for info message along the way */
 
 void infof(struct UrlData *data, char *fmt, ...)
@@ -77,14 +88,14 @@ void failf(struct UrlData *data, char *fmt, ...)
   va_list ap;
   va_start(ap, fmt);
   if(data->errorbuffer)
-    vsprintf(data->errorbuffer, fmt, ap);
+    vsnprintf(data->errorbuffer, CURL_ERROR_SIZE, fmt, ap);
   else /* no errorbuffer receives this, write to data->err instead */
     vfprintf(data->err, fmt, ap);
   va_end(ap);
 }
 
 /* sendf() sends the formated data to the server */
-int sendf(int fd, struct UrlData *data, char *fmt, ...)
+size_t sendf(int fd, struct UrlData *data, char *fmt, ...)
 {
   size_t bytes_written;
   char *s;
@@ -96,11 +107,12 @@ int sendf(int fd, struct UrlData *data, char *fmt, ...)
     return 0; /* failure */
   if(data->bits.verbose)
     fprintf(data->err, "> %s", s);
+
 #ifndef USE_SSLEAY
   bytes_written = swrite(fd, s, strlen(s));
-#else
-  if (data->use_ssl) {
-    bytes_written = SSL_write(data->ssl, s, strlen(s));
+#else /* USE_SSLEAY */
+  if (data->ssl.use) {
+    bytes_written = SSL_write(data->ssl.handle, s, strlen(s));
   } else {
     bytes_written = swrite(fd, s, strlen(s));
   }
@@ -109,25 +121,184 @@ int sendf(int fd, struct UrlData *data, char *fmt, ...)
   return(bytes_written);
 }
 
-/* ssend() sends plain (binary) data to the server */
-size_t ssend(int fd, struct UrlData *data, void *mem, size_t len)
+/*
+ * ftpsendf() sends the formated string as a ftp command to a ftp server
+ */
+size_t ftpsendf(int fd, struct connectdata *conn, char *fmt, ...)
 {
   size_t bytes_written;
+  char *s;
+  va_list ap;
+  va_start(ap, fmt);
+  s = mvaprintf(fmt, ap);
+  va_end(ap);
+  if(!s)
+    return 0; /* failure */
+  if(conn->data->bits.verbose)
+    fprintf(conn->data->err, "> %s\n", s);
 
-  if(data->bits.verbose)
-    fprintf(data->err, "> [binary output]\n");
-#ifndef USE_SSLEAY
-   bytes_written = swrite(fd, mem, len);
-#else
-  if (data->use_ssl) {
-    bytes_written = SSL_write(data->ssl, mem, len);
-  } else {
-    bytes_written = swrite(fd, mem, len);
+#ifdef KRB4
+  if(conn->sec_complete && conn->data->cmdchannel) {
+    bytes_written = sec_fprintf(conn, conn->data->cmdchannel, s);
+    bytes_written += fprintf(conn->data->cmdchannel, "\r\n");
+    fflush(conn->data->cmdchannel);
   }
-#endif /* USE_SSLEAY */
+  else
+#endif /* KRB4 */
+    {
+      bytes_written = swrite(fd, s, strlen(s));
+      bytes_written += swrite(fd, "\r\n", 2);
+    }
+  free(s); /* free the output string */
+  return(bytes_written);
+}
+
+
+/* ssend() sends plain (binary) data to the server */
+size_t ssend(int fd, struct connectdata *conn, void *mem, size_t len)
+{
+  size_t bytes_written;
+  struct UrlData *data=conn->data; /* conn knows data, not vice versa */
+
+#ifdef USE_SSLEAY
+  if (data->ssl.use) {
+    bytes_written = SSL_write(data->ssl.handle, mem, len);
+  }
+  else {
+#endif
+#ifdef KRB4
+    if(conn->sec_complete) {
+      bytes_written = sec_write(conn, fd, mem, len);
+    }
+    else
+#endif /* KRB4 */
+      bytes_written = swrite(fd, mem, len);
+#ifdef USE_SSLEAY
+  }
+#endif
+
   return bytes_written;
 }
 
+/* client_write() sends data to the write callback(s)
+
+   The bit pattern defines to what "streams" to write to. Body and/or header.
+   The defines are in sendf.h of course.
+ */
+CURLcode client_write(struct UrlData *data,
+                      int type,
+                      char *ptr,
+                      size_t len)
+{
+  size_t wrote;
+
+  if(0 == len)
+    len = strlen(ptr);
+
+  if(type & CLIENTWRITE_BODY) {
+    wrote = data->fwrite(ptr, 1, len, data->out);
+    if(wrote != len) {
+      failf (data, "Failed writing body");
+      return CURLE_WRITE_ERROR;
+    }
+  }
+  if((type & CLIENTWRITE_HEADER) && data->writeheader) {
+    wrote = data->fwrite(ptr, 1, len, data->writeheader);
+    if(wrote != len) {
+      failf (data, "Failed writing header");
+      return CURLE_WRITE_ERROR;
+    }
+  }
+  
+  return CURLE_OK;
+}
 
 
+/*
+ * add_buffer_init() returns a fine buffer struct
+ */
+send_buffer *add_buffer_init(void)
+{
+  send_buffer *blonk;
+  blonk=(send_buffer *)malloc(sizeof(send_buffer));
+  if(blonk) {
+    memset(blonk, 0, sizeof(send_buffer));
+    return blonk;
+  }
+  return NULL; /* failed, go home */
+}
+
+/*
+ * add_buffer_send() sends a buffer and frees all associated memory.
+ */
+size_t add_buffer_send(int sockfd, struct connectdata *conn, send_buffer *in)
+{
+  size_t amount;
+  if(conn->data->bits.verbose) {
+    fputs("> ", conn->data->err);
+    /* this data _may_ contain binary stuff */
+    fwrite(in->buffer, in->size_used, 1, conn->data->err);
+  }
+
+  amount = ssend(sockfd, conn, in->buffer, in->size_used);
+
+  if(in->buffer)
+    free(in->buffer);
+  free(in);
+
+  return amount;
+}
+
+
+/* 
+ * add_bufferf() builds a buffer from the formatted input
+ */
+CURLcode add_bufferf(send_buffer *in, char *fmt, ...)
+{
+  CURLcode result = CURLE_OUT_OF_MEMORY;
+  char *s;
+  va_list ap;
+  va_start(ap, fmt);
+  s = mvaprintf(fmt, ap); /* this allocs a new string to append */
+  va_end(ap);
+
+  if(s) {
+    result = add_buffer(in, s, strlen(s));
+    free(s);
+  }
+  return result;
+}
+
+/*
+ * add_buffer() appends a memory chunk to the existing one
+ */
+CURLcode add_buffer(send_buffer *in, void *inptr, size_t size)
+{
+  char *new_rb;
+  int new_size;
+
+  if(size > 0) {
+    if(!in->buffer ||
+       ((in->size_used + size) > (in->size_max - 1))) {
+      new_size = (in->size_used+size)*2;
+      if(in->buffer)
+        /* we have a buffer, enlarge the existing one */
+        new_rb = (char *)realloc(in->buffer, new_size);
+      else
+        /* create a new buffer */
+        new_rb = (char *)malloc(new_size);
+
+      if(!new_rb)
+        return CURLE_OUT_OF_MEMORY;
+
+      in->buffer = new_rb;
+      in->size_max = new_size;
+    }
+    memcpy(&in->buffer[in->size_used], inptr, size);
+      
+    in->size_used += size;
+  }
+
+  return CURLE_OK;
+}
 

lib/sendf.h

@@ -40,9 +40,29 @@
  * ------------------------------------------------------------
  ****************************************************************************/
 
+size_t ftpsendf(int fd, struct connectdata *, char *fmt, ...);
 size_t sendf(int fd, struct UrlData *, char *fmt, ...);
-size_t ssend(int fd, struct UrlData *, void *fmt, size_t len);
+size_t ssend(int fd, struct connectdata *, void *fmt, size_t len);
 void infof(struct UrlData *, char *fmt, ...);
 void failf(struct UrlData *, char *fmt, ...);
 
+struct send_buffer {
+  char *buffer;
+  long size_max;
+  long size_used;
+};
+typedef struct send_buffer send_buffer;
+
+#define CLIENTWRITE_BODY   1
+#define CLIENTWRITE_HEADER 2
+#define CLIENTWRITE_BOTH   (CLIENTWRITE_BODY|CLIENTWRITE_HEADER)
+
+CURLcode client_write(struct UrlData *data, int type, char *ptr,
+                      size_t len);
+send_buffer *add_buffer_init(void);
+CURLcode add_buffer(send_buffer *in, void *inptr, size_t size);
+CURLcode add_bufferf(send_buffer *in, char *fmt, ...);
+size_t add_buffer_send(int sockfd, struct connectdata *conn, send_buffer *in);
+
+
 #endif

lib/setup.h

@@ -57,6 +57,10 @@
 #endif
 #endif
 
+#ifndef __cplusplus        /* (rabe) */
+typedef char bool;
+#endif                     /* (rabe) */
+
 #include <stdio.h>
 #ifndef OS
 #ifdef WIN32
@@ -158,13 +162,4 @@ int fileno( FILE *stream);
 
 #endif
 
-/*
- * FIXME: code for getting a passwd in windows/non termcap/signal systems?
- */
-#ifndef WIN32
-#define get_password(x) getpass(x)
-#else
-#define get_password(x) 
-#endif
-
 #endif /* __CONFIG_H */

lib/speedcheck.c

@@ -50,21 +50,24 @@
 #include "sendf.h"
 #include "speedcheck.h"
 
+void speedinit(struct UrlData *data)
+{
+  memset(&data->keeps_speed, 0, sizeof(struct timeval));
+}
+
 CURLcode speedcheck(struct UrlData *data,
                     struct timeval now)
 {
-  static struct timeval keeps_speed;
-
-  if((data->current_speed >= 0) &&
+  if((data->progress.current_speed >= 0) &&
      data->low_speed_time &&
-     (tvlong(keeps_speed) != 0) &&
-     (data->current_speed < data->low_speed_limit)) {
+     (tvlong(data->keeps_speed) != 0) &&
+     (data->progress.current_speed < data->low_speed_limit)) {
 
     /* We are now below the "low speed limit". If we are below it
        for "low speed time" seconds we consider that enough reason
        to abort the download. */
     
-    if( tvdiff(now, keeps_speed) > data->low_speed_time) {
+    if( tvdiff(now, data->keeps_speed) > data->low_speed_time) {
       /* we have been this slow for long enough, now die */
       failf(data,
 	    "Operation too slow. "
@@ -76,7 +79,7 @@ CURLcode speedcheck(struct UrlData *data,
   }
   else {
     /* we keep up the required speed all right */
-    keeps_speed = now;
+    data->keeps_speed = now;
   }
   return CURLE_OK;
 }

lib/speedcheck.h

@@ -44,6 +44,7 @@
 
 #include "timeval.h"
 
+void speedinit(struct UrlData *data);
 CURLcode speedcheck(struct UrlData *data,
                     struct timeval now);
 

lib/ssluse.c

@@ -94,10 +94,10 @@ int SSL_cert_stuff(struct UrlData *data,
        */
       strcpy(global_passwd, data->cert_passwd);
       /* Set passwd callback: */
-      SSL_CTX_set_default_passwd_cb(data->ctx, passwd_callback);
+      SSL_CTX_set_default_passwd_cb(data->ssl.ctx, passwd_callback);
     }
 
-    if (SSL_CTX_use_certificate_file(data->ctx,
+    if (SSL_CTX_use_certificate_file(data->ssl.ctx,
 				     cert_file,
 				     SSL_FILETYPE_PEM) <= 0) {
       failf(data, "unable to set certificate file (wrong password?)\n");
@@ -106,14 +106,14 @@ int SSL_cert_stuff(struct UrlData *data,
     if (key_file == NULL)
       key_file=cert_file;
 
-    if (SSL_CTX_use_PrivateKey_file(data->ctx,
+    if (SSL_CTX_use_PrivateKey_file(data->ssl.ctx,
 				    key_file,
 				    SSL_FILETYPE_PEM) <= 0) {
       failf(data, "unable to set public key file\n");
       return(0);
     }
     
-    ssl=SSL_new(data->ctx);
+    ssl=SSL_new(data->ssl.ctx);
     x509=SSL_get_certificate(ssl);
     
     if (x509 != NULL)
@@ -127,7 +127,7 @@ int SSL_cert_stuff(struct UrlData *data,
     
     /* Now we know that a key and cert have been set against
      * the SSL context */
-    if (!SSL_CTX_check_private_key(data->ctx)) {
+    if (!SSL_CTX_check_private_key(data->ssl.ctx)) {
       failf(data, "Private key does not match the certificate public key\n");
       return(0);
     }
@@ -140,7 +140,7 @@ int SSL_cert_stuff(struct UrlData *data,
 
 #endif
 
-#if SSL_VERIFY_CERT
+#ifdef USE_SSLEAY
 int cert_verify_callback(int ok, X509_STORE_CTX *ctx)
 {
   X509 *err_cert;
@@ -164,7 +164,7 @@ UrgSSLConnect (struct UrlData *data)
     SSL_METHOD *req_method;
 
     /* mark this is being ssl enabled from here on out. */
-    data->use_ssl = 1;
+    data->ssl.use = TRUE;
 
     /* Lets get nice error messages */
     SSL_load_error_strings();
@@ -195,7 +195,7 @@ UrgSSLConnect (struct UrlData *data)
     /* Setup all the global SSL stuff */
     SSLeay_add_ssl_algorithms();
 
-    switch(data->ssl_version) {
+    switch(data->ssl.version) {
     default:
       req_method = SSLv23_client_method();
       break;
@@ -207,9 +207,9 @@ UrgSSLConnect (struct UrlData *data)
       break;
     }
     
-    data->ctx = SSL_CTX_new(req_method);
+    data->ssl.ctx = SSL_CTX_new(req_method);
 
-    if(!data->ctx) {
+    if(!data->ssl.ctx) {
       failf(data, "SSL: couldn't create a context!");
       return 1;
     }
@@ -221,22 +221,31 @@ UrgSSLConnect (struct UrlData *data)
       }
     }
 
-#if SSL_VERIFY_CERT
-    SSL_CTX_set_verify(data->ctx,
+    if(data->ssl.verifypeer){
+      SSL_CTX_set_verify(data->ssl.ctx,
                          SSL_VERIFY_PEER|SSL_VERIFY_FAIL_IF_NO_PEER_CERT|
                          SSL_VERIFY_CLIENT_ONCE,
                          cert_verify_callback);
-#endif
+      if (!SSL_CTX_load_verify_locations(data->ssl.ctx,
+                                         data->ssl.CAfile,
+                                         data->ssl.CApath)) {
+        failf(data,"error setting cerficate verify locations\n");
+        return 2;
+      }
+    }
+    else
+      SSL_CTX_set_verify(data->ssl.ctx, SSL_VERIFY_NONE, cert_verify_callback);
+
 
