Commit Graph

619 Commits

Author SHA1 Message Date
Thomas Köppe
c4312f8367 Added some missing inclusions (#1274) 2017-10-06 19:47:53 -05:00
Alex Fabijanic
b3ef37af1f initialize response address 2017-10-06 19:45:51 -05:00
Alex Fabijanic
f9eac16631 Add const modificator to SocketAddress argument in ICMPClient methods #1267 2017-10-06 19:38:29 -05:00
Alex Fabijanic
e0a5ebc06e don't inline methods containing private symbols #915 2017-10-06 19:20:56 -05:00
Alex Fabijanic
14076f75b4 Internet connection is not available on build servers, making tests fail #1365 2017-10-06 18:22:54 -05:00
Alex Fabijanic
71531d1a75 Move to standard integer types #1147 2017-10-05 17:03:11 -05:00
Alex Fabijanic
426368c016 Move to standard integer types #1147 2017-10-05 14:36:02 -05:00
Alex Fabijanic
f68e0174b6 ICMPSocket does not check reply address #1921 2017-10-05 14:01:41 -05:00
micheleselea
c4f0727bc7 Fix in close and open (#1911)
* Fix in close and open

close(): fix possible throw on broken socket that could lead in CLOSE_WAIT leak (especially under linux OS)
open(): fix open function to deal with control socket created from outside and login without user

* Fix FTPSClientSession ctor

* fix: check if controlsocket is valid

* removed read server reply with no username

If we read server reply with no username, we can end up with a throw (if server reply is negative) and the testLoginFailed1 fail
2017-09-30 23:24:10 -05:00
Alex Fabijanic
03e97dd0f7 Run ping with custom data size #1912 2017-09-28 11:18:45 -05:00
Aleksandar Fabijanic
238306a6ed C++11 support #1793 (#1795)
* * C++14 detection
* C++14 RWLock (disabled: std::shared_timed_mutex has separate read and write unlock)

* * C++11 Event - disable because
std::condition_variable nas no built-in concept of reset
std::condition_variable may be unblocked spuriously, must handle?

* * Fix std::condition_variable handling

* * C++11 semaphore

* * Use Semaphore max paramater

* * Implement manual reset Event and re-enable
* Disable std semaphore, C++ doesn't have native semaphore, better to use native ones for now

* * C++11 Thread (in progress)

* * Fix auto reset event

* * std::*mutex with timeout doesn't guarantee that the timeout will be respected
* Thread in unix has special member for signal handling

* * Workaround for AppleClang lack of thread_local

* * C++11 format

* * C++11 Logger

* * Fix Poco::format

* * std::function event delegate

* * Fix StdFunctionDelegate for GCC

* * C++11 move semantics

* * Fix format test

* * Add cpp11_changes files

* * Auto detect compiler C++11/14 support
* Option to force disable C++11/14

* * GCC/Clang need to enable C++11/14 support via command line switches

* * Fixes for VS2013 C++11 support

* * Makefile C++11/14 compiler detection (in progress)

* * Simple GCC version detection

* AppleClang version detection

* Fix gcc c++11 build script
Update c++11 status document

* Use predefined constant instead of --version

* Fix CC & CXX to gcc 4.8 for all matrix rows.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Restore apt-get update -qq

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Display g++ version and full g++ command line

* Fix typo

* Use command line/environment variables for CC & CXX

* Display GCCVERSION

* Use $(CXX) for setting GCCVERSION

* Do not use <cstdint>, otherwise NumberFormatter wo'nt compile

* restore silent mode on make

* Use predefined constants __clang_major__ and __clang_minor__

* Reverse clang++ options

* Add c++11/14 detection

* -std=c++14 is not valid for clang 3.4

* Run tests for both gcc 4.6 & gcc 4.8

* Restore Apple clang setup

* Align setup of std=c++11 from CLANGVERSION with check of clang version in Platform_POSIX.h

On Linux ubuntu at Travis, clang 3.4 does not support std=c++14

* Add OSX run with clang

* Add forgetted Linux guard on install

* Display clang's predefined constants on OSX

* Comment out Apple temporarly

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* export CC & CXX for OSX

* Test when clang is from Apple

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Remove typo

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Align test with comment

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Add Cygwin-clang

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Install CppUnit for running the unit tests on OSX. 

