This commit was manufactured by cvs2svn to create branch

'OpenSSL_0_9_7-stable'.

doc/crypto/ASN1_STRING_length.pod

@@ -0,0 +1,81 @@
+=pod
+
+=head1 NAME
+
+ASN1_STRING_dup, ASN1_STRING_cmp, ASN1_STRING_set, ASN1_STRING_length,
+ASN1_STRING_length_set, ASN1_STRING_type, ASN1_STRING_data -
+ASN1_STRING utility functions
+
+=head1 SYNOPSIS
+
+ int ASN1_STRING_length(ASN1_STRING *x);
+ unsigned char * ASN1_STRING_data(ASN1_STRING *x);
+
+ ASN1_STRING * ASN1_STRING_dup(ASN1_STRING *a);
+
+ int ASN1_STRING_cmp(ASN1_STRING *a, ASN1_STRING *b);
+
+ int ASN1_STRING_set(ASN1_STRING *str, const void *data, int len);
+
+ int ASN1_STRING_type(ASN1_STRING *x);
+
+ int ASN1_STRING_to_UTF8(unsigned char **out, ASN1_STRING *in);
+
+=head1 DESCRIPTION
+
+These functions allow an B<ASN1_STRING> structure to be manipulated.
+
+ASN1_STRING_length() returns the length of the content of B<x>.
+
+ASN1_STRING_data() returns an internal pointer to the data of B<x>.
+Since this is an internal pointer it should B<not> be freed or
+modified in any way.
+
+ASN1_STRING_dup() returns a copy of the structure B<a>.
+
+ASN1_STRING_cmp() compares B<a> and B<b> returning 0 if the two
+are identical. The string types and content are compared.
+
+ASN1_STRING_set() sets the data of string B<str> to the buffer
+B<data> or length B<len>. The supplied data is copied. If B<len>
+is -1 then the length is determined by strlen(data).
+
+ASN1_STRING_type() returns the type of B<x>, using standard constants
+such as B<V_ASN1_OCTET_STRING>.
+
+ASN1_STRING_to_UTF8() converts the string B<in> to UTF8 format, the
+converted data is allocated in a buffer in B<*out>. The length of
+B<out> is returned or a negative error code. The buffer B<*out>
+should be free using OPENSSL_free().
+
+=head1 NOTES
+
+Almost all ASN1 types in OpenSSL are represented as an B<ASN1_STRING>
+structure. Other types such as B<ASN1_OCTET_STRING> are simply typedefed
+to B<ASN1_STRING> and the functions call the B<ASN1_STRING> equivalents.
+B<ASN1_STRING> is also used for some B<CHOICE> types which consist
+entirely of primitive string types such as B<DirectoryString> and
+B<Time>.
+
+These functions should B<not> be used to examine or modify B<ASN1_INTEGER>
+or B<ASN1_ENUMERATED> types: the relevant B<INTEGER> or B<ENUMERATED>
+utility functions should be used instead.
+
+In general it cannot be assumed that the data returned by ASN1_STRING_data()
+is null terminated or does not contain embedded nulls. The actual format
+of the data will depend on the actual string type itself: for example
+for and IA5String the data will be ASCII, for a BMPString two bytes per
+character in big endian format, UTF8String will be in UTF8 format.
+
+Similar care should be take to ensure the data is in the correct format
+when calling ASN1_STRING_set().
+
+=head1 RETURN VALUES
+
+=head1 SEE ALSO
+
+L<ERR_get_error(3)|ERR_get_error(3)>
+
+=head1 HISTORY
+
+=cut

doc/crypto/ASN1_STRING_new.pod

@@ -0,0 +1,44 @@
+=pod
+
+=head1 NAME
+
+ASN1_STRING_new, ASN1_STRING_type_new, ASN1_STRING_free -
+ASN1_STRING allocation functions
+
+=head1 SYNOPSIS
+
+ ASN1_STRING * ASN1_STRING_new(void);
+ ASN1_STRING * ASN1_STRING_type_new(int type);
+ void ASN1_STRING_free(ASN1_STRING *a);
+
+=head1 DESCRIPTION
+
+ASN1_STRING_new() returns an allocated B<ASN1_STRING> structure. Its type
+is undefined.
+
+ASN1_STRING_type_new() returns an allocated B<ASN1_STRING> structure of
+type B<type>.
+
+ASN1_STRING_free() frees up B<a>.
+
+=head1 NOTES
+
+Other string types call the B<ASN1_STRING> functions. For example
+ASN1_OCTET_STRING_new() calls ASN1_STRING_type(V_ASN1_OCTET_STRING).
+
+=head1 RETURN VALUES
+
+ASN1_STRING_new() and ASN1_STRING_type_new() return a valid
+ASN1_STRING structure or B<NULL> if an error occurred.
+
+ASN1_STRING_free() does not return a value.
+
+=head1 SEE ALSO
+
+L<ERR_get_error(3)|ERR_get_error(3)>
+
+=head1 HISTORY
+
+TBA
+
+=cut