     /* Lets make an SSL structure */
-    data->ssl = SSL_new (data->ctx);
-    SSL_set_connect_state (data->ssl);
+    data->ssl.handle = SSL_new (data->ssl.ctx);
+    SSL_set_connect_state (data->ssl.handle);
 
-    data->server_cert = 0x0;
+    data->ssl.server_cert = 0x0;
 
     /* pass the raw socket into the SSL layers */
-    SSL_set_fd (data->ssl, data->firstsocket);
-    err = SSL_connect (data->ssl);
+    SSL_set_fd (data->ssl.handle, data->firstsocket);
+    err = SSL_connect (data->ssl.handle);
 
     if (-1 == err) {
       err = ERR_get_error(); 
@@ -244,8 +253,9 @@ UrgSSLConnect (struct UrlData *data)
       return 10;
     }
 
-
-    infof (data, "SSL connection using %s\n", SSL_get_cipher (data->ssl));
+    /* Informational message */
+    infof (data, "SSL connection using %s\n",
+           SSL_get_cipher(data->ssl.handle));
   
     /* Get server's certificate (note: beware of dynamic allocation) - opt */
     /* major serious hack alert -- we should check certificates
@@ -253,40 +263,42 @@ UrgSSLConnect (struct UrlData *data)
      * attack
      */
 
-    data->server_cert = SSL_get_peer_certificate (data->ssl);
-    if(!data->server_cert) {
+    data->ssl.server_cert = SSL_get_peer_certificate (data->ssl.handle);
+    if(!data->ssl.server_cert) {
       failf(data, "SSL: couldn't get peer certificate!");
       return 3;
     }
     infof (data, "Server certificate:\n");
   
-    str = X509_NAME_oneline (X509_get_subject_name (data->server_cert), NULL, 0);
+    str = X509_NAME_oneline (X509_get_subject_name (data->ssl.server_cert),
+                             NULL, 0);
     if(!str) {
       failf(data, "SSL: couldn't get X509-subject!");
       return 4;
     }
     infof(data, "\t subject: %s\n", str);
-    Free (str);
+    CRYPTO_free(str);
 
-    str = X509_NAME_oneline (X509_get_issuer_name  (data->server_cert), NULL, 0);
+    str = X509_NAME_oneline (X509_get_issuer_name  (data->ssl.server_cert),
+                             NULL, 0);
     if(!str) {
       failf(data, "SSL: couldn't get X509-issuer name!");
       return 5;
     }
     infof(data, "\t issuer: %s\n", str);
-    Free (str);
+    CRYPTO_free(str);
 
     /* We could do all sorts of certificate verification stuff here before
        deallocating the certificate. */
 
+    if(data->ssl.verifypeer) {
+      data->ssl.certverifyresult=SSL_get_verify_result(data->ssl.handle);
+      infof(data, "Verify result: %d\n", data->ssl.certverifyresult);
+    }
+    else
+      data->ssl.certverifyresult=0;
 
-#if SSL_VERIFY_CERT
-    infof(data, "Verify result: %d\n", SSL_get_verify_result(data->ssl));
-#endif
-
-
-
-    X509_free (data->server_cert);
+    X509_free(data->ssl.server_cert);
 #else /* USE_SSLEAY */
     /* this is for "-ansi -Wall -pedantic" to stop complaining!   (rabe) */
     (void) data;

lib/telnet.c

@@ -714,7 +714,7 @@ void telrcv(struct UrlData *data,
 	    break;   /* Ignore \0 after CR */
 	 }
 
-	 data->fwrite((char *)&c, 1, 1, data->out);
+	 client_write(data, CLIENTWRITE_BODY, (char *)&c, 1);
 	 continue;
 
       case TS_DATA:
@@ -728,7 +728,7 @@ void telrcv(struct UrlData *data,
 	    telrcv_state = TS_CR;
 	 }
 
-	 data->fwrite((char *)&c, 1, 1, data->out);
+	 client_write(data, CLIENTWRITE_BODY, (char *)&c, 1);
 	 continue;
 
       case TS_IAC:
@@ -752,7 +752,7 @@ void telrcv(struct UrlData *data,
 	   telrcv_state = TS_SB;
 	   continue;
 	case IAC:
-	   data->fwrite((char *)&c, 1, 1, data->out);
+          client_write(data, CLIENTWRITE_BODY, (char *)&c, 1);
           break;
 	case DM:
 	case NOP:
@@ -861,8 +861,9 @@ void telwrite(struct UrlData *data,
 #ifndef USE_SSLEAY
       bytes_written = swrite(data->firstsocket, outbuf, out_count);
 #else
-      if (data->use_ssl) {
-        bytes_written = SSL_write(data->ssl, (char *)outbuf, out_count);
+      if (data->ssl.use) {
+        bytes_written = SSL_write(data->ssl.handle, (char *)outbuf,
+                                  out_count);
       }
       else {
         bytes_written = swrite(data->firstsocket, outbuf, out_count);
@@ -918,8 +919,8 @@ CURLcode telnet(struct connectdata *conn)
 #ifndef USE_SSLEAY
 	    nread = sread (sockfd, buf, BUFSIZE - 1);
 #else
-	    if (data->use_ssl) {
-	       nread = SSL_read (data->ssl, buf, BUFSIZE - 1);
+	    if (data->ssl.use) {
+	       nread = SSL_read (data->ssl.handle, buf, BUFSIZE - 1);
 	    }
 	    else {
 	       nread = sread (sockfd, buf, BUFSIZE - 1);

lib/url.c

@@ -52,7 +52,6 @@
 
 #include <errno.h>
 
-
 #if defined(WIN32) && !defined(__GNUC__) || defined(__MINGW32__)
 #include <winsock.h>
 #include <time.h>
@@ -110,7 +109,7 @@
 #include "progress.h"
 #include "cookie.h"
 #include "strequal.h"
-#include "writeout.h"
+#include "escape.h"
 
 /* And now for the protocols */
 #include "ftp.h"
@@ -125,6 +124,14 @@
 #define _MPRINTF_REPLACE /* use our functions only */
 #include <curl/mprintf.h>
 
+#ifdef KRB4
+#include "security.h"
+#endif
+/* The last #include file should be: */
+#ifdef MALLOCDEBUG
+#include "memdebug.h"
+#endif
+
 /* -- -- */
 
 
@@ -144,19 +151,19 @@ void curl_free(void)
 void static urlfree(struct UrlData *data, bool totally)
 {
 #ifdef USE_SSLEAY
-  if (data->use_ssl) {
-    if(data->ssl) {
-      (void)SSL_shutdown(data->ssl);
-      SSL_set_connect_state(data->ssl);
+  if (data->ssl.use) {
+    if(data->ssl.handle) {
+      (void)SSL_shutdown(data->ssl.handle);
+      SSL_set_connect_state(data->ssl.handle);
 
-      SSL_free (data->ssl);
-      data->ssl = NULL;
+      SSL_free (data->ssl.handle);
+      data->ssl.handle = NULL;
     }
-    if(data->ctx) {
-      SSL_CTX_free (data->ctx);
-      data->ctx = NULL;
+    if(data->ssl.ctx) {
+      SSL_CTX_free (data->ssl.ctx);
+      data->ssl.ctx = NULL;
     }
-    data->use_ssl = FALSE; /* get back to ordinary socket usage */
+    data->ssl.use = FALSE; /* get back to ordinary socket usage */
   }
 #endif /* USE_SSLEAY */
 
@@ -181,6 +188,12 @@ void static urlfree(struct UrlData *data, bool totally)
     data->bits.httpproxy=FALSE;
   }
   
+  if(data->bits.rangestringalloc) {
+    free(data->range);
+    data->range=NULL;
+    data->bits.rangestringalloc=0; /* free now */
+  }
+
   if(data->ptr_proxyuserpwd) {
     free(data->ptr_proxyuserpwd);
     data->ptr_proxyuserpwd=NULL;
@@ -224,6 +237,10 @@ void static urlfree(struct UrlData *data, bool totally)
     if(data->free_referer)
       free(data->referer);
 
+    if(data->bits.urlstringalloc)
+      /* the URL is allocated, free it! */
+      free(data->url);
+
     cookie_cleanup(data->cookies);
 
     free(data);
@@ -248,6 +265,17 @@ CURLcode curl_close(CURL *curl)
   return CURLE_OK;
 }
 
+int my_getpass(void *clientp, char *prompt, char* buffer, int buflen )
+{
+  char *retbuf;
+  retbuf = getpass_r(prompt, buffer, buflen);
+  if(NULL == retbuf)
+    return 1;
+  else
+    return 0; /* success */
+}
+
+
 CURLcode curl_open(CURL **curl, char *url)
 {
   /* We don't yet support specifying the URL at this point */
@@ -270,13 +298,6 @@ CURLcode curl_open(CURL **curl, char *url)
 
     data-> headersize=HEADERSIZE;
 
-#if 0
-    /* Let's set some default values: */
-    curl_setopt(data, CURLOPT_FILE, stdout); /* default output to stdout */
-    curl_setopt(data, CURLOPT_INFILE, stdin);  /* default input from stdin */
-    curl_setopt(data, CURLOPT_STDERR, stderr);  /* default stderr to stderr! */
-#endif
-
     data->out = stdout; /* default output to stdout */
     data->in  = stdin;  /* default input from stdin */
     data->err  = stderr;  /* default stderr to stderr */
@@ -290,6 +311,9 @@ CURLcode curl_open(CURL **curl, char *url)
     /* use fread as default function to read input */
     data->fread = (size_t (*)(char *, size_t, size_t, FILE *))fread;
 
+    /* set the default passwd function */
+    data->fpasswd = my_getpass;
+
     data->infilesize = -1; /* we don't know any size */
 
     data->current_speed = -1; /* init to negative == impossible */
@@ -334,6 +358,9 @@ CURLcode curl_setopt(CURL *curl, CURLoption option, ...)
   case CURLOPT_POST:
     data->bits.http_post = va_arg(param, long)?TRUE:FALSE;
     break;
+  case CURLOPT_FILETIME:
+    data->bits.get_filetime = va_arg(param, long)?TRUE:FALSE;
+    break;
   case CURLOPT_FTPLISTONLY:
     data->bits.ftp_list_only = va_arg(param, long)?TRUE:FALSE;
     break;
@@ -365,7 +392,7 @@ CURLcode curl_setopt(CURL *curl, CURLoption option, ...)
     break;
 
   case CURLOPT_SSLVERSION:
-    data->ssl_version = va_arg(param, long);
+    data->ssl.version = va_arg(param, long);
     break;
 
   case CURLOPT_COOKIEFILE:
@@ -416,9 +443,7 @@ CURLcode curl_setopt(CURL *curl, CURLoption option, ...)
     data->url = va_arg(param, char *);
     break;
   case CURLOPT_PORT:
-    /* this typecast is used to fool the compiler to NOT warn for a
-       "cast from pointer to integer of different size" */
-    data->port = (unsigned short)(va_arg(param, long));
+    data->port = va_arg(param, long);
     break;
   case CURLOPT_POSTFIELDS:
     data->postfields = va_arg(param, char *);
@@ -437,12 +462,18 @@ CURLcode curl_setopt(CURL *curl, CURLoption option, ...)
     data->proxy = va_arg(param, char *);
     data->bits.httpproxy = data->proxy?1:0;
     break;
+  case CURLOPT_HTTPPROXYTUNNEL:
+    data->bits.tunnel_thru_httpproxy = va_arg(param, long)?TRUE:FALSE;
+    break;
   case CURLOPT_PROXYPORT:
     data->proxyport = va_arg(param, long);
     break;
   case CURLOPT_TIMEOUT:
     data->timeout = va_arg(param, long);
     break;
+  case CURLOPT_MAXREDIRS:
+    data->maxredirs = va_arg(param, long);
+    break;
   case CURLOPT_USERAGENT:
     data->useragent = va_arg(param, char *);
     break;
@@ -460,6 +491,12 @@ CURLcode curl_setopt(CURL *curl, CURLoption option, ...)
   case CURLOPT_PROGRESSDATA:
     data->progress_client = va_arg(param, void *);
     break;
+  case CURLOPT_PASSWDFUNCTION:
+    data->fpasswd = va_arg(param, curl_passwd_callback);
+    break;
+  case CURLOPT_PASSWDDATA:
+    data->passwd_client = va_arg(param, void *);
+    break;
   case CURLOPT_PROXYUSERPWD:
     data->proxyuserpwd = va_arg(param, char *);
     data->bits.proxy_user_passwd = data->proxyuserpwd?1:0;
@@ -477,9 +514,6 @@ CURLcode curl_setopt(CURL *curl, CURLoption option, ...)
   case CURLOPT_WRITEFUNCTION:
     data->fwrite = va_arg(param, curl_write_callback);
     break;
-  case CURLOPT_WRITEINFO:
-    data->writeinfo = va_arg(param, char *);
-    break;
   case CURLOPT_READFUNCTION:
     data->fread = va_arg(param, curl_read_callback);
     break;
@@ -495,6 +529,20 @@ CURLcode curl_setopt(CURL *curl, CURLoption option, ...)
   case CURLOPT_QUOTE:
     data->quote = va_arg(param, struct curl_slist *);
     break;
+  case CURLOPT_INTERFACE:
+    data->device = va_arg(param, char *);
+    break;
+  case CURLOPT_KRB4LEVEL:
+    data->krb4_level = va_arg(param, char *);
+    data->bits.krb4=data->krb4_level?TRUE:FALSE;
+    break;
+  case CURLOPT_SSL_VERIFYPEER:
+    data->ssl.verifypeer = va_arg(param, long);
+    break;
+  case CURLOPT_CAINFO:
+    data->ssl.CAfile = va_arg(param, char *);
+    data->ssl.CApath = NULL; /*This does not work on windows.*/
+    break;
   default:
     /* unknown tag and its companion, just ignore: */
     return CURLE_READ_ERROR; /* correct this */
@@ -519,8 +567,8 @@ int GetLine(int sockfd, char *buf, struct UrlData *data)
       (nread<BUFSIZE) && read_rc;
       nread++, ptr++) {
 #ifdef USE_SSLEAY
-    if (data->use_ssl) {
-      read_rc = SSL_read(data->ssl, ptr, 1);
+    if (data->ssl.use) {
+      read_rc = SSL_read(data->ssl.handle, ptr, 1);
     }
     else {
 #endif
@@ -566,10 +614,15 @@ CURLcode curl_write(CURLconnect *c_conn, char *buf, size_t amount,
   data = conn->data;
 