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Looking for libCppUnit.1.dylib

* Install libCppUnit*.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Exclude Redis for OSX. To be restored later on.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Remove some display lines

* Reenable tests

* Add run tests on cmake build with ctest -VV

* Split runtests.sh into ignored.sh and excluded.sh. ignored.sh will be
used for excluding tests from ctest.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Add ignored.sh to exclude CPPUNIT_IGNORE test on ctest

* Set proper path

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Exclude Data* from CTest. Run tests on ARM

* Add POCO_BASE for tests using test files under $POCO_BASE/component/testsuite

* Replace MB by KB otherwise Travis & AppVeyor are timing out.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Specialize runtests.sh & excluded.sh according to OSX or Linux

* Split OSX/Linux runtests.sh

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Removed invalid comment

* Make runtests.sh executable

* c++11 atomic<> default constructor is never initialized

* Initialize to 0 in any case.

* Comment out set -ev for now

* Comment out set -ev for now

* reset verbose mode

* reset verbose mode

*  catch SIGPIPE as returnfrom testrunner to avoid failure in networking tests

* Use source for export ignored tests to avoid the permission denied problem

* chmod 775

* Apply patches from @RangeIReale

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Add cpp11-appleclang

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Initialize AtomicCounter to 0 since the c++11 default constructor don't
do it.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* avoid SIGPIPE on send

* fix WbeSocketServer

* Remove SIGPIPE workaround

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Fix WebSocketServer

* Restore set -ev

* Restore set -ev

* Fix WebSocketServer responding loop

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Cleanup the exit test.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Add important comment.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Copy also cpp11-* so that mkrelease works!

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Avoid PocoDoc preprocessisng error.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* * Sync script and Platform_POSIX.h compiler versions

* VS2013 & VS2015 fixups.

Signed-off-by: Francis ANDRE <zosrothko@orange.fr>

* Backport to c++03 so that Travis and Appveyor be ok

Signed-off-by: Francis ANDRE <zosrothko@orange.fr>

* Add Copyright (c) 2016, Applied Informatics Software Engineering GmbH.
and Contributors.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* VS2013 fixup

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Backport to c++03

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* VS2013 fixup

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Backport to c++03

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* VS2013 fixup

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* BAckport c++03

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* OSX: clang++ rejects return (*static_cast<Derived*>(this));
cannot cast protected base class 'cpp_ex4::StoplightContext<cpp_ex4::Stoplight>' to 'cpp_ex4::Stoplight'


Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* BAckport to c++03

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Add generated file for now

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* deleted as source file name in lowercase.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Rename stoplight_sm.h to Stoplight_sm.h

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Rename stoplight_sm.h to Stoplight_sm.h

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Rename stoplight_sm.h to Stoplight_sm.h

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Put cmake and ctest on the DOS PATH

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Cygwin does not support timed_mutex

* Put installed CMake /CTest on the DOS PATH.

* Repackaging

* Build VS2013 & VS2014 first.

* Add std::string Foundation_API format(const std::string& fmt, const Any& value);

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Restore INLINES macros when compiling with c++03

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Rename stoplight_sm.h to Stoplight_sm.h

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* VS2013 fixup.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Baclport to c++03

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* cpp_ex4::StoplightContext<cpp_ex4::Stoplight> is an inaccessible base of ˜cpp_ex4::Stoplight

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Undef _XOPEN_SOURCE under Cygwin otherwise ulong is undefined.
usr/include/mysql/my_global.h:997:9: error: 'ulong' does not name a type
 typedef ulong nesting_map;  /* Used for flags of nesting constructs */
         ^
In file included from src/SQLExecutor.cpp:38:0:

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Fix merge failure. Set -ev option on shell.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Removed

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Undefine _XOPEN_SOURCE so that typedef unsigned long ulong be defined.