doc/crypto/ASN1_STRING_print_ex.pod

@@ -0,0 +1,96 @@
+=pod
+
+=head1 NAME
+
+ASN1_STRING_print_ex, ASN1_STRING_print_ex_fp - ASN1_STRING output routines.
+
+=head1 SYNOPSIS
+
+ #include <openssl/asn1.h>
+
+ int ASN1_STRING_print_ex(BIO *out, ASN1_STRING *str, unsigned long flags);
+ int ASN1_STRING_print_ex_fp(FILE *fp, ASN1_STRING *str, unsigned long flags);
+ int ASN1_STRING_print(BIO *out, ASN1_STRING *str);
+
+
+=head1 DESCRIPTION
+
+These functions output an B<ASN1_STRING> structure. B<ASN1_STRING> is used to
+represent all the ASN1 string types.
+
+ASN1_STRING_print_ex() outputs B<str> to B<out>, the format is determined by
+the options B<flags>. ASN1_STRING_print_ex_fp() is identical except it outputs
+to B<fp> instead.
+
+ASN1_STRING_print() prints B<str> to B<out> but using a different format to
+ASN1_STRING_print_ex(). It replaces unprintable characters (other than CR, LF)
+with '.'.
+
+=head1 NOTES
+
+ASN1_STRING_print() is a legacy function which should be avoided in new applications.
+
+Although there are a large number of options frequently B<ASN1_STRFLAGS_RFC2253> is 
+suitable, or on UTF8 terminals B<ASN1_STRFLAGS_RFC2253 & ~ASN1_STRFLAGS_ESC_MSB>.
+
+The complete set of supported options for B<flags> is listed below.
+
+Various characters can be escaped. If B<ASN1_STRFLGS_ESC_2253> is set the characters
+determined by RFC2253 are escaped. If B<ASN1_STRFLGS_ESC_CTRL> is set control
+characters are escaped. If B<ASN1_STRFLGS_ESC_MSB> is set characters with the
+MSB set are escaped: this option should B<not> be used if the terminal correctly
+interprets UTF8 sequences.
+
+Escaping takes several forms.
+
+If the character being escaped is a 16 bit character then the form "\WXXXX" is used
+using exactly four characters for the hex representation. If it is 32 bits then
+"\UXXXXXXXX" is used using eight characters of its hex representation. These forms
+will only be used if UTF8 conversion is not set (see below).
+
+Printable characters are normally escaped using the backslash '\' character. If
+B<ASN1_STRFLGS_ESC_QUOTE> is set then the whole string is instead surrounded by
+double quote characters: this is arguably more readable than the backslash
+notation. Other characters use the "\XX" using exactly two characters of the hex
+representation.
+
+If B<ASN1_STRFLGS_UTF8_CONVERT> is set then characters are converted to UTF8
+format first. If the terminal supports the display of UTF8 sequences then this
+option will correctly display multi byte characters.
+
+If B<ASN1_STRFLGS_IGNORE_TYPE> is set then the string type is not interpreted at
+all: everything is assumed to be one byte per character. This is primarily for
+debugging purposes and can result in confusing output in multi character strings.
+
+If B<ASN1_STRFLGS_SHOW_TYPE> is set then the string type itself is printed out
+before its value (for example "BMPSTRING"), this actually uses ASN1_tag2str().
+
+The content of a string instead of being interpreted can be "dumped": this just
+outputs the value of the string using the form #XXXX using hex format for each
+octet.
+
+If B<ASN1_STRFLGS_DUMP_ALL> is set then any type is dumped.
+
+Normally non character string types (such as OCTET STRING) are assumed to be
+one byte per character, if B<ASN1_STRFLAGS_DUMP_UNKNOWN> is set then they will
+be dumped instead.
+
+When a type is dumped normally just the content octets are printed, if 
+B<ASN1_STRFLGS_DUMP_DER> is set then the complete encoding is dumped
+instead (including tag and length octets).
+
+B<ASN1_STRFLGS_RFC2253> includes all the flags required by RFC2253. It is
+equivalent to:
+ ASN1_STRFLGS_ESC_2253 | ASN1_STRFLGS_ESC_CTRL | ASN1_STRFLGS_ESC_MSB |
+ ASN1_STRFLGS_UTF8_CONVERT | ASN1_STRFLGS_DUMP_UNKNOWN ASN1_STRFLGS_DUMP_DER
+
+=head1 SEE ALSO
+
+L<X509_NAME_print_ex(3)|L<X509_NAME_print_ex(3)>,
+L<ASN1_tag2str(3)|ASN1_tag2str(3)>
+
+=head1 HISTORY
+
+TBA
+
+=cut