 #ifdef USE_SSLEAY
-  if (data->use_ssl) {
-    bytes_written = SSL_write(data->ssl, buf, amount);
+  if (data->ssl.use) {
+    bytes_written = SSL_write(data->ssl.handle, buf, amount);
   }
   else {
+#endif
+#ifdef KRB4
+    if(conn->sec_complete)
+      bytes_written = sec_write(conn, conn->writesockfd, buf, amount);
+    else
 #endif
       bytes_written = swrite(conn->writesockfd, buf, amount);
 #ifdef USE_SSLEAY
@@ -592,10 +645,15 @@ CURLcode curl_read(CURLconnect *c_conn, char *buf, size_t buffersize,
   data = conn->data;
 
 #ifdef USE_SSLEAY
-  if (data->use_ssl) {
-    nread = SSL_read (data->ssl, buf, buffersize);
+  if (data->ssl.use) {
+    nread = SSL_read (data->ssl.handle, buf, buffersize);
   }
   else {
+#endif
+#ifdef KRB4
+    if(conn->sec_complete)
+      nread = sec_read(conn, conn->sockfd, buf, buffersize);
+    else
 #endif
       nread = sread (conn->sockfd, buf, buffersize);
 #ifdef USE_SSLEAY
@@ -611,6 +669,12 @@ CURLcode curl_disconnect(CURLconnect *c_connect)
 
   struct UrlData *data = conn->data;
 
+  if(conn->hostent_buf) /* host name info */
+    free(conn->hostent_buf);
+
+  if(conn->path) /* the URL path part */
+    free(conn->path);
+
   free(conn); /* free the connection oriented data */
 
   /* clean up the sockets and SSL stuff from the previous "round" */
@@ -619,38 +683,18 @@ CURLcode curl_disconnect(CURLconnect *c_connect)
   return CURLE_OK;
 }
 
-/*
- * NAME curl_connect()
- *
- * DESCRIPTION
- *
- * Connects to the peer server and performs the initial setup. This function
- * writes a connect handle to its second argument that is a unique handle for
- * this connect. This allows multiple connects from the same handle returned
- * by curl_open().
- *
- * EXAMPLE
- *
- * CURLCode result;
- * CURL curl;
- * CURLconnect connect;
- * result = curl_connect(curl, &connect);
- */
-
-CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
+static CURLcode _connect(CURL *curl, CURLconnect **in_connect)
 {
   char *tmp;
   char *buf;
   CURLcode result;
-  char resumerange[12]="";
+  char resumerange[40]="";
   struct UrlData *data = curl;
   struct connectdata *conn;
-
-  /* I believe the longest possible name in a DNS is set to 255 letters, FQDN.
-     Although the buffer required for storing all possible aliases and IP
-     numbers is according to Stevens' Unix Network Programming 2nd editor,
-     p. 304: 8192 bytes. Let's go with that! */
-  char hostent_buf[8192];
+#ifdef HAVE_SIGACTION
+  struct sigaction sigact;
+#endif
+  int urllen;
 
   if(!data || (data->handle != STRUCT_OPEN))
     return CURLE_BAD_FUNCTION_ARGUMENT; /* TBD: make error codes */
@@ -671,19 +715,43 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
   conn->data = data; /* remember our daddy */
   conn->state = CONN_INIT;
 
+  conn->upload_bufsize = UPLOAD_BUFSIZE; /* the smallest upload buffer size
+                                            we use */
+
   buf = data->buffer; /* this is our buffer */
 
-#if 0
+#ifdef HAVE_SIGACTION
+  sigaction(SIGALRM, NULL, &sigact);
+  sigact.sa_handler = alarmfunc;
+  sigact.sa_flags &= ~SA_RESTART;
+  sigaction(SIGALRM, &sigact, NULL);
+#else
+  /* no sigaction(), revert to the much lamer signal() */
+#ifdef HAVE_SIGNAL
   signal(SIGALRM, alarmfunc);
 #endif
 
+#endif
+
+  /* We need to allocate memory to store the path in. We get the size of the
+     full URL to be sure, and we need to make it at least 256 bytes since
+     other parts of the code will rely on this fact */
+#define LEAST_PATH_ALLOC 256
+  urllen=strlen(data->url);
+  if(urllen < LEAST_PATH_ALLOC)
+    urllen=LEAST_PATH_ALLOC;
+  
+  conn->path=(char *)malloc(urllen);
+  if(NULL == conn->path)
+    return CURLE_OUT_OF_MEMORY; /* really bad error */
+
   /* Parse <url> */
   /* We need to parse the url, even when using the proxy, because
    * we will need the hostname and port in case we are trying
    * to SSL connect through the proxy -- and we don't know if we
    * will need to use SSL until we parse the url ...
    */
-  if((2 == sscanf(data->url, "%64[^:]://%" URL_MAX_LENGTH_TXT "[^\n]",
+  if((2 == sscanf(data->url, "%64[^:]://%[^\n]",
                   conn->proto,
                   conn->path)) && strequal(conn->proto, "file")) {
     /* we deal with file://<host>/<path> differently since it
@@ -703,11 +771,11 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
     strcpy(conn->path, "/");
 
     if (2 > sscanf(data->url,
-                   "%64[^\n:]://%256[^\n/]%" URL_MAX_LENGTH_TXT "[^\n]",
+                   "%64[^\n:]://%256[^\n/]%[^\n]",
                    conn->proto, conn->gname, conn->path)) {
       
       /* badly formatted, let's try the browser-style _without_ 'http://' */
-      if((1 > sscanf(data->url, "%256[^\n/]%" URL_MAX_LENGTH_TXT "[^\n]",
+      if((1 > sscanf(data->url, "%256[^\n/]%[^\n]",
                      conn->gname, conn->path)) ) {
         failf(data, "<url> malformed");
         return CURLE_URL_MALFORMAT;
@@ -742,16 +810,19 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
 
     if(*data->userpwd != ':') {
       /* the name is given, get user+password */
-      sscanf(data->userpwd, "%127[^:]:%127[^@]",
+      sscanf(data->userpwd, "%127[^:]:%127[^\n]",
              data->user, data->passwd);
       }
     else
       /* no name given, get the password only */
-      sscanf(data->userpwd+1, "%127[^@]", data->passwd);
+      sscanf(data->userpwd+1, "%127[^\n]", data->passwd);
 
     /* check for password, if no ask for one */
     if( !data->passwd[0] ) {
-      strncpy(data->passwd, getpass("password: "), sizeof(data->passwd));
+      if(!data->fpasswd ||
+         data->fpasswd(data->passwd_client,
+                       "password:", data->passwd, sizeof(data->passwd)))
+        return CURLE_BAD_PASSWORD_ENTERED;
     }
   }
 
@@ -761,16 +832,21 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
 
     if(*data->proxyuserpwd != ':') {
       /* the name is given, get user+password */
-      sscanf(data->proxyuserpwd, "%127[^:]:%127[^@]",
+      sscanf(data->proxyuserpwd, "%127[^:]:%127[^\n]",
              data->proxyuser, data->proxypasswd);
       }
     else
       /* no name given, get the password only */
-      sscanf(data->proxyuserpwd+1, "%127[^@]", data->proxypasswd);
+      sscanf(data->proxyuserpwd+1, "%127[^\n]", data->proxypasswd);
 
     /* check for password, if no ask for one */
     if( !data->proxypasswd[0] ) {
-      strncpy(data->proxypasswd, getpass("proxy password: "), sizeof(data->proxypasswd));
+      if(!data->fpasswd ||
+         data->fpasswd( data->passwd_client,
+                        "proxy password:",
+                        data->proxypasswd,
+                        sizeof(data->proxypasswd)))
+        return CURLE_BAD_PASSWORD_ENTERED;
     }
 
   }
@@ -784,20 +860,29 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
     /* If proxy was not specified, we check for default proxy environment
        variables, to enable i.e Lynx compliance:
 
-       HTTP_PROXY http://some.server.dom:port/
-       HTTPS_PROXY http://some.server.dom:port/
-       FTP_PROXY http://some.server.dom:port/
-       GOPHER_PROXY http://some.server.dom:port/
-       NO_PROXY host.domain.dom  (a comma-separated list of hosts which should
-       not be proxied, or an asterisk to override all proxy variables)
-       ALL_PROXY seems to exist for the CERN www lib. Probably the first to
-       check for.
+       http_proxy=http://some.server.dom:port/
+       https_proxy=http://some.server.dom:port/
+       ftp_proxy=http://some.server.dom:port/
+       gopher_proxy=http://some.server.dom:port/
+       no_proxy=domain1.dom,host.domain2.dom
+                                 (a comma-separated list of hosts which should
+                                  not be proxied, or an asterisk to override
+                                  all proxy variables)
+       all_proxy=http://some.server.dom:port/
+                                 (seems to exist for the CERN www lib. Probably
+                                  the first to check for.)
 
+       For compatibility, the all-uppercase versions of these variables are
+       checked if the lowercase versions don't exist.
        */
-    char *no_proxy=GetEnv("NO_PROXY");
+    char *no_proxy=NULL;
     char *proxy=NULL;
     char proxy_env[128];
 
+    no_proxy=GetEnv("no_proxy");
+    if(!no_proxy)
+      no_proxy=GetEnv("NO_PROXY");
+
     if(!no_proxy || !strequal("*", no_proxy)) {
       /* NO_PROXY wasn't specified or it wasn't just an asterisk */
       char *nope;
@@ -820,23 +905,31 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
 	char *envp = proxy_env;
 	char *prox;
 
-	/* Now, build <PROTOCOL>_PROXY and check for such a one to use */
-	while(*protop) {
-	  *envp++ = toupper(*protop++);
-	}
-	/* append _PROXY */
-	strcpy(envp, "_PROXY");
-#if 0
-	infof(data, "DEBUG: checks the environment variable %s\n", proxy_env);
-#endif
+	/* Now, build <protocol>_proxy and check for such a one to use */
+	while(*protop)
+	  *envp++ = tolower(*protop++);
+
+	/* append _proxy */
+	strcpy(envp, "_proxy");
+
 	/* read the protocol proxy: */
 	prox=GetEnv(proxy_env);
 
+	if(!prox) {
+          /* There was no lowercase variable, try the uppercase version: */
+	  for(envp = proxy_env; *envp; envp++)
+	    *envp = toupper(*envp);
+	  prox=GetEnv(proxy_env);
+	}
+
 	if(prox && *prox) { /* don't count "" strings */
 	  proxy = prox; /* use this */
 	}
-        else
-          proxy = GetEnv("ALL_PROXY"); /* default proxy to use */
+	else {
+	  proxy = GetEnv("all_proxy"); /* default proxy to use */
+	  if(!proxy)
+            proxy=GetEnv("ALL_PROXY");
+	}
 
         if(proxy && *proxy) {
           /* we have a proxy here to set */
@@ -844,7 +937,7 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
           data->bits.proxystringalloc=1; /* this needs to be freed later */
           data->bits.httpproxy=1;
         }
-      } /* if (!nope) - it wasn't specfied non-proxy */
+      } /* if (!nope) - it wasn't specified non-proxy */
     } /* NO_PROXY wasn't specified or '*' */
     if(no_proxy)
       free(no_proxy);
@@ -879,8 +972,9 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
   if(data->resume_from) {
     if(!data->bits.set_range) {
       /* if it already was in use, we just skip this */
-      sprintf(resumerange, "%d-", data->resume_from);
-      data->range=resumerange; /* tell ourselves to fetch this range */
+      snprintf(resumerange, sizeof(resumerange), "%d-", data->resume_from);
+      data->range=strdup(resumerange); /* tell ourselves to fetch this range */
+      data->bits.rangestringalloc = TRUE; /* mark as allocated */
       data->bits.set_range = 1; /* switch on range usage */
     }
   }
@@ -922,7 +1016,7 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
     conn->curl_close = http_close;
 
 #else /* USE_SSLEAY */
-    failf(data, "SSL is disabled, https: not supported!");
+    failf(data, "libcurl was built with SSL disabled, https: not supported!");
     return CURLE_UNSUPPORTED_PROTOCOL;
 #endif /* !USE_SSLEAY */
   }
@@ -948,7 +1042,10 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
     data->remote_port = PORT_FTP;
     conn->protocol |= PROT_FTP;
 
-    if(data->bits.httpproxy) {
+    if(data->bits.httpproxy &&
+       !data->bits.tunnel_thru_httpproxy) {
+      /* Unless we have asked to tunnel ftp operations through the proxy, we
+         switch and use HTTP operations only */
       conn->curl_do = http;
       conn->curl_done = http_done;
       conn->curl_close = http_close;
@@ -1018,6 +1115,11 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
 
     conn->curl_do = file;
     /* no done() function */
+
+    result = Transfer(conn, -1, -1, FALSE, NULL, /* no download */
+                      -1, NULL); /* no upload */
+
+    return CURLE_OK;
   }
 
   else {
@@ -1047,13 +1149,7 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
        user+password pair in a string like:
        ftp://user:password@ftp.my.site:8021/README */
     char *ptr=NULL; /* assign to remove possible warnings */
-#if 0
-    if(':' == *conn->name) {
-      failf(data, "URL malformat: user can't be zero length");
-      return CURLE_URL_MALFORMAT_USER;
-    }
-#endif
-    if(ptr=strchr(conn->name, '@')) {
+    if((ptr=strchr(conn->name, '@'))) {
       /* there's a user+password given here, to the left of the @ */
 
       data->user[0] =0;
@@ -1061,16 +1157,37 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
 
       if(*conn->name != ':') {
         /* the name is given, get user+password */
-        sscanf(conn->name, "%127[^:]:%127[^@]",
+        sscanf(conn->name, "%127[^:@]:%127[^@]",
                data->user, data->passwd);
       }
       else
         /* no name given, get the password only */
         sscanf(conn->name+1, "%127[^@]", data->passwd);
 
+      if(data->user[0]) {
+        char *newname=curl_unescape(data->user, 0);
+        if(strlen(newname) < sizeof(data->user)) {
+          strcpy(data->user, newname);
+        }
+        /* if the new name is longer than accepted, then just use
+           the unconverted name, it'll be wrong but what the heck */
+        free(newname);
+      }
+
       /* check for password, if no ask for one */
       if( !data->passwd[0] ) {
-        strncpy(data->passwd, getpass("password: "), sizeof(data->passwd));
+        if(!data->fpasswd ||
+           data->fpasswd(data->passwd_client,
+                         "password:",data->passwd,sizeof(data->passwd)))
+          return CURLE_BAD_PASSWORD_ENTERED;
+      }
+      else {
+        /* we have a password found in the URL, decode it! */
+        char *newpasswd=curl_unescape(data->passwd, 0);
+        if(strlen(newpasswd) < sizeof(data->passwd)) {
+          strcpy(data->passwd, newpasswd);
+        }
+        free(newpasswd);
       }
 
       conn->name = ++ptr;
@@ -1090,10 +1207,12 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
       *tmp++ = '\0';
       data->port = atoi(tmp);
     }
+    data->remote_port = data->port; /* it is the same port */
 