Signed-off-by: FrancisANDRE <zosrothko@orange.fr>

* Remove __*VISIBLE defines as GNU forbids their usage outside the system includes

Signed-off-by: Francis ANDRE <zosrothko@orange.fr>

* Use c++03 as last c++ version.

Signed-off-by: Francis ANDRE <zosrothko@orange.fr>

* MySQL needs ulong which are not visible when compiling with
c++11 and above.

Signed-off-by: Francis ANDRE <zosrothko@orange.fr>

* Add BUILD_CC=gcc to avoid strip issue

* Remove reference to FSM

* make ?= operator does not work with

* Remove BUILD_CC

* Remove FSM since it is now in its own branch

* Don't build Cygwin64 with std=c++14 until this issue be fixed https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77441

* enable c++11

* remove changes tracking file (moved to project)

* C++11: fixes for Travis CI (#1798)

* Use c++11 compliant compiler: g++ 5 & clang 3.5.0

* Fix typo on IntType

* Use Linux-clang config instead of Linux

* Check arm-linux-gnueabi-g++ version. Dump compiler constants

* Fix missing namespace qualifier

* Ignore ltdl for Cygwin platform

* Fix typo

* Commented out arm-linux-gnueabi-g++ since the version 4.7.0 which does
not support c++11 standard.

* Restore logic as in develop branch

* Add CXXFLAGS for launching the CXX compilation. CXXFLAGS defaulted to
-std=c++11

* Replace auto_ptr by unique_ptr

* Comment out failing tests until be fixed

* C++11: Fix AppVeyor CI yaml (#1812)

* C++11 support #1793: It's time. g++4.9, clang3.3 and VS2015 will become
minimum compiler versions.

* Comment out image with VS2017 since it crashes.

* New release of OpenSSL: 1_1_0f

* Refactor the download steps of OpenSSL

* C++11 support: Add VS2017 Win32 & x64 (#1816)

* Add VS2017.

* Use VSSetup to get VS150COMNTOOLS

* Fix VS150COMNTOOLS value

* Ignore Install-Module

* Install VSSetUp PS module.

* Fix typo missing $

* Split InstallPath againt "="

* Adjust VS150COMMONTOOLS path.

* Exit failed is VS commontools is undefined.

* Get the InstalledPath property value

* Replace mysql-5.7.17 by mysql-5.7.18

* Replace mysql-5.7.17 by mysql-5.7.18

* c++11 thread priority and affinity, and more (#1811)

* * c++11 thread priority and affinity
* AtomicCounter wasn't using std::atomic
* Buffer wasn't setting all fields on move
* Poco::Event is not directly available in c++11, keep using the native versions
* UnWindows.h define NOMINMAX to reduce conflicts
* Remove "-static" from Linux GCC parameters, this conflicts with c++11 classes (https://stackoverflow.com/questions/7090623/c0x-thread-static-linking-problem)
* Link "-lrt" before "-lpthread" on Linux, otherwise won't build on GCC 5+

* * Fixed std::thread POSIX handle

* * Fix OSX compilation

* raise cmake min version to 3.2.0

* Remove CMake jobs since it is already tested on Travis (#1832)

* Remove CMake jobs since it is already tested on Travis

* Add clang-3.7

* Ubuntu Trusty LLVM repo for gettign clang versions.

* Remove 3.3 & 3.4

* Add clang 3.8, clang 3.9 & clang 4.0

* Add CXXFLAGS=-stdlib=libc++

* use libstdc++ instead of libc++

* Remove for now -g compiler option to avoid this spurious error on clang
error: debug information for auto is not yet supported

* Use -ftemplate-backtrace-limit=0 to dump the recursive template stack.

* Cleanup

* Use good gcc version

* Fix gcc-5 & g++-5 name

* Use libc++ instead of libstdc++

* Restore installation of gcc-5

* Readd agt-get update for gcc-5/g++-5

* Reswitch from libc++ to libstdc++

* Increase template evaluation stack size up to 340 to avoid a stack
overflow in parsing 
TupleType t = std::make_tuple(0,1,2,3,4,5,6,7,8,9,10,11,12,13,14...

* Add clang 3.4, 3.5, 3.6

* Adjust template evaluation stack for clang  3.4, 3.5, 3.6, 3.7

* Update notifications

* Add gcc-4.9

* Backport Travis CXXFLAGS to config/Linux-clang

* Remove CXXFLAGS from clang tests

* Test g++ first, then clang++ and then arm

* remove all non-c++11 code and POCO_ENABLE_CPP11 (default now)

* remove old file

* windows c++11 fixes

* fix conversion warning

* C++11: Use Mutex_POSIX for Cygwin since it does not support std::timed_mutex (#1864)

* Use Mutex_POSIX for Cygwin since it does not support std::timed_mutex

* Use Mutex_POSIX for Cygwin since it does not support std::timed_mutex

* Add Mutext_POSIX.cpp

* Json/parser (#1858)

* move parser implementation in separate file

* parser replacement, first attempt (wip)

* making tests pass (mostly done, 3 to go)

* add missing headers

* honor no-null-byte-in-string setting

* few more fixes

* add -std=c99

* fix cmake JSON build and formatting

* accept utf-8, fix tests

* skip calling isdigit()

* #740, #1860, license update, a warning fix

* var emtpy()->clear(); json array and object tidy up

* vs 2015 32-bit project update

* update VS2017 projects

* remove ProGen and all VS files older than 2015

* fixed GH #1865: AbstractEvent::hasDelegates() is not thread-safe

* merge develop

* add missing files

* add VS generated dir to gitignore

* add WEC2013.cmake

* fix exception message

* fix VS2017 ODBC project filter

* integrate PageCompiler changes from poco-1.7.9

* vs2015 openssl build

* added POCO_DEPRECATED macro

* added POCO_NO_DEPRECATED to disarm POCO_DEPRECATED macro

* PageCompiler: support <%@ include file="<path>" %> syntax for includes.

* remove '$Id$' headers

* remove '$Id$' headers

* remove old VMS and VxWorks build support

* Poco::NamedMutex and Poco::NamedEvent (System V Semaphores implementation): files are now opened with O_RDONLY | O_CREAT instead of O_WRONLY | O_CREAT, allowing sharing between different users. Furthermore, ftok() is called with 'p' as project ID argument.

* C++11 (#1887)

* Use Mutex_POSIX for Cygwin since it does not support std::timed_mutex

* Use Mutex_POSIX for Cygwin since it does not support std::timed_mutex

* Add Mutext_POSIX.cpp

* Remove clang 3.4, 3.5, 3.6, 3.7, 3.8, 3.9 since LLVM does not provide
anymore their packages for Ubuntu Trusy

* Remove ubuntu-toolchain-r/test/ubuntu since it is already installed

* Comment our the llvl-toolchain-trusty since its checksum is buggy
See https://bugs.llvm.org/show_bug.cgi?id=34572

* Display g++ version (#1889)

* merge change from develop|

* Support for PKCS#12 (#1876)

* rebuild openssl binaries; warning and stlye fixes

* fix vs2015 projects and openssl linking

* add PKCS12Container

* remove comments

* style

* add ECKey* and CryptoException

* EC key, unify RSA and EC under same inheritance, add constructor from PKCS12, couple of EC key tests