     /* Connect to target host right on */
-    if(!(conn->hp = GetHost(data, conn->name, hostent_buf, sizeof(hostent_buf)))) {
-      failf(data, "Couldn't resolv host '%s'", conn->name);
+    conn->hp = GetHost(data, conn->name, &conn->hostent_buf);
+    if(!conn->hp) {
+      failf(data, "Couldn't resolve host '%s'", conn->name);
       return CURLE_COULDNT_RESOLVE_HOST;
     }
   }
@@ -1147,8 +1266,9 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
     }
 
     /* connect to proxy */
-    if(!(conn->hp = GetHost(data, proxyptr, hostent_buf, sizeof(hostent_buf)))) {
-      failf(data, "Couldn't resolv proxy '%s'", proxyptr);
+    conn->hp = GetHost(data, proxyptr, &conn->hostent_buf);
+    if(!conn->hp) {
+      failf(data, "Couldn't resolve proxy '%s'", proxyptr);
       return CURLE_COULDNT_RESOLVE_PROXY;
     }
 
@@ -1164,6 +1284,126 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
   conn->serv_addr.sin_family = conn->hp->h_addrtype;
   conn->serv_addr.sin_port = htons(data->port);
 
+#ifndef WIN32 
+  /* We don't generally like checking for OS-versions, we should make this
+     HAVE_XXXX based, although at the moment I don't have a decent test for
+     this! */
+
+  /* sck 8/31/2000 add support for specifing device to bind socket to */
+  /* I am using this, but it may not work everywhere, only tested on
+     RedHat 6.2 */
+#ifdef HAVE_INET_NTOA
+
+#ifndef INADDR_NONE
+#define INADDR_NONE (unsigned long) ~0
+#endif
+
+  if (data->device && (strlen(data->device)<255)) {
+    struct sockaddr_in sa;
+    struct hostent *h=NULL;
+    char *hostdataptr=NULL;
+    size_t size;
+    char myhost[256] = "";
+    unsigned long in;
+
+    if(if2ip(data->device, myhost, sizeof(myhost))) {
+      h = GetHost(data, myhost, &hostdataptr);
+    }
+    else {
+      if(strlen(data->device)>1) {
+        h = GetHost(data, data->device, &hostdataptr);
+      }
+      if(h) {
+        /* we know data->device is shorter than the myhost array */
+        strcpy(myhost, data->device);
+      }
+    }
+
+    if(! *myhost) {
+      /* need to fix this
+         h=GetHost(data,
+         getmyhost(*myhost,sizeof(myhost)),
+         hostent_buf,
+         sizeof(hostent_buf));
+      */
+      printf("in here\n");
+    }
+
+    infof(data, "We connect from %s\n", myhost);
+
+    if ( (in=inet_addr(myhost)) != INADDR_NONE ) {
+
+      if ( h ) {
+        memset((char *)&sa, 0, sizeof(sa));
+        memcpy((char *)&sa.sin_addr,
+               h->h_addr,
+               h->h_length);
+        sa.sin_family = AF_INET;
+        sa.sin_addr.s_addr = in;
+        sa.sin_port = 0; /* get any port */
+	
+        if( bind(data->firstsocket, (struct sockaddr *)&sa, sizeof(sa)) >= 0) {
+          /* we succeeded to bind */
+          struct sockaddr_in add;
+	
+          size = sizeof(add);
+          if(getsockname(data->firstsocket, (struct sockaddr *) &add,
+                         (int *)&size)<0) {
+            failf(data, "getsockname() failed");
+            return CURLE_HTTP_PORT_FAILED;
+          }
+        }
+        else {
+          switch(errno) {
+          case EBADF:
+            failf(data, "Invalid descriptor: %d", errno);
+            break;
+          case EINVAL:
+            failf(data, "Invalid request: %d", errno);
+            break;
+          case EACCES:
+            failf(data, "Address is protected, user not superuser: %d", errno);
+            break;
+          case ENOTSOCK:
+            failf(data,
+                  "Argument is a descriptor for a file, not a socket: %d",
+                  errno);
+            break;
+          case EFAULT:
+            failf(data, "Inaccessable memory error: %d", errno);
+            break;
+          case ENAMETOOLONG:
+            failf(data, "Address too long: %d", errno);
+            break;
+          case ENOMEM:
+            failf(data, "Insufficient kernel memory was available: %d", errno);
+            break;
+          default:
+            failf(data,"errno %d\n");
+          } /* end of switch */
+	
+          return CURLE_HTTP_PORT_FAILED;
+        } /* end of else */
+	
+      } /* end of if  h */
+      else {
+	failf(data,"could't find my own IP address (%s)", myhost);
+	return CURLE_HTTP_PORT_FAILED;
+      }
+    } /* end of inet_addr */
+
+    else {
+      failf(data, "could't find my own IP address (%s)", myhost);
+      return CURLE_HTTP_PORT_FAILED;
+    }
+
+    if(hostdataptr)
+      free(hostdataptr); /* allocated by GetHost() */
+
+  } /* end of device selection support */
+#endif  /* end of HAVE_INET_NTOA */
+#endif /* end of not WIN32 */
+
   if (connect(data->firstsocket,
               (struct sockaddr *) &(conn->serv_addr),
               sizeof(conn->serv_addr)
@@ -1174,12 +1414,38 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
     case ECONNREFUSED:
       failf(data, "Connection refused");
       break;
+    case EFAULT:
+      failf(data, "Invalid socket address: %d",errno);
+      break;
+    case EISCONN:
+      failf(data, "Socket already connected: %d",errno);
+      break;
+    case ETIMEDOUT:
+      failf(data, "Timeout while accepting connection, server busy: %d",errno);
+      break;
+    case ENETUNREACH:
+      failf(data, "Network is unreachable: %d",errno);
+      break;
+    case EADDRINUSE:
+      failf(data, "Local address already in use: %d",errno);
+      break;
+    case EINPROGRESS:
+      failf(data, "Socket is nonblocking and connection can not be completed immediately: %d",errno);
+      break;
+    case EALREADY:
+      failf(data, "Socket is nonblocking and a previous connection attempt not completed: %d",errno);
+      break;
+    case EAGAIN:
+      failf(data, "No more free local ports: %d",errno);
+      break;
+    case EACCES:
+    case EPERM:
+      failf(data, "Attempt to connect to broadcast address without socket broadcast flag or local firewall rule violated: %d",errno);
+      break;
 #endif
-#ifdef EINTR
     case EINTR:
       failf(data, "Connection timeouted");
       break;
-#endif
     default:
       failf(data, "Can't connect to server: %d", errno);
       break;
@@ -1188,12 +1454,15 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
   }
 
   if(data->bits.proxy_user_passwd) {
-    char authorization[512];
-    sprintf(data->buffer, "%s:%s", data->proxyuser, data->proxypasswd);
-    base64Encode(data->buffer, authorization);
-
-    data->ptr_proxyuserpwd = maprintf("Proxy-authorization: Basic %s\015\012",
-				      authorization);
+    char *authorization;
+    snprintf(data->buffer, BUFSIZE, "%s:%s",
+             data->proxyuser, data->proxypasswd);
+    if(base64_encode(data->buffer, strlen(data->buffer),
+                    &authorization) >= 0) {
+      data->ptr_proxyuserpwd =
+        maprintf("Proxy-authorization: Basic %s\015\012", authorization);
+      free(authorization);
+    }
   }
   if((conn->protocol&PROT_HTTP) || data->bits.httpproxy) {
     if(data->useragent) {
@@ -1223,10 +1492,6 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
     infof(data, "Connected to %s (%s)\n", conn->hp->h_name, inet_ntoa(in));
   }
 
-#if 0 /* Kerberos experiements! Beware! Take cover! */
-  kerberos_connect(data, name);
-#endif
-
 #ifdef __EMX__
   /* 20000330 mgs
    * the check is quite a hack...
@@ -1242,6 +1507,52 @@ CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
   return CURLE_OK;
 }
 
+CURLcode curl_connect(CURL *curl, CURLconnect **in_connect)
+{
+  CURLcode code;
+  struct connectdata *conn;
+
+  /* call the stuff that needs to be called */
+  code = _connect(curl, in_connect);
+
+  if(CURLE_OK != code) {
+    /* We're not allowed to return failure with memory left allocated
+       in the connectdata struct, free those here */
+    conn = (struct connectdata *)*in_connect;
+    if(conn) {
+      if(conn->path)
+        free(conn->path);
+      if(conn->hostent_buf)
+        free(conn->hostent_buf);
+      free(conn);
+      *in_connect=NULL;
+    }
+  }
+  return code;
+}
+
+
+/*
+ * NAME curl_connect()
+ *
+ * DESCRIPTION
+ *
+ * Connects to the peer server and performs the initial setup. This function
+ * writes a connect handle to its second argument that is a unique handle for
+ * this connect. This allows multiple connects from the same handle returned
+ * by curl_open().
+ *
+ * EXAMPLE
+ *
+ * CURLCode result;
+ * CURL curl;
+ * CURLconnect connect;
+ * result = curl_connect(curl, &connect);
+ */
+
+
+
+
 CURLcode curl_done(CURLconnect *c_connect)
 {
   struct connectdata *conn = c_connect;
@@ -1293,13 +1604,6 @@ CURLcode curl_do(CURLconnect *in_conn)
 
   conn->state = CONN_DO; /* we have entered this state */
 
-#if 0
-  if(conn->bytecount) {
-    double ittook = tvdiff (tvnow(), conn->now);
-    infof(data, "%i bytes transfered in %.3lf seconds (%.0lf bytes/sec).\n",
-          conn->bytecount, ittook, (double)conn->bytecount/(ittook!=0.0?ittook:1));
-  }
-#endif
   return CURLE_OK;
 }
 

lib/urldata.h

@@ -99,6 +99,11 @@
 /* Download buffer size, keep it fairly big for speed reasons */
 #define BUFSIZE (1024*50)
 
+/* Upload buffer size, keep it smallish to get faster progress meter
+   updates. This should probably become dynamic and adjust to the upload
+   speed. */
+#define UPLOAD_BUFSIZE (1024*2)
+
 /* Initial size of the buffer to store headers in, it'll be enlarged in case
    of need. */
 #define HEADERSIZE 256
@@ -123,6 +128,20 @@ typedef enum {
   CONN_LAST   /* illegal state */
 } ConnState;
 
+#ifdef KRB4
+struct krb4buffer {
+  void *data;
+  size_t size;
+  size_t index;
+  int eof_flag;
+};
+enum protection_level { 
+    prot_clear, 
+    prot_safe, 
+    prot_confidential, 
+    prot_private 
+};
+#endif
 
 /*
  * The connectdata struct contains all fields and variables that should be
@@ -150,16 +169,20 @@ struct connectdata {
 #define PROT_LDAP    (1<<7)
 #define PROT_FILE    (1<<8)
 
+  char *hostent_buf; /* pointer to allocated memory for name info */
   struct hostent *hp;
   struct sockaddr_in serv_addr;
   char proto[64];
   char gname[256];
   char *name;
-  char path[URL_MAX_LENGTH];
+  char *path; /* formerly staticly this size: URL_MAX_LENGTH */
   char *ppath;
   long bytecount;
   struct timeval now;
 
+  long upload_bufsize; /* adjust as you see fit, never bigger than BUFSIZE
+                          never smaller than UPLOAD_BUFSIZE */
+
   /* These two functions MUST be set by the curl_connect() function to be
      be protocol dependent */
   CURLcode (*curl_do)(struct connectdata *connect);
@@ -188,6 +211,19 @@ struct connectdata {
                             the same we read from. -1 disables */
   long *writebytecountp; /* return number of bytes written or NULL */
 
+#ifdef KRB4
+
+  enum protection_level command_prot;
+  enum protection_level data_prot;
+  enum protection_level request_data_prot;
+
+  size_t buffer_size;
+
+  struct krb4buffer in_buffer, out_buffer;
+  int sec_complete;
+  void *app_data;
+
+#endif
 };
 
 struct Progress {
@@ -208,11 +244,19 @@ struct Progress {
   double ulspeed;
 
   struct timeval start;
+  struct timeval t_startsingle;
   /* various data stored for possible later report */
-  struct timeval t_nslookup;
-  struct timeval t_connect;
-  struct timeval t_pretransfer;
+  double t_nslookup;
+  double t_connect;
+  double t_pretransfer;
   int httpcode;
+  time_t filetime; /* If requested, this is might get set. It may be 0 if
+                      the time was unretrievable */
+
+#define CURR_TIME 5
+
+  double speeder[ CURR_TIME ];
+  int speeder_c;
 };
 
 /****************************************************************************
@@ -245,6 +289,8 @@ struct FTP {
 };
 
 struct Configbits {
+  bool get_filetime;
+  bool tunnel_thru_httpproxy;
   bool ftp_append;
   bool ftp_ascii;
   bool ftp_list_only;
@@ -262,7 +308,6 @@ struct Configbits {
   bool mute;
   bool no_body;
   bool proxy_user_passwd;
-  bool proxystringalloc; /* the http proxy string is malloc()'ed */
   bool set_port;
   bool set_range;
   bool upload;
@@ -270,6 +315,11 @@ struct Configbits {
   bool user_passwd;
   bool verbose;
   bool this_is_a_follow; /* this is a followed Location: request */
+  bool krb4; /* kerberos4 connection requested */
+
+  bool proxystringalloc; /* the http proxy string is malloc()'ed */
+  bool rangestringalloc; /* the range string is malloc()'ed */
+  bool urlstringalloc;   /* the URL string is malloc()'ed */
 };
 
 /* What type of interface that intiated this struct */
@@ -280,6 +330,21 @@ typedef enum {
   CURLI_LAST
 } CurlInterface;
 
+struct ssldata {
+  bool use;              /* use ssl encrypted communications TRUE/FALSE */
+  long version;          /* what version the client wants to use */
+  long certverifyresult; /* result from the certificate verification */
+  long verifypeer;       /* set TRUE if this is desired */
+  char *CApath;          /* DOES NOT WORK ON WINDOWS */
+  char *CAfile;          /* cerficate to verify peer against */
+#ifdef USE_SSLEAY
+  /* these ones requires specific SSL-types */
+  SSL_CTX* ctx;
+  SSL*     handle;
+  X509*    server_cert;
+#endif /* USE_SSLEAY */
+};
+
 /*
  * As of April 11, 2000 we're now trying to split up the urldata struct in
  * three different parts:
@@ -313,6 +378,10 @@ struct UrlData {
                      proxy string features a ":[port]" that one will override
                      this. */
 