* simplify EVPPKey, ad EC tests

* EVPPKey test and fixes

* fix linux build

* PKCS12 tests and fixes

* linux build, fix crash

* fix leaks

* uncomment ifstream tests, some minor fixes

* fix stream tests and some tidy-up

* remove $Id

* add ECDSA

* update makefile

* align PKCS12 constructors signatures with X509

* EVPPKey EC curve name constructor

* ECDSA fixes and tests

* linux build, wrap tests in try/catch to get full exception message

* style

* update VS projects

* remove openssl, modify VS projects for git submodule directories

* add openssl submodule

* add _CRT_SECURE_NO_WARNINGS

* port crypto dev fixes from 1.8

* lock whole Event:set()

* fix openssl include path

* Add <memory> for std::unique_ptr (#1891)

* Missing include <iostream> (#1893)

* add file/stream load/save capabilities to EVPPKey

* add EVPTest

* add #include <typeinfo>

* fix posix compile

* update samples VS projects

* fix g++ test compile error
2017-09-20 22:13:19 -05:00
Günter Obiltschnig
7896a1b7dd remove old VMS and VxWorks build support 2017-09-09 11:07:39 +02:00
Günter Obiltschnig
06373b5d0c remove '$Id$' headers 2017-09-09 10:50:32 +02:00
Alex Fabijanic
c81de1d34b fix FTPS win build, license, header guards; fix few warnings 2017-08-31 12:20:49 -05:00
micheleselea
dd573b98d8 Add FTPS (FTP over explicit SSL) implementation (#1866)
* FTPSClientSession

Add class to support FTPS (explicit FTP over SSL). Testcase done but just  TestSuite_x64_vs140.vcxproj modified

* FTPS project file

Added FTPS files to v140 project

* FTPS makefile

Added FTPS file to makefile

* testsuit compile

Fix compile of testsuite

* comment misspelled

checked misspelled comment, and re-add WebSocket testsuite

* remove warning

reorder constructor inizializzations in order to remove gcc -Wreorder warning

* testsuite compile

Correct Makefile in testsuite

* Makefile testsuite

add DialogServer to testsuite makefile

* test build

fix build of testsuite

* add FTPSStreamFactory

* vs140 32bit

* build and test for vs150
2017-08-31 07:30:10 -05:00
sebastien-guay
a9982e68ba "SocketReactor::addEventHandler" and "SocketReactor::removeEventHandler" must protect the access to "NotifierPtr pNotifier" (#1709)
Those two methods already use FastMutex::ScopedLock lock(_mutex), but
the scope is not large enough to protect "NotifierPtr pNotifier" that is
accessed by calling SocketNotifier::addObserver() and
SocketNotifier::removeObserver().

It is mentioned in SocketReator.h that it is safe to call
addEventHandler() and removeEventHandler() from another thread while the
SocketReactor is running. My current use of the SocketReactor
encountered an issue where the SocketNotifier::_events has been
corrupted by a concurent write access done by
SocketReactor::addEventHandler() and
SocketReactor::removeEventHandler().
The call stack show that the SocketReactor::addEventHandler is stuck in
a while loop in gcc/libstdc++/tree.cc Rb_tree_insert_and_rebalance()

I clearly see in my logs that it happened while my
SocketConnector::unregisterConnector() and
SocketConnector::registerConnector() were called by two different
threads.