+  
+  long header_size;  /* size of read header(s) in bytes */
+  long request_size; /* the amount of bytes sent in the request(s) */
+
   /*************** Request - specific items ************/
 
   union {
@@ -335,7 +404,7 @@ struct UrlData {
   char *url;   /* what to get */
   char *freethis; /* if non-NULL, an allocated string for the URL */
   char *hostname; /* hostname to connect, as parsed from url */
-  unsigned short port; /* which port to use (if non-protocol bind) set
+  long port; /* which port to use (if non-protocol bind) set
                 CONF_PORT to use this */
   unsigned short remote_port; /* what remote port to connect to, not the proxy
 				 port! */
@@ -344,18 +413,26 @@ struct UrlData {
   char *userpwd;  /* <user:password>, if used */
   char *range; /* range, if used. See README for detailed specification on
                   this syntax. */
+
+  /* stuff related to HTTP */
+
+  long followlocation;
+  long maxredirs; /* maximum no. of http(s) redirects to follow */
+  char *referer;
+  bool free_referer; /* set TRUE if 'referer' points to a string we
+                        allocated */
+  char *useragent;   /* User-Agent string */
   char *postfields; /* if POST, set the fields' values here */
   long postfieldsize; /* if POST, this might have a size to use instead of
                          strlen(), and then the data *may* be binary (contain
                          zero bytes) */
 
-  bool free_referer; /* set TRUE if 'referer' points to a string we
-                        allocated */
-  char *referer;
-  char *useragent;   /* User-Agent string */
-
+  /* stuff related to FTP */
   char *ftpport; /* port to send with the PORT command */
 
+  /* general things */
+  char *device;  /* Interface to use */
+
   /* function that stores the output:*/
   curl_write_callback fwrite;
 
@@ -366,6 +443,10 @@ struct UrlData {
   curl_progress_callback fprogress;
   void *progress_client; /* pointer to pass to the progress callback */
 
+  /* function to call instead of the internal for password */
+  curl_passwd_callback fpasswd;
+  void *passwd_client; /* pointer to pass to the passwd callback */
+
   long timeout; /* in seconds, 0 means no timeout */
   long infilesize; /* size of file to upload, -1 means unknown */
 
@@ -387,8 +468,6 @@ struct UrlData {
 
   char *cookie;       /* HTTP cookie string to send */
 
-  short    use_ssl;   /* use ssl encrypted communications */
-
   char *newurl; /* This can only be set if a Location: was in the
 		   document headers */
 
@@ -400,12 +479,8 @@ struct UrlData {
 
   struct CookieInfo *cookies;
 
-  long ssl_version; /* what version the client wants to use */
-#ifdef USE_SSLEAY
-  SSL_CTX* ctx;
-  SSL*     ssl;
-  X509*    server_cert;
-#endif /* USE_SSLEAY */
+  struct ssldata ssl; /* this is for ssl-stuff */
+
   long crlf;
   struct curl_slist *quote;     /* before the transfer */
   struct curl_slist *postquote; /* after the transfer */
@@ -418,8 +493,11 @@ struct UrlData {
   char *headerbuff; /* allocated buffer to store headers in */
   int headersize;   /* size of the allocation */
 
+#if 0
+  /* this was removed in libcurl 7.4 */
   char *writeinfo;  /* if non-NULL describes what to output on a successful
                        completion */
+#endif
 
   struct Progress progress;
 
@@ -444,6 +522,13 @@ struct UrlData {
   char *ptr_ref; /* free later if not NULL! */
   char *ptr_cookie; /* free later if not NULL! */
   char *ptr_host; /* free later if not NULL */
+
+  char *krb4_level;
+#ifdef KRB4
+  FILE *cmdchannel;
+#endif
+
+  struct timeval keeps_speed; /* this should be request-specific */
 };
 
 #define LIBCURL_NAME "libcurl"

lib/version.c

@@ -79,8 +79,14 @@ char *curl_version(void)
   ptr=strchr(ptr, '\0');
 #endif
 
+#ifdef KRB4
+  sprintf(ptr, " (krb4 enabled)");
+  ptr += strlen(ptr);
+#endif
+
 #ifdef USE_ZLIB
   sprintf(ptr, " (zlib %s)", zlibVersion());
+  ptr += strlen(ptr);
 #endif
 
   return version;

maketgz

@@ -62,25 +62,25 @@ findprog()
 # brand new version number:
 #
  
-if { findprog autoconf >/dev/null 2>/dev/null; } then
-  echo "- No autoconf found, we leave configure as it is"
-else
-  # Replace version number in configure.in file:
-
-  CONF="configure.in"
-
-  sed 's/^AM_INIT_AUTOMAKE.*/AM_INIT_AUTOMAKE(curl,"'$version'")/g' $CONF >$CONF.new
-
-  # Save old  file
-  cp -p $CONF $CONF.old
-
-  # Make new configure.in
-  mv $CONF.new $CONF
- 
-  # Update the configure script
-  echo "Runs autoconf"
-  autoconf
-fi
+#if { findprog autoconf >/dev/null 2>/dev/null; } then
+#  echo "- No autoconf found, we leave configure as it is"
+#else
+#  # Replace version number in configure.in file:
+#
+#  CONF="configure.in"
+#
+#  sed 's/^AM_INIT_AUTOMAKE.*/AM_INIT_AUTOMAKE(curl,"'$version'")/g' $CONF >$CONF.new
+#
+#  # Save old  file
+#  cp -p $CONF $CONF.old
+#
+#  # Make new configure.in
+#  mv $CONF.new $CONF
+# 
+#  # Update the configure script
+#  echo "Runs autoconf"
+#  autoconf
+#fi
 
 ############################################################################
 #

memanalyze.pl

@@ -0,0 +1,95 @@
+#!/usr/bin/perl
+#
+# Example input:
+#
+# MEM mprintf.c:1094 malloc(32) = e5718
+# MEM mprintf.c:1103 realloc(e5718, 64) = e6118
+# MEM sendf.c:232 free(f6520)
+
+do {
+    if($ARGV[0] eq "-v") {
+        $verbose=1;
+    }
+} while (shift @ARGV);
+
+while(<STDIN>) {
+    chomp $_;
+    $line = $_;
+    if($verbose) {
+        print "IN: $line\n";
+    }
+    if($line =~ /^MEM ([^:]*):(\d*) (.*)/) {
+        # generic match for the filename+linenumber
+        $source = $1;
+        $linenum = $2;
+        $function = $3;
+
+        if($function =~ /free\(0x([0-9a-f]*)/) {
+            $addr = $1;
+            if($sizeataddr{$addr} <= 0) {
+                print "FREE ERROR: No memory allocated: $line\n";
+            }
+            else {
+                $totalmem -= $sizeataddr{$addr};
+                $sizeataddr{$addr}=0;
+                $getmem{$addr}=""; # forget after a good free()
+            }
+        }
+        elsif($function =~ /malloc\((\d*)\) = 0x([0-9a-f]*)/) {
+            $size = $1;
+            $addr = $2;
+            $sizeataddr{$addr}=$size;
+            $totalmem += $size;
+
+            $getmem{$addr}="$source:$linenum";
+        }
+        elsif($function =~ /realloc\(0x([0-9a-f]*), (\d*)\) = 0x([0-9a-f]*)/) {
+            $oldaddr = $1;
+            $newsize = $2;
+            $newaddr = $3;
+
+            $totalmem -= $sizeataddr{$oldaddr};
+            $sizeataddr{$oldaddr}=0;
+
+            $totalmem += $newsize;
+            $sizeataddr{$newaddr}=$newsize;
+
+            $getmem{$oldaddr}="";
+            $getmem{$newaddr}="$source:$linenum";
+        }
+        elsif($function =~ /strdup\(0x([0-9a-f]*)\) \((\d*)\) = 0x([0-9a-f]*)/) {
+            # strdup(a5b50) (8) = df7c0
+
+            $dup = $1;
+            $size = $2;
+            $addr = $3;
+            $getmem{$addr}="$source:$linenum";
+            $sizeataddr{$addr}=$size;
+
+            $totalmem += $size;
+        }
+        else {
+            print "Not recognized input line: $function\n";
+        }        
+    }
+    else {
+        print "Not recognized prefix line: $line\n";
+    }
+    if($verbose) {
+        print "TOTAL: $totalmem\n";
+    }
+}
+
+if($totalmem) {
+    print "Leak detected: memory still allocated: $totalmem bytes\n";
+
+    for(keys %sizeataddr) {
+        $addr = $_;
+        $size = $sizeataddr{$addr};
+        if($size) {
+            print "At $addr, there's $size bytes.\n";
+            print " allocated by ".$getmem{$addr}."\n";
+        }
+    }
+}
+

packages/Linux/RPM/README

@@ -0,0 +1,5 @@
+Author: Daniel (I'm not trustworthy, replace this!)
+
+Paul Marquis's 'make_curl_rpm' script is a fine example on how to automate the
+jobs. You need to fill in your own name and email at least.
+

packages/Linux/RPM/curl-ssl.spec

@@ -1,26 +1,26 @@
-%define ver	@VERSION@
+%define ver	7.4.2
 %define rel	1
 %define prefix	/usr
 
 Summary:	get a file from a FTP, GOPHER or HTTP server.
-Name:		@PACKAGE@-ssl
+Name:		curl-ssl
 Version:	%ver
 Release:	%rel
 Copyright:	MPL
 Group:		Utilities/Console
-Source:		@PACKAGE@-%{version}.tar.gz
-URL:		http://@PACKAGE@.haxx.se
+Source:		curl-%{version}.tar.gz
+URL:		http://curl.haxx.se
 BuildPrereq:	openssl
 BuildRoot:	/tmp/%{name}-%{version}-%{rel}-root
 Packager:	Fill In As You Wish
 Docdir:		%{prefix}/doc
 
 %description
-@PACKAGE@-ssl is a client to get documents/files from servers, using 
+curl-ssl is a client to get documents/files from servers, using 
 any of the supported protocols.  The command is designed to 
 work without user interaction or any kind of interactivity.
 
-@PACKAGE@-ssl offers a busload of useful tricks like proxy support, 
+curl-ssl offers a busload of useful tricks like proxy support, 
 user authentication, ftp upload, HTTP post, file transfer 
 resume and more.
 
@@ -31,7 +31,7 @@ Authors:
 
 
 %prep
-%setup -n @PACKAGE@-@VERSION@
+%setup -n curl-7.4.2
 
 
 %build
@@ -74,7 +74,7 @@ find ${RPM_BUILD_ROOT}%{prefix} -type f | sed -e "s#^${RPM_BUILD_ROOT}##g" >> fi
 
 
 %clean
-(cd ..; rm -rf @PACKAGE@-@VERSION@ ${RPM_BUILD_ROOT})
+(cd ..; rm -rf curl-7.4.2 ${RPM_BUILD_ROOT})
 
 
 %files -f file-lists
@@ -90,7 +90,7 @@ find ${RPM_BUILD_ROOT}%{prefix} -type f | sed -e "s#^${RPM_BUILD_ROOT}##g" >> fi
 %doc MPL-1.0.txt
 %doc README
 %doc README.curl
-%doc README.lib@PACKAGE@
+%doc README.libcurl
 %doc RESOURCES
 %doc TODO
 %doc %{name}-ssl.spec.in

packages/Linux/RPM/curl.spec

@@ -1,25 +1,25 @@
-%define ver	@VERSION@
+%define ver	7.4.2
 %define rel	1
 %define prefix	/usr
 
 Summary:	get a file from a FTP, GOPHER or HTTP server.
-Name:		@PACKAGE@
+Name:		curl
 Version:	%ver
 Release:	%rel
 Copyright:	MPL
 Group:		Utilities/Console
 Source:		%{name}-%{version}.tar.gz
-URL:		http://@PACKAGE@.haxx.se
+URL:		http://curl.haxx.se
 BuildRoot:	/tmp/%{name}-%{version}-%{rel}-root
 Packager:	Fill In As You Wish
 Docdir:		%{prefix}/doc
 
 %description
-@PACKAGE@ is a client to get documents/files from servers, using 
+curl is a client to get documents/files from servers, using 
 any of the supported protocols.  The command is designed to 
 work without user interaction or any kind of interactivity.
 
-@PACKAGE@ offers a busload of useful tricks like proxy support, 
+curl offers a busload of useful tricks like proxy support, 
 user authentication, ftp upload, HTTP post, file transfer 
 resume and more.
 
@@ -88,7 +88,7 @@ find ${RPM_BUILD_ROOT}%{prefix} -type f | sed -e "s#^${RPM_BUILD_ROOT}##g" >> fi
 %doc MPL-1.0.txt
 %doc README
 %doc README.curl
-%doc README.lib@PACKAGE@
+%doc README.libcurl
 %doc RESOURCES
 %doc TODO
 %doc %{name}-ssl.spec.in

packages/Linux/RPM/make_curl_rpm

@@ -0,0 +1,62 @@
+#! /bin/sh
+# script to build curl RPM from src RPM (SSL and non-SSL versions)
+
+# initialize
+top_dir=/usr/src/redhat
+sources_dir=$top_dir/SOURCES
+specs_dir=$top_dir/SPECS
+rpms_dir=$top_dir/RPMS
+arch=`rpm --showrc | awk 'NF == 3 && $2 == "_arch" { print $3 }'`
+
+# fill in your own name and email here
+packager_name="Mr Joe Packager Person"
+packager_email='<Joe@packager.person>'
+
+# make sure we're running as root
+if test `id -u` -ne `id -u root`
+then
+   echo "you must build the RPM as root"
+   exit 1
+fi
+
+# get version and release number
+if test $# -lt 1
+then
+   echo "version number?"
+   read version
+else
+   version=$1
+fi
+
+if test $# -lt 2
+then
+   echo "release number?"
+   read release
+else
+   release=$2
+fi
+
+# build all the files
+targets="curl curl-ssl"
+for target in $targets
+do
+   # make sure src RPM exist
+   src_rpm="$target-$version-$release.src.rpm"
+   if test -f $src_rpm
+   then
+      rpm -ivh $src_rpm
+
+      # replace packager in spec file
+      sed -e 's/^Packager: .*/Packager: $packager_name $packager_email/' $specs_dir/$target.spec > $specs_dir/$target-$version-$arch.spec
+
+      # build it
+      if ! rpm -ba $specs_dir/$target-$version-$arch.spec
+      then
+         echo "error building $target for $arch -- check output above"
+      fi
+
+      echo "$target rpm is now in $rpms_dir/$arch"
+   else
+      echo $src_rpm does not exist
+   fi
+done

packages/README

@@ -0,0 +1,27 @@
+                                  _   _ ____  _     
+                              ___| | | |  _ \| |    
+                             / __| | | | |_) | |    
+                            | (__| |_| |  _ <| |___ 
+                             \___|\___/|_| \_\_____|
+
+PACKAGES
+
+ This directory and all its subdirectories are for special package
+information, template, scripts and docs. The files herein should be of use for
+those of you who want to package curl in a binary or source format using one
+of those custom formats.
+
+ The hierarchy for these directories is something like this:
+
+   packages/[OS]/[FORMAT]/
+ 
+ Currently, we have Win32 and Linux for [OS]. There might be different formats
+for the same OS so for Linux we have RPM as format.
+
+ We might need to add some differentiation for CPU as well, as there is
+Linux-RPMs for several CPUs. However, it might not be necessary since the
+packaging should be pretty much the same no matter what CPU that is used.
+    
+ For each unique OS-FORMAT pair, there's a directory to "fill"! I'd like to
+see a single README with as much details as possible, and then I'd like some
+template files for the package process.

packages/Win32/README

@@ -0,0 +1,50 @@
+Author: J<>rn Hartroth
+
+DESCRIPTION
+
+Packaging of the curl binaries for Win32 should at this point in time be based
+on the InfoZip (zip/unzip) archiver family as the de-facto standard for
+Windows archives. A package should contain the main binary curl.exe along with
+the appropriate documentation and license information files. For development
+releases, you should also include the header directory and probably the
+compiled binaries of libcurl and the appropriate Makefiles/project definition
+files for the compiler used.
+
+A simple packaging mechanism can be based on a set of batch files which call
+zip.exe with the appropriate files from the curl distribution - see the
+samples included below (Long lines have been split with "\" as the split
+marker, you'll want to rejoin the pieces to be all on one line in the batch
+file). Call any of these batch files - after compiling the curl binaries -
+with a single parameter specifying the name of the archive file to be created.
+
+It is implicitely assumed that all of the binary files (curl.exe, libcurl.a,
+etc) have previously been copied to the main directory of the curl source
+package (the directory where the main README resides), because that is where
+they should end up in the zip archive. The archive should *not* be built with
+absolute path information because the user will want to locally extract the
+archive contents and shift the binaries to his executable directory.
+
+SCRIPT_TEMPLATES
+
+curlpkg.bat:
+	zip -9 %1 curl.exe CHANGES LEGAL MPL-1.0.txt README \
+	 docs/FAQ docs/FEATURES docs/README.curl docs/README.win32 docs/TODO
+
+curldevpkg.bat:
+	zip -9 %1 curl.exe include\README include\curl\*.h CHANGES docs\* \
+	  curl.spec curl-ssl.spec LEGAL lib/Makefile.m32 src/Makefile.m32 \
+	  libcurl.a libcurl.def libcurl.dll libcurldll.a MPL-1.0.txt README
+
+PROCEDURE_EXAMPLE
+
+A standard packaging routine (for MingW32) using the above batch files could
+go like this:
+(No SSL)					(With SSL)
+cd <curl-sourcedir>\lib			cd <curl-sourcedir>\lib
+make -f Makefile.m32			make -f Makefile.m32 SSL=1
+cd ..\src					cd ..\src
+make -f Makefile.m32			make -f Makefile.m32 SSL=1
+cd ..						cd ..
+copy lib\libcurl.a .			copy lib\libcurl.a .
+copy src\curl.exe .			copy src\curl.exe .
+curlpkg curl-win32-nossl.zip		curlpkg curl-win32-ssl.zip

src/Makefile.am

@@ -3,14 +3,17 @@
 #
 
 # Some flags needed when trying to cause warnings ;-)
-# CFLAGS = -Wall -pedantic
+# CFLAGS = -g -DMALLOCDEBUG # -Wall -pedantic
 #CPPFLAGS = -DGLOBURL -DCURL_SEPARATORS
 
 INCLUDES = -I$(top_srcdir)/include
 
-bin_PROGRAMS = curl
+bin_PROGRAMS = curl #memtest
 
-curl_SOURCES = main.c hugehelp.c urlglob.c
+#memtest_SOURCES = memtest.c
+#memtest_LDADD = $(top_srcdir)/lib/libcurl.la
+
+curl_SOURCES = main.c hugehelp.c urlglob.c writeout.c
 curl_LDADD = $(top_srcdir)/lib/libcurl.la
 curl_DEPENDENCIES = $(top_srcdir)/lib/libcurl.la
 BUILT_SOURCES = hugehelp.c
@@ -22,7 +25,7 @@ EXTRA_DIST = mkhelp.pl Makefile.vc6
 AUTOMAKE_OPTIONS = foreign no-dependencies
 
 MANPAGE=$(top_srcdir)/docs/curl.1
-README=$(top_srcdir)/docs/README.curl
+README=$(top_srcdir)/docs/MANUAL
 MKHELP=$(top_srcdir)/src/mkhelp.pl
 
 # This generates the hugehelp.c file

src/Makefile.m32

@@ -1,15 +1,15 @@
 #############################################################
 ## Makefile for building curl.exe with MingW32 (GCC-2.95) and
-## optionally OpenSSL (0.9.4)
+## optionally OpenSSL (0.9.6)
 ##
-## Use: make -f Makefile.m32 [SSL=1]
+## Use: make -f Makefile.m32 [SSL=1] [DYN=1]
 ##
 ## Comments to: Troy Engel <tengel@sonic.net> or
 ##              Joern Hartroth <hartroth@acm.org>
 
 CC = gcc
 STRIP = strip -s
-OPENSSL_PATH = ../../openssl-0.9.5a
+OPENSSL_PATH = ../../openssl-0.9.6
 
 # We may need these someday
 # PERL = perl
@@ -25,10 +25,16 @@ COMPILE = $(CC) $(INCLUDES) $(CFLAGS)
 LINK = $(CC) $(CFLAGS) $(LDFLAGS) -o $@
 
 curl_PROGRAMS = curl.exe
-curl_OBJECTS =  main.o hugehelp.o urlglob.o
-curl_SOURCES = main.c hugehelp.c urlglob.c
+curl_OBJECTS = main.o hugehelp.o urlglob.o writeout.o
+curl_SOURCES = main.c hugehelp.c urlglob.c writeout.c
+ifdef DYN
+  curl_DEPENDENCIES = ../lib/libcurldll.a ../lib/libcurl.dll
+  curl_LDADD = -L../lib -lcurldll
+else
   curl_DEPENDENCIES = ../lib/libcurl.a
-curl_LDADD = -L../lib -lcurl -lwsock32
+  curl_LDADD = -L../lib -lcurl
+endif
+curl_LDADD += -lwsock32
 ifdef SSL
   curl_LDADD += -L$(OPENSSL_PATH)/out -leay32 -lssl32 -lRSAglue
 endif
@@ -37,12 +43,12 @@ PROGRAMS = $(curl_PROGRAMS)
 SOURCES = $(curl_SOURCES)
 OBJECTS = $(curl_OBJECTS)
 
-all: curl
+all: curl.exe
 
-curl: $(curl_OBJECTS) $(curl_DEPENDENCIES)
-	-@erase curl.exe
+curl.exe: $(curl_OBJECTS) $(curl_DEPENDENCIES)
+	-@erase $@
 	$(LINK) $(curl_OBJECTS) $(curl_LDADD)
-	$(STRIP) $(curl_PROGRAMS)
+	$(STRIP) $@
 
 # We don't have nroff normally under win32
 # hugehelp.c: ../README.curl ../curl.1 mkhelp.pl

src/Makefile.vc6

@@ -4,6 +4,7 @@
 ##      (default is release)
 ##
 ## Comments to: Troy Engel <tengel@sonic.net>
+## Updated by: Craig Davison <cd@securityfocus.com>
 
 PROGRAM_NAME = curl.exe
 
@@ -11,27 +12,34 @@ PROGRAM_NAME = curl.exe
 ## Nothing more to do below this line!
 
 ## Release
-CCR = cl.exe /ML /O2 /D "NDEBUG"
+CCR = cl.exe /MD /O2 /D "NDEBUG"
 LINKR = link.exe /incremental:no /libpath:"../lib"
 
 ## Debug
-CCD = cl.exe /MLd /Gm /ZI /Od /D "_DEBUG" /GZ
+CCD = cl.exe /MDd /Gm /ZI /Od /D "_DEBUG" /GZ
 LINKD = link.exe /incremental:yes /debug
 
-CFLAGS = /nologo /W3 /GX /D "WIN32" /D "_CONSOLE" /D "_MBCS" /YX /FD /c
+CFLAGS = /I "../include" /nologo /W3 /GX /D "WIN32" /D "_CONSOLE" /D "_MBCS" /YX /FD /c
 LFLAGS = /nologo /out:$(PROGRAM_NAME) /subsystem:console /machine:I386
-LINKLIBS = kernel32.lib wsock32.lib libcurl.lib
+LINKLIBS = wsock32.lib libcurl.lib
+LINKLIBS_DEBUG = wsock32.lib libcurld.lib
 
 RELEASE_OBJS= \
 	hugehelpr.obj \
+	writeoutr.obj \
+	urlglobr.obj \
 	mainr.obj
 
 DEBUG_OBJS= \
 	hugehelpd.obj \
+	writeoutd.obj \
+	urlglobd.obj \
 	maind.obj
 	
 LINK_OBJS= \
 	hugehelp.obj \
+	writeout.obj \
+	urlglob.obj \
 	main.obj
 
 all : release
@@ -40,17 +48,25 @@ release: $(RELEASE_OBJS)
 	$(LINKR) $(LFLAGS) $(LINKLIBS) $(LINK_OBJS)
 
 debug: $(DEBUG_OBJS)
-	$(LINKD) $(LFLAGS) $(LINKLIBS) $(LINK_OBJS)
+	$(LINKD) $(LFLAGS) $(LINKLIBS_DEBUG) $(LINK_OBJS)
 
 ## Release
 hugehelpr.obj: hugehelp.c
 	$(CCR) $(CFLAGS) /Zm200 hugehelp.c
+writeoutr.obj: writeout.c
+	$(CCR) $(CFLAGS) writeout.c
+urlglobr.obj: urlglob.c
+	$(CCR) $(CFLAGS) urlglob.c
 mainr.obj: main.c
 	$(CCR) $(CFLAGS) main.c
 
 ## Debug
 hugehelpd.obj: hugehelp.c
 	$(CCD) $(CFLAGS) /Zm200 hugehelp.c
+writeoutd.obj: writeout.c
+	$(CCD) $(CFLAGS) writeout.c
+urlglobd.obj: urlglob.c
+	$(CCD) $(CFLAGS) urlglob.c
 maind.obj: main.c
 	$(CCD) $(CFLAGS) main.c
 