#0 0x00a80a7b in std::_Rb_tree_insert_and_rebalance () from
/usr/lib/libstdc++.so.6
#1 0x06ccb430 in std::_Rb_tree<Poco::Net::SocketNotification*,
Poco::Net::SocketNotification*,
std::_IdentityPoco::Net::SocketNotification*,
std::lessPoco::Net::SocketNotification*,
std::allocatorPoco::Net::SocketNotification* >::_M_insert
(this=0xac75dc90, __x=0x0, __p=0xac7ed0c8, __v=@0xb5fb0c40) at
/usr/lib/gcc/i386-redhat-linux/3.4.6/../../../../include/c++/3.4.6/bits/stl_tree.h:816
#2 0x06ccb15d in std::_Rb_tree<Poco::Net::SocketNotification*,
Poco::Net::SocketNotification*,
std::_IdentityPoco::Net::SocketNotification*,
std::lessPoco::Net::SocketNotification*,
std::allocatorPoco::Net::SocketNotification* >::insert_equal
(this=0xac75dc90, __v=@0xb5fb0c40) at
/usr/lib/gcc/i386-redhat-linux/3.4.6/../../../../include/c++/3.4.6/bits/stl_tree.h:858
#3 0x06ccad86 in std::multiset<Poco::Net::SocketNotification*,
std::lessPoco::Net::SocketNotification*,
std::allocatorPoco::Net::SocketNotification* >::insert (this=0xac75dc90,
__x=@0xb5fb0c40)
at
/usr/lib/gcc/i386-redhat-linux/3.4.6/../../../../include/c++/3.4.6/bits/stl_multiset.h:306
#4 0x06cca72b in Poco::Net::SocketNotifier::addObserver
(this=0xac75dc70, pReactor=0x8d27958, observer=@0xb5fb0cd0) at
src/SocketNotifier.cpp:45
#5 0x06cd060f in Poco::Net::SocketReactor::addEventHandler
(this=0x8d27958, socket=@0xac7f5a74, observer=@0xb5fb0cd0) at
src/SocketReactor.cpp:178

SocketReactor::run() is already protecting the access to the
SocketNotifier.
2017-08-10 10:10:09 +02:00
TimTim
5b7d9d2b18 avoid multipart boundary conflict (#1753)
when nesting multipart, descendant parts currently may have the same boundary as the top-level multipart.
2017-08-10 10:07:19 +02:00
micheleselea
5fe775dfc3 SQlite 3.19.3 and Socket::select fix (#1825)
* SQlite 3.19.3

changelog
http://www.sqlite.org/releaselog/3_19_3.html

* epoll_event memset fix

with -std=c++14 the sizeof(eventsIn) give as result the number of array elements not the bytes number, so the memset is wrong