src/hugehelp.c

@@ -87,15 +87,15 @@ puts (
 "          sent to stdout to be in text mode for win32 systems.\n"
 "\n"
 "     -c/--continue\n"
-"          Continue/Resume  a   previous   file   transfer.   This\n"
-"          instructs  curl  to continue appending data on the file\n"
-"          where it was previously left,  possibly  because  of  a\n"
-"          broken  connection to the server. There must be a named\n"
-"          physical file to append to for  this  to  work.   Note:\n"
-"          Upload  resume  is depening on a command named SIZE not\n"
-"          always present in all ftp servers! Upload resume is for\n"
-"          FTP  only.   HTTP resume is only possible with HTTP/1.1\n"
-"          or later servers.\n"
+"          Deprecated. Use '-C -' instead.  Continue/Resume a pre<EFBFBD>\n"
+"          vious  file  transfer.  This instructs curl to continue\n"
+"          appending data on the  file  where  it  was  previously\n"
+"          left,  possibly  because  of a broken connection to the\n"
+"          server. There must be a named physical file  to  append\n"
+"          to  for  this to work.  Note: Upload resume is depening\n"
+"          on a command named SIZE not always present in  all  ftp\n"
+"          servers! Upload resume is for FTP only.  HTTP resume is\n"
+"          only possible with HTTP/1.1 or later servers.\n"
 "\n"
 "     -C/--continue-at <offset>\n"
 "          Continue/Resume a previous file transfer at  the  given\n"
@@ -114,12 +114,17 @@ puts (
 "          cut off).  The data is expected  to  be  \"url-encoded\".\n"
 "          This  will  cause  curl  to pass the data to the server\n"
 "          using  the  content-type  application/x-www-form-urlen<65>\n"
-"          coded. Compare to -F.\n"
+"          coded. Compare to -F. If more than one -d/--data option\n"
+"          is used on the same command line, the data pieces spec<65>\n"
+"          ified  will be merged together with a separating &-let<65>\n"
+"          ter. Thus, using '-d name=daniel -d skill=lousy'  would\n"
+"          generate a post chunk that looks like\n"
 "\n"
 "          If you start the data with the letter @, the rest\n"
 "          should be a file name to read the data from, or - if\n"
 "          you want curl to read the data from stdin.  The con<6F>\n"
-"          tents of the file must already be url-encoded.\n"
+"          tents of the file must already be url-encoded. Multiple\n"
+"          files can also be specified.\n"
 "\n"
 "          To post data purely binary, you should instead use the\n"
 "          --data-binary option.\n"
@@ -162,6 +167,11 @@ puts (
 "          that this certificate is the private key and the pri<72>\n"
 "          vate certificate concatenated!\n"
 "\n"
+"     --cacert <CA certificate>\n"
+"          (HTTPS) Tells curl to use the specified certificate\n"
+"          file to verify the peer. The certificate must be in PEM\n"
+"          format.\n"
+"\n"
 "     -f/--fail\n"
 "          (HTTP) Fail silently (no output at all) on server\n"
 "          errors. This is mostly done like this to better enable\n"
@@ -204,30 +214,45 @@ puts (
 "          as one of the internal ones curl would use, your exter<65>\n"
 "          nally set header will be used instead of the internal\n"
 "          one. This allows you to make even trickier stuff than\n"
-"          curl would normally do. You should not  replace  inter<65>\n"
-"          nally  set  headers without knowing perfectly well what\n"
-"          you're doing. Replacing an  internal  header  with  one\n"
-"          without  content  on  the  right side of the colon will\n"
+"          curl would normally do. You should not replace\n"
+"          internally set headers without knowing perfectly well\n"
+"          what you're doing. Replacing an internal header with\n"
+"          one without content on the right side of the colon will\n"
 "          prevent that header from appearing.\n"
 "\n"
 "     -i/--include\n"
 "          (HTTP) Include the HTTP-header in the output. The HTTP-\n"
 "          header includes things like server-name, date of the\n"
 "          document, HTTP-version and more...\n"
+"\n"
+"     --interface <name>\n"
+"          Perform an operation using a specified interface. You\n"
+"          can enter interface name, IP address or host name. An\n"
+"          example could look like:\n"
+"\n"
+"          curl --interface eth0:1 http://www.netscape.com/\n"
+);
+ puts(
+"\n"
 "     -I/--head\n"
 "          (HTTP/FTP) Fetch the HTTP-header only! HTTP-servers\n"
 "          feature the command HEAD which this uses to get nothing\n"
 "          but the header of a document. When used on a FTP file,\n"
 "          curl displays the file size only.\n"
 "\n"
+"     --krb4 <level>\n"
+"          (FTP) Enable kerberos4 authentication and use. The\n"
+"          level must be entered and should be one of 'clear',\n"
+"          'safe', 'confidential' or 'private'. Should you use a\n"
+"          level that is not one of these, 'private' will instead\n"
+"          be used.\n"
+"\n"
 "     -K/--config <config file>\n"
 "          Specify which config file to read curl arguments from.\n"
 "          The config file is a text file in which command line\n"
 "          arguments can be written which then will be used as if\n"
 "          they were written on the actual command line. If the\n"
 "          first column of a config line is a '#' character, the\n"
-);
- puts(
 "          rest of the line will be treated as a comment.\n"
 "\n"
 "          Specify the filename as '-' to make curl read the file\n"
@@ -296,12 +321,20 @@ puts (
 "          or use several variables like:\n"
 "\n"
 "            curl http://{site,host}.host[1-5].com -o \"#1_#2\"\n"
-"\n"
 "     -O/--remote-name\n"
 "          Write output to a local file named like the remote file\n"
 "          we get. (Only the file part of the remote file is used,\n"
 "          the path is cut off.)\n"
 "\n"
+"     -p/--proxytunnel\n"
+"          When an HTTP proxy is used, this option will cause non-\n"
+"          HTTP protocols to attempt to tunnel through the proxy\n"
+"          instead of merely using it to do HTTP-like operations.\n"
+"          The tunnel approach is made with the HTTP proxy CONNECT\n"
+"          request and requires that the proxy allows direct con<6F>\n"
+"          nect to the remote port number curl wants to tunnel\n"
+"          through to.\n"
+"\n"
 "     -P/--ftpport <address>\n"
 "          (FTP) Reverses the initiator/listener roles when con<6F>\n"
 "          necting with ftp. This switch makes Curl use the PORT\n"
@@ -320,6 +353,7 @@ puts (
 "\n"
 "          -           (any single-letter string) to make it pick\n"
 "                      the machine's default\n"
+"\n"
 "     -q   If used as the first parameter on the command line, the\n"
 "          $HOME/.curlrc file will not be read and used as a con<6F>\n"
 "          fig file.\n"
@@ -372,15 +406,17 @@ puts (
 "     -s/--silent\n"
 "          Silent mode. Don't show progress meter or error mes<65>\n"
 "          sages.  Makes Curl mute.\n"
+"\n"
 "     -S/--show-error\n"
 "          When used with -s it makes curl show error message if\n"
 "          it fails.\n"
 "\n"
 "     -t/--upload\n"
-"          Transfer the stdin data to  the  specified  file.  Curl\n"
-"          will  read  everything  from  stdin until EOF and store\n"
-"          with the supplied name. If this is used  on  a  http(s)\n"
-"          server, the PUT command will be used.\n"
+"          Deprecated. Use '-T -' instead.  Transfer the stdin\n"
+"          data to the specified file. Curl will read everything\n"
+"          from stdin until EOF and store with the supplied name.\n"
+"          If this is used on a http(s) server, the PUT command\n"
+"          will be used.\n"
 "\n"
 "     -T/--upload-file <file>\n"
 "          Like -t, but this transfers the specified local file.\n"
@@ -404,6 +440,10 @@ puts (
 "          tion. If no password is specified, curl will ask for it\n"
 "          interactively.\n"
 "\n"
+"     --url <URL>\n"
+"          Set the URL to fetch. This option is mostly handy when\n"
+"          you wanna specify URL in a config file.\n"
+"\n"
 "     -v/--verbose\n"
 "          Makes the fetching more verbose/talkative. Mostly\n"
 "          usable for debugging. Lines starting with '>' means\n"
@@ -423,9 +463,9 @@ puts (
 "          particular file you specify it \"@filename\" and to tell\n"
 "          curl to read the format from stdin you write \"@-\".\n"
 "\n"
-"          The variables present in  the  output  format  will  be\n"
-"          substituted  by the value or text that curl thinks fit,\n"
-"          as described below. All variables  are  specified  like\n"
+"          The variables present in the output format will be sub<75>\n"
+"          stituted by the value or text that curl thinks fit, as\n"
+"          described below. All variables are specified like\n"
 "          %{variable_name} and to output a normal % you just\n"
 "          write them like %%. You can output a newline by using\n"
 "          \\n, a carrige return with \\r and a tab space with \\t.\n"
@@ -435,6 +475,8 @@ puts (
 "          doubled when using this option.\n"
 "\n"
 "          Available variables are at this point:\n"
+);
+ puts(
 "\n"
 "          url_effective  The URL that was fetched last. This is\n"
 "                         mostly meaningful if you've told curl to\n"
@@ -463,8 +505,6 @@ puts (
 "                         transfer commands and negotiations that\n"
 "                         are specific to the particular proto<74>\n"
 "                         col(s) involved.\n"
-);
- puts(
 "\n"
 "          size_download  The total amount of bytes that were\n"
 "                         downloaded.\n"
@@ -472,11 +512,18 @@ puts (
 "          size_upload    The total amount of bytes that were\n"
 "                         uploaded.\n"
 "\n"
+"          size_header    The total amount of bytes of the down<77>\n"
+"                         loaded headers.\n"
+"\n"
+"          size_request   The total amount of bytes that were sent\n"
+"                         in the HTTP request.\n"
+"\n"
 "          speed_download The average download speed that curl\n"
 "                         measured for the complete download.\n"
 "\n"
 "          speed_upload   The average upload speed that curl mea<65>\n"
-"                         sured for the complete download.\n"
+"                         sured for the complete upload.\n"
+"\n"
 "     -x/--proxy <proxyhost[:port]>\n"
 "          Use specified proxy. If the port number is not speci<63>\n"
 "          fied, it is assumed at port 1080.\n"
@@ -536,7 +583,6 @@ puts (
 "          instead. If the file name is a plain '-', it is instead\n"
 "          written to stdout. This option has no point when you're\n"
 "          using a shell with decent redirecting capabilities.\n"
-"\n"
 "FILES\n"
 "     ~/.curlrc\n"
 "          Default config file.\n"
@@ -580,6 +626,7 @@ puts (
 "\n"
 "     4    URL user malformatted. The user-part of the URL syntax\n"
 "          was not correct.\n"
+"\n"
 "     5    Couldn't resolve proxy. The given proxy host could not\n"
 "          be resolved.\n"
 "\n"
@@ -587,7 +634,6 @@ puts (
 "          resolved.\n"
 "\n"
 "     7    Failed to connect to host.\n"
-"\n"
 "     8    FTP weird server reply. The server sent data curl\n"
 "          couldn't parse.\n"
 "\n"
@@ -605,7 +651,7 @@ puts (
 "     13   FTP weird PASV reply, Curl couldn't parse the reply\n"
 "          sent to the PASV request.\n"
 "\n"
-"     14   FTP  weird 227 formay. Curl couldn't parse the 227-line\n"
+"     14   FTP weird 227 format. Curl couldn't parse the 227-line\n"
 "          the server sent.\n"
 "\n"
 "     15   FTP can't get host. Couldn't resolve the host IP we got\n"
@@ -632,13 +678,13 @@ puts (
 "\n"
 "     23   Write error. Curl couldn't write data to a local\n"
 "          filesystem or similar.\n"
+"\n"
 "     24   Malformat user. User name badly specified.\n"
 "\n"
 "     25   FTP couldn't STOR file. The server denied the STOR\n"
 "          operation.\n"
 "\n"
 "     26   Read error. Various reading problems.\n"
-"\n"
 "     27   Out of memory. A memory allocation request failed.\n"
 "\n"
 "     28   Operation timeout. The specified time-out period was\n"
@@ -677,12 +723,28 @@ puts (
 "     41   Function not found. A required LDAP function was not\n"
 "          found.\n"
 "\n"
+"     42   Aborted by callback. An application told curl to abort\n"
+"          the operation.\n"
+"\n"
+"     43   Internal error. A function was called with a bad param<61>\n"
+"          eter.\n"
+"\n"
+"     44   Internal error. A function was called in a bad order.\n"
+"\n"
+"     45   Interface error. A specified outgoing interface could\n"
+"          not be used.\n"
+"\n"
+"     46   Bad password entered. An error was signalled when the\n"
+"          password was entered.\n"
+"     47   Too many redirects. When following redirects, curl hit\n"
+"          the maximum amount.\n"
+"\n"
 "     XX   There will appear more error codes here in future\n"
 "          releases. The existing ones are meant to never change.\n"
 "\n"
 "BUGS\n"
-"     If you do find any (or have other suggestions), mail  Daniel\n"
-"     Stenberg <Daniel.Stenberg@haxx.se>.\n"
+"     If you do find bugs, mail them to curl-bug@haxx.se.\n"
+"\n"
 "AUTHORS / CONTRIBUTORS\n"
 "      - Daniel Stenberg <Daniel.Stenberg@haxx.se>\n"
 "      - Rafael Sagula <sagula@inf.ufrgs.br>\n"
@@ -704,6 +766,8 @@ puts (
 "      - Douglas E. Wegscheid <wegscd@whirlpool.com>\n"
 "      - Mark Butler <butlerm@xmission.com>\n"
 "      - Eric Thelin <eric@generation-i.com>\n"
+);
+ puts(
 "      - Marc Boucher <marc@mbsi.ca>\n"
 "      - Greg Onufer <Greg.Onufer@Eng.Sun.COM>\n"
 "      - Doug Kaufman <dkaufman@rahul.net>\n"
@@ -733,6 +797,14 @@ puts (
 "      - Fred Noz <FNoz@siac.com>\n"
 "      - Caolan McNamara <caolan@csn.ul.ie>\n"
 "      - Albert Chin-A-Young <china@thewrittenword.com>\n"
+"      - Stephen Kick <skick@epicrealm.com>\n"
+"      - Martin Hedenfalk <mhe@stacken.kth.se>\n"
+"      - Richard Prescott\n"
+"      - Jason S. Priebe <priebe@wral-tv.com>\n"
+"      - T. Bharath <TBharath@responsenetworks.com>\n"
+"      - Alexander Kourakos <awk@users.sourceforge.net>\n"
+"      - James Griffiths <griffiths_james@yahoo.com>\n"
+"\n"
 "WWW\n"
 "     http://curl.haxx.se\n"
 "\n"
@@ -762,8 +834,6 @@ puts (
 "  Get a gopher document from funet's gopher server:\n"
 "\n"
 "        curl gopher://gopher.funet.fi\n"
-);
- puts(
 "\n"
 "  Get a web page from a server using port 8000:\n"
 "\n"
@@ -885,9 +955,11 @@ puts (
 "\n"
 "        curl -T localfile -a ftp://ftp.upload.com/remotefile\n"
 "\n"
-"  NOTE: Curl does not support ftp upload through a proxy! The reason for this\n"
-"  is simply that proxies are seldomly configured to allow this and that no\n"
-"  author has supplied code that makes it possible!\n"
+"  Curl also supports ftp upload through a proxy, but only if the proxy is\n"
+"  configured to allow that kind of tunneling. If it does, you can run curl in\n"
+"  a fashion similar to:\n"
+"\n"
+"        curl --proxytunnel -x proxy:port -T localfile ftp.upload.com\n"
 "\n"
 " HTTP\n"
 "\n"
@@ -1004,6 +1076,8 @@ puts (
 "\n"
 "        curl -F \"file=@cooltext.txt\" -F \"yourname=Daniel\" \\\n"
 "             -F \"filedescription=Cool text file with cool text inside\" \\\n"
+);
+ puts(
 "             http://www.post.com/postit.cgi\n"
 "\n"
 "  So, to send two files in one post you can do it in two ways:\n"
@@ -1026,6 +1100,8 @@ puts (
 "\n"
 "        curl -e www.coolsite.com http://www.showme.com/\n"
 "\n"
+"  NOTE: The referer field is defined in the HTTP spec to be a full URL.\n"
+"\n"
 "USER AGENT\n"
 "\n"
 "  A HTTP request has the option to include information about the browser\n"
@@ -1058,8 +1134,6 @@ puts (
 "  headers that looks like 'Set-Cookie: <data>' where the data part then\n"
 "  typically contains a set of NAME=VALUE pairs (separated by semicolons ';'\n"
 "  like \"NAME1=VALUE1; NAME2=VALUE2;\"). The server can also specify for what\n"
-);
- puts(
 "  path the \"cookie\" should be used for (by specifying \"path=value\"), when the\n"
 "  cookie should expire (\"expire=DATE\"), for what domain to use it\n"
 "  (\"domain=NAME\") and if it should be used on secure connections only\n"
@@ -1146,17 +1220,26 @@ puts (
 "CONFIG FILE\n"
 "\n"
 "  Curl automatically tries to read the .curlrc file (or _curlrc file on win32\n"
-"  systems) from the user's home dir on startup. The config file should be\n"
-"  made up with normal command line switches. Comments can be used within the\n"
-"  file. If the first letter on a line is a '#'-letter the rest of the line\n"
-"  is treated as a comment.\n"
+"  systems) from the user's home dir on startup.\n"
+"\n"
+"  The config file could be made up with normal command line switches, but you\n"
+"  can also specify the long options without the dashes to make it more\n"
+"  readable. You can separate the options and the parameter with spaces, or\n"
+"  with = or :. Comments can be used within the file. If the first letter on a\n"
+"  line is a '#'-letter the rest of the line is treated as a comment.\n"
+"\n"
+"  If you want the parameter to contain spaces, you must inclose the entire\n"
+"  parameter within double quotes (\"). Within those quotes, you specify a\n"
+"  quote as \\\".\n"
+"\n"
+"  NOTE: You must specify options and their arguments on the same line.\n"
 "\n"
 "  Example, set default time out and proxy in a config file:\n"
 "\n"
 "        # We want a 30 minute timeout:\n"
 "        -m 1800\n"
 "        # ... and we use a proxy for all accesses:\n"
-"        -x proxy.our.domain.com:8080\n"
+"        proxy = proxy.our.domain.com:8080\n"
 "\n"
 "  White spaces ARE significant at the end of lines, but all white spaces\n"
 "  leading up to the first characters of each line are ignored.\n"
@@ -1170,14 +1253,14 @@ puts (
 "  without URL by making a config file similar to:\n"
 "\n"
 "        # default url to get\n"
-"        http://help.with.curl.com/curlhelp.html\n"
+"        url = \"http://help.with.curl.com/curlhelp.html\"\n"
 "\n"
 "  You can specify another config file to be read by using the -K/--config\n"
 "  flag. If you set config file name to \"-\" it'll read the config from stdin,\n"
 "  which can be handy if you want to hide options from being visible in process\n"
 "  tables etc:\n"
 "\n"
-"        echo \"-u user:passwd\" | curl -K - http://that.secret.site.com\n"
+"        echo \"user = user:passwd\" | curl -K - http://that.secret.site.com\n"
 "\n"
 "EXTRA HEADERS\n"
 "\n"
@@ -1228,13 +1311,14 @@ puts (
 "  connect to the client on the given (as parameters to the PORT command) IP\n"
 "  number and port.\n"
 "\n"
-"  The -P flag to curl allows for different options. Your machine may have\n"
+"  The -P flag to curl supports a few different options. Your machine may have\n"
 "  several IP-addresses and/or network interfaces and curl allows you to select\n"
 "  which of them to use. Default address can also be used:\n"
 "\n"
 "        curl -P - ftp.download.com\n"
 "\n"
-"  Download with PORT but use the IP address of our 'le0' interface:\n"
+"  Download with PORT but use the IP address of our 'le0' interface (this does\n"
+"  not work on windows):\n"
 "\n"
 "        curl -P le0 ftp.download.com\n"
 "\n"
@@ -1242,6 +1326,16 @@ puts (
 "\n"
 "        curl -P 192.168.0.10 ftp.download.com\n"
 "\n"
+"NETWORK INTERFACE\n"
+"\n"
+"  Get a web page from a server using a specified port for the interface:\n"
+"\n"
+"	curl --interface eth0:1 http://www.netscape.com/\n"
+"\n"
+"  or\n"
+"\n"
+"	curl --interface 192.168.1.10 http://www.netscape.com/\n"
+"\n"
 "HTTPS\n"
 "\n"
 "  Secure HTTP requires SSL libraries to be installed and used when curl is\n"
@@ -1253,6 +1347,8 @@ puts (
 "        curl https://www.secure-site.com\n"
 "\n"
 "  Curl is also capable of using your personal certificates to get/post files\n"
+);
+ puts(
 "  from sites that require valid certificates. The only drawback is that the\n"
 "  certificate needs to be in PEM-format. PEM is a standard and open format to\n"
 "  store certificates with, but it is not used by the most commonly used\n"
@@ -1328,8 +1424,6 @@ puts (
 "TIME CONDITIONS\n"
 "\n"
 " HTTP allows a client to specify a time condition for the document it\n"
-);
- puts(
 " requests. It is If-Modified-Since or If-Unmodified-Since. Curl allow you to\n"
 " specify them with the -z/--time-cond flag.\n"
 "\n"
@@ -1439,13 +1533,26 @@ puts (
 "\n"
 "        curl -w 'We downloaded %{size_download} bytes\\n' www.download.com\n"
 "\n"
+"KERBEROS4 FTP TRANSFER\n"
+"\n"
+"  Curl supports kerberos4 for FTP transfers. You need the kerberos package\n"
+"  installed and used at curl build time for it to be used.\n"
+"\n"
+"  First, get the krb-ticket the normal way, like with the kauth tool. Then use\n"
+"  curl in way similar to:\n"
+"\n"
+"        curl --krb4 private ftp://krb4site.com -u username:fakepwd\n"
+"\n"
+"  There's no use for a password on the -u switch, but a blank one will make\n"
+"  curl ask for one and you already entered the real password to kauth.\n"
+"\n"
 "MAILING LIST\n"
 "\n"
 "  We have an open mailing list to discuss curl, its development and things\n"
 "  relevant to this.\n"
 "\n"
-"  To subscribe, mail curl-request@contactor.se with \"subscribe <your email\n"
-"  address>\" in the body.\n"
+"  To subscribe, mail curl-request@contactor.se with \"subscribe <fill in your\n"
+"  email address>\" in the body.\n"
 "\n"
 "  To post to the list, mail curl@contactor.se.\n"
 "\n"

src/setup.h

@@ -26,9 +26,9 @@
  *
  * ------------------------------------------------------------
  * Main author:
- * - Daniel Stenberg <Daniel.Stenberg@haxx.nu>
+ * - Daniel Stenberg <daniel@haxx.se>
  *
- * 	http://curl.haxx.nu
+ * 	http://curl.haxx.se
  *
  * $Source$
  * $Revision$

src/urlglob.c

@@ -24,9 +24,9 @@
  *
  * ------------------------------------------------------------
  * Main author:
- * - Daniel Stenberg <Daniel.Stenberg@haxx.nu>
+ * - Daniel Stenberg <daniel@haxx.se>
  *
- * 	http://curl.haxx.nu
+ * 	http://curl.haxx.se
  *
  * $Source$
  * $Revision$
@@ -45,7 +45,11 @@
 #include <curl/curl.h>
 #include "urlglob.h"
 
-char glob_buffer[URL_MAX_LENGTH];
+#ifdef MALLOCDEBUG
+#include "../lib/memdebug.h"
+#endif
+
+char *glob_buffer;
 URLGlob *glob_expand;
 
 int glob_word(char*, int);
@@ -206,10 +210,13 @@ int glob_word(char *pattern, int pos) {
 
 int glob_url(URLGlob** glob, char* url, int *urlnum)
 {
-  if (strlen(url)>URL_MAX_LENGTH) {
-    printf("Illegally sized URL\n");
-    return CURLE_URL_MALFORMAT;
-  }
+  /*
+   * We can deal with any-size, just make a buffer with the same length
+   * as the specified URL!
+   */
+  glob_buffer=(char *)malloc(strlen(url)+1);
+  if(NULL == glob_buffer)
+    return CURLE_OUT_OF_MEMORY;
 
   glob_expand = (URLGlob*)malloc(sizeof(URLGlob));
   glob_expand->size = 0;
@@ -218,6 +225,25 @@ int glob_url(URLGlob** glob, char* url, int *urlnum)
   return CURLE_OK;
 }
 
+void glob_cleanup(URLGlob* glob) {
+  int i, elem;
+
+  for (i = glob->size - 1; i >= 0; --i) {
+    if (!(i & 1)) {	/* even indexes contain literals */
+      free(glob->literal[i/2]);
+    } else {		/* odd indexes contain sets or ranges */
+      if (glob->pattern[i/2].type == UPTSet) {
+	for (elem = glob->pattern[i/2].content.Set.size - 1; elem >= 0; --elem) {
+	  free(glob->pattern[i/2].content.Set.elements[elem]);
+	}
+	free(glob->pattern[i/2].content.Set.elements);
+      }
+    }
+  }
+  free(glob);
+  free(glob_buffer);
+}
+
 char *next_url(URLGlob *glob)
 {
   static int beenhere = 0;

src/urlglob.h

@@ -26,9 +26,9 @@
  *
  * ------------------------------------------------------------
  * Main author:
- * - Daniel Stenberg <Daniel.Stenberg@haxx.nu>
+ * - Daniel Stenberg <daniel@haxx.se>
  *
- * 	http://curl.haxx.nu
+ * 	http://curl.haxx.se
  *
  * $Source$
  * $Revision$
@@ -70,5 +70,6 @@ typedef struct {
 int glob_url(URLGlob**, char*, int *);
 char* next_url(URLGlob*);
 char* match_url(char*, URLGlob); 
+void glob_cleanup(URLGlob* glob);
 
 #endif

src/version.h

@@ -1,3 +1,3 @@
 #define CURL_NAME "curl"
-#define CURL_VERSION "7.2"
+#define CURL_VERSION "7.5"
 #define CURL_ID CURL_NAME " " CURL_VERSION " (" OS ") "

src/writeout.c

@@ -38,12 +38,16 @@
  * ------------------------------------------------------------
  ****************************************************************************/
 
-#include "setup.h"
-
 #include <stdio.h>
 #include <string.h>
 
-#include "strequal.h"
+#include <curl/curl.h>
+#include <curl/types.h>
+#include <curl/easy.h>
+
+#define _MPRINTF_REPLACE /* we want curl-functions instead of native ones */
+#include <curl/mprintf.h>
+
 #include "writeout.h"
 
 typedef enum {
@@ -57,6 +61,8 @@ typedef enum {
   VAR_SPEED_DOWNLOAD,
   VAR_SPEED_UPLOAD,
   VAR_HTTP_CODE,
+  VAR_HEADER_SIZE,
+  VAR_REQUEST_SIZE,
   VAR_EFFECTIVE_URL,
   VAR_NUM_OF_VARS /* must be the last */
 } replaceid;
@@ -74,6 +80,8 @@ static struct variable replacements[]={
   {"time_namelookup", VAR_NAMELOOKUP_TIME},
   {"time_connect", VAR_CONNECT_TIME},
   {"time_pretransfer", VAR_PRETRANSFER_TIME},
+  {"size_header", VAR_HEADER_SIZE},
+  {"size_request", VAR_REQUEST_SIZE},
   {"size_download", VAR_SIZE_DOWNLOAD},
   {"size_upload", VAR_SIZE_UPLOAD},
   {"speed_download", VAR_SPEED_DOWNLOAD},
@@ -81,10 +89,14 @@ static struct variable replacements[]={
   {NULL}
 };
 
-void WriteOut(struct UrlData *data)
+void ourWriteOut(CURL *curl, char *writeinfo)
 {
   FILE *stream = stdout;
-  char *ptr=data->writeinfo;
+  char *ptr=writeinfo;
+  char *stringp;
+  long longinfo;
+  double doubleinfo;
+
   while(*ptr) {
     if('%' == *ptr) {
       if('%' == ptr[1]) {
@@ -105,37 +117,67 @@ void WriteOut(struct UrlData *data)
             if(strequal(ptr, replacements[i].name)) {
               switch(replacements[i].id) {
               case VAR_EFFECTIVE_URL:
-                fprintf(stream, "%s", data->url?data->url:"");
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_EFFECTIVE_URL, &stringp))
+                  fputs(stringp, stream);
                 break;
               case VAR_HTTP_CODE:
-                fprintf(stream, "%03d", data->progress.httpcode);
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_HTTP_CODE, &longinfo))
+                  fprintf(stream, "%03d", longinfo);
+                break;
+              case VAR_HEADER_SIZE:
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_HEADER_SIZE, &longinfo))
+                  fprintf(stream, "%d", longinfo);
+                break;
+              case VAR_REQUEST_SIZE:
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_REQUEST_SIZE, &longinfo))
+                  fprintf(stream, "%d", longinfo);
                 break;
               case VAR_TOTAL_TIME:
-                fprintf(stream, "%.3f", data->progress.timespent);
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_TOTAL_TIME, &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
                 break;
               case VAR_NAMELOOKUP_TIME:
-                fprintf(stream, "%.3f", tvdiff(data->progress.t_nslookup,
-                                               data->progress.start));
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_NAMELOOKUP_TIME,
+                                     &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
                 break;
               case VAR_CONNECT_TIME:
-                fprintf(stream, "%.3f", tvdiff(data->progress.t_connect,
-                                               data->progress.start));
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_CONNECT_TIME, &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
                 break;
               case VAR_PRETRANSFER_TIME:
-                fprintf(stream, "%.3f", tvdiff(data->progress.t_pretransfer,
-                                               data->progress.start));
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_PRETRANSFER_TIME, &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
                 break;
               case VAR_SIZE_UPLOAD:
-                fprintf(stream, "%.0f", data->progress.uploaded);
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_SIZE_UPLOAD, &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
                 break;
               case VAR_SIZE_DOWNLOAD:
-                fprintf(stream, "%.0f", data->progress.downloaded);
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_SIZE_DOWNLOAD, &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
                 break;
               case VAR_SPEED_DOWNLOAD:
-                fprintf(stream, "%.2f", data->progress.dlspeed);
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_SPEED_DOWNLOAD, &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
                 break;
               case VAR_SPEED_UPLOAD:
-                fprintf(stream, "%.2f", data->progress.ulspeed);
+                if(CURLE_OK ==
+                   curl_easy_getinfo(curl, CURLINFO_SPEED_UPLOAD, &doubleinfo))
+                  fprintf(stream, "%.3f", doubleinfo);
+                break;
+              default:
                 break;
               }
               break;

src/writeout.h

@@ -40,8 +40,6 @@
  * ------------------------------------------------------------
  ****************************************************************************/
 
-#include "urldata.h"
-
-void WriteOut(struct UrlData *data);
+void ourWriteOut(CURL *curl, char *out);
 
 #endif

tests/Makefile.am

@@ -0,0 +1,17 @@
+
+all:
+install:
+
+curl:
+	@(cd ..; make)
+
+test:
+	perl runtests.pl
+
+quiet-test:
+	perl runtests.pl -s -a
+
+clean:
+	rm -rf log
+	find . -name "*~" | xargs rm -f
+

tests/README

@@ -0,0 +1,69 @@
+                                  _   _ ____  _     
+                              ___| | | |  _ \| |    
+                             / __| | | | |_) | |    
+                            | (__| |_| |  _ <| |___ 
+                             \___|\___/|_| \_\_____|
+
+The cURL Test Suite
+
+Requires:
+  perl
+
+Run:
+  'make test'. This invokes the 'runtests.pl' perl script. Edit the top
+  variables of that script in case you have some specific needs.
+
+  The script breaks on the first test that doesn't do OK. Use -a to prevent
+  the script to abort on the first error. Run the script with -v for more
+  verbose output.
+
+  Use -s fort shorter output, or pass a string with test numbers to run
+  specific tests only (like ./runtests.pl "3 4" to test 3 and 4 only)
+
+Memory:
+  The test script will check that all allocated memory is freed properly IF
+  curl has been built with the MALLOCDEBUG define set. The script will
+  automatically detect if that is the case, and it will use the ../memanalyze
+  script to analyze the memory debugging output.
+
+Logs:
+  All logs are generated in the logs/ subdirctory (it is emtpied first
+  in the runtests.sh script)
+
+Data:
+  All test-data are put in the data/ subdirctory.
+
+  For each tests there exist a few files, all with their own separate and
+  special purpose. Replace N with the test number:
+
+    nameN.txt: test description as displayed when run
+
+    commandN.txt: command line options for this test
+
+    protN.txt: the full dump of the protocol communication that curl is
+               expected to use when performing this test
+
+    replyN.txt: the full dump the server should reply to curl for this test.
+               If the final result that curl should've got is not in this
+               file, you can instead name the file replyN0001.txt. This enables
+               you to fiddle more. ;-)
+
+    stdoutN.txt: if this file is present, curl's stdout is compared against
+               this file to see that they're identical. If this is present,
+               curl will not be run with -o but instead all output is compared
+               against this file!
+
+    errorN.txt: if this file is present, it should contain the error number
+               curl is supposed to return when this test is run.
+
+    uploadN.txt: if this file is present, it should contain the same data as
+               the log/upload.N does, after a curl upload has been performed.
+
+    ftpdN.txt: this file may contain instructions how to modify the behaviour
+               of the ftp server. It uses a simple syntax that is left to
+               describe here!
+
+FIX:
+
+  * Make httpserver.pl work when we PUT without Content-Length:
+

tests/data/command1.txt

@@ -0,0 +1 @@
+http://%HOSTIP:%HOSTPORT/1

tests/data/command10.txt

@@ -0,0 +1,3 @@
+http://%HOSTIP:%HOSTPORT/we/want/10 -T data/command10.txt
+
+

tests/data/command100.txt

@@ -0,0 +1 @@
+ftp://%HOSTIP:%FTPPORT/

tests/data/command101.txt

@@ -0,0 +1,3 @@
+ftp://%HOSTIP:%FTPPORT/ -P %HOSTIP
+
+

tests/data/command102.txt

@@ -0,0 +1 @@
+ftp://%HOSTIP:%FTPPORT/102