* fix for older gcc

on older version of gcc struct epoll_event eventsOut[epollSize] = { 0 }; does not compile
2017-07-25 14:20:19 +02:00
Yasuhiro Horimoto
fd1d113acb Fix VisualStudio warning C4245 in NetworkInterface.h (#1823) 2017-07-23 21:09:57 +02:00
Matt Tucker
4da941f869 TCPServerDispatcher: fix thread accounting leak (#1801)
* TCPServerDispatcher::run: catch errors in connection creation and handling to prevent threads from exiting without accounting for them
2017-07-10 22:37:11 +02:00
Stefan Weil
8ccee8928c Fix typos in comments, documentation and strings (#1615)
All of them were found using codespell.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
2017-07-06 00:17:38 +02:00
Alex Fabijanic
72b826651f src/Var.cpp: 2 * sanity check in wrong place ? #1784 2017-07-03 21:03:50 +02:00
Reg Arvidson
e0b64badfe Fix always true condition, possible out-of-bounds 2017-02-22 13:09:32 -08:00
Guenter Obiltschnig
f453dd145c added Visual Studio 2017 project files 2017-02-20 16:46:44 +01:00
Christopher Baker
2267feb475 Fixed position of scope parameter. Fixes #1552 2017-02-19 23:41:12 -06:00
Guenter Obiltschnig
fe8e4bfc27 fix incorrect auto_ptr usage in testsuite 2017-02-18 07:04:41 +01:00
Guenter Obiltschnig
6ecaa675cc style fixes 2017-02-17 13:43:48 +01:00
Guenter Obiltschnig
5b0df5fcd5 fixed test due to changed URI::encode() behavior 2017-02-17 13:42:50 +01:00
Guenter Obiltschnig
c1b868d642 detect malformed MIME messages earlier 2017-02-12 15:45:17 +01:00
Guenter Obiltschnig
aa803f3657 fixed GH #1570: IPv6AddressImpl::toString() returns wrong output for IPv6 address "::" 2017-02-11 19:02:01 +01:00
Guenter Obiltschnig
3eda8f40bb fix a potential DoS (through memory exhaustion) by restricting HTML form field name and value sizes 2017-01-24 13:15:19 +01:00
Guenter Obiltschnig
11845334dc fix potential DoS vulnerability (through memory exhaustion) by restricting maximum line length 2017-01-24 13:15:12 +01:00
Guenter Obiltschnig
29f259cc24 fix a potential DoS vulnerability by restricting the length of the HTTP chunk size in chunked transfer encoding 2017-01-24 13:15:04 +01:00
Guenter Obiltschnig
a44217d0a2 remove unneeded declaration introduced through cherry-pick 2017-01-12 16:12:20 +01:00
Guenter Obiltschnig
5ac7959b11 fixed GH #1212: Lost WebSocket Frames after Client Websocket Handshake is complete
Conflicts:
	Net/include/Poco/Net/HTTPServerRequestImpl.h
2017-01-12 16:03:52 +01:00
proller
cc2f754d4d Net::HTTPSession: split timeout to connection, recieve, send 2017-01-11 22:00:56 +03:00
theAirC
2a7568c737 Detection of closed websocket connection
When the websocket connection is closed, a non-empty frame (n>0) may be received with FRAME_OP_CLOSE flag bit set.
Tested with chrome 54.0.2840.99 ; after establishing the websocket connection, if the connection is not closed by javascript and the tab is closed, a frame of length n=2 is received with FRAME_OP_CLOSE flag set.
2016-12-08 22:04:15 +02:00
vm2mv
9141f38035 * Net: fixed adding content part to mail message with empty content-disposition 2016-11-22 13:53:27 +01:00
Guenter Obiltschnig
0b0c808408 rename TCPConnectionFilter to TCPServerConnectionFilter; pass a StreamSocket to accept() 2016-11-11 14:30:37 +01:00
Guenter Obiltschnig
fe59f959ef GH #1485: add TCPConnectionFilter 2016-11-11 14:13:04 +01:00
Guenter Obiltschnig
f38a1a38de more Net testsuite fixes 2016-11-09 13:04:40 +01:00
Guenter Obiltschnig
696a9de502 GH #1481, GH #1449: fix broken tests due to changed dns resolution 2016-11-09 12:22:49 +01:00
Xu Yifeng
0ad635c5d4 Add a new bind() and bind6() interface and make REUSEPORT as an option.
SO_REUSEADDR is to reuse a socket in time-wait state, this is often turned on
by server applicationis, but if a port is being used, the server could fail,
this is expected. But with SO_REUSEPORT option, it even can open a port which
is being used, this is a more error prone than SO_REUSEADDR, because we
normally don't want to share a port with irrelevant application without
notice, the behavior should not be default.
2016-11-09 17:05:50 +08:00
Guenter Obiltschnig
5df7f69b08 fixed GH #1472: HTTP(S)StreamFactory should send a User-Agent header 2016-10-26 12:59:33 +02:00
Guenter Obiltschnig
b22501f692 another fix for GH #1458 - handle case when all sockets are invalid 2016-10-18 22:25:01 +02:00
Guenter Obiltschnig
6d373825fa fixed GH #1458: always pass 1 to epoll_create() 2016-10-17 18:33:41 +02:00
Guenter Obiltschnig
8293f93f7a use std::unique_ptr instead of std::auto_ptr with C++11 or newer compilers to avoid std::auto_ptr deprecation warnings 2016-10-14 12:01:25 +02:00
Guenter Obiltschnig
6195c97a86 make port constructor explicit; no longer prefer IPv4 addresses over IPv6 addresses 2016-10-10 18:44:55 +02:00
Guenter Obiltschnig
92a0f0f0aa improved EchoServer error handling 2016-10-07 22:04:10 +02:00
Guenter Obiltschnig
835d0365d9 fixed bad disconnect handling in onSocketReadable 2016-10-07 22:03:58 +02:00