doc/crypto/X509_NAME_ENTRY_get_object.pod

@@ -0,0 +1,72 @@
+=pod
+
+X509_NAME_ENTRY_get_object, X509_NAME_ENTRY_get_data,
+X509_NAME_ENTRY_set_object, X509_NAME_ENTRY_set_data,
+X509_NAME_ENTRY_create_by_txt, X509_NAME_ENTRY_create_by_NID,
+X509_NAME_ENTRY_create_by_OBJ - X509_NAME_ENTRY utility functions
+
+=head1 NAME
+
+=head1 SYNOPSIS
+
+ASN1_OBJECT * X509_NAME_ENTRY_get_object(X509_NAME_ENTRY *ne);
+ASN1_STRING * X509_NAME_ENTRY_get_data(X509_NAME_ENTRY *ne);
+
+int X509_NAME_ENTRY_set_object(X509_NAME_ENTRY *ne, ASN1_OBJECT *obj);
+int X509_NAME_ENTRY_set_data(X509_NAME_ENTRY *ne, int type, unsigned char *bytes, int len);
+
+X509_NAME_ENTRY *X509_NAME_ENTRY_create_by_txt(X509_NAME_ENTRY **ne, char *field, int type, unsigned char *bytes, int len);
+X509_NAME_ENTRY *X509_NAME_ENTRY_create_by_NID(X509_NAME_ENTRY **ne, int nid, int type,unsigned char *bytes, int len);
+X509_NAME_ENTRY *X509_NAME_ENTRY_create_by_OBJ(X509_NAME_ENTRY **ne, ASN1_OBJECT *obj, int type,unsigned char *bytes, int len);
+
+=head1 DESCRIPTION
+
+X509_NAME_ENTRY_get_object() retrieves the field name of B<ne> in
+and B<ASN1_OBJECT> structure.
+
+X509_NAME_ENTRY_get_data() retrieves the field value of B<ne> in
+and B<ASN1_STRING> structure.
+
+X509_NAME_ENTRY_set_object() sets the field name of B<ne> to B<obj>.
+
+X509_NAME_ENTRY_set_data() sets the field value of B<ne> to string type
+B<type> and value determined by B<bytes> and B<len>.
+
+X509_NAME_ENTRY_create_by_txt(), X509_NAME_ENTRY_create_by_NID()
+and X509_NAME_ENTRY_create_by_OBJ() create and return an 
+B<X509_NAME_ENTRY> structure.
+
+=head1 NOTES
+
+X509_NAME_ENTRY_get_object() and X509_NAME_ENTRY_get_data() can be
+used to examine an B<X509_NAME_ENTRY> function as returned by 
+X509_NAME_get_entry() for example.
+
+X509_NAME_ENTRY_create_by_txt(), X509_NAME_ENTRY_create_by_NID(),
+and X509_NAME_ENTRY_create_by_OBJ() create and return an 
+
+X509_NAME_ENTRY_create_by_txt(), X509_NAME_ENTRY_create_by_OBJ(),
+X509_NAME_ENTRY_create_by_NID() and X509_NAME_ENTRY_set_data()
+are seldom used in practice because B<X509_NAME_ENTRY> structures
+are almost always part of B<X509_NAME> structures and the
+corresponding B<X509_NAME> functions are typically used to
+create and add new entries in a single operation.
+
+The arguments of these functions support similar options to the similarly
+named ones of the corresponding B<X509_NAME> functions such as
+X509_NAME_add_entry_by_txt(). So for example B<type> can be set to
+B<MBSTRING_ASC> but in the case of X509_set_data() the field name must be
+set first so the relevant field information can be looked up internally.
+
+=head1 RETURN VALUES
+
+=head1 SEE ALSO
+
+L<ERR_get_error(3)|ERR_get_error(3)>, L<d2i_X509_NAME(3)|d2i_X509_NAME(3)>,
+L<OBJ_nid2obj(3),OBJ_nid2obj(3)>
+
+=head1 HISTORY
+
+TBA
+
+=cut

doc/crypto/X509_NAME_add_entry_by_txt.pod

@@ -0,0 +1,110 @@
+=pod
+
+X509_NAME_add_entry_by_txt, X509_NAME_add_entry_by_OBJ, X509_NAME_add_entry_by_NID,
+X509_NAME_add_entry, X509_NAME_delete_entry - X509_NAME modification functions
+
+=head1 NAME
+
+=head1 SYNOPSIS
+
+int X509_NAME_add_entry_by_txt(X509_NAME *name, char *field, int type, unsigned char *bytes, int len, int loc, int set);
+int X509_NAME_add_entry_by_OBJ(X509_NAME *name, ASN1_OBJECT *obj, int type, unsigned char *bytes, int len, int loc, int set);
+int X509_NAME_add_entry_by_NID(X509_NAME *name, int nid, int type, unsigned char *bytes, int len, int loc, int set);
+int X509_NAME_add_entry(X509_NAME *name,X509_NAME_ENTRY *ne, int loc, int set);
+X509_NAME_ENTRY *X509_NAME_delete_entry(X509_NAME *name, int loc);
+
+=head1 DESCRIPTION
+
+X509_NAME_add_entry_by_txt(), X509_NAME_add_entry_by_OBJ() and
+X509_NAME_add_entry_by_NID() add a field whose name is defined
+by a string B<field>, an object B<obj> or a NID B<nid> respectively.
+The field value to be added is in B<bytes> of length B<len>. If
+B<len> is -1 then the field length is calculated internally using
+strlen(bytes).
+
+The type of field is determined by B<type> which can either be a
+definition of the type of B<bytes> (such as B<MBSTRING_ASC>) or a
+standard ASN1 type (such as B<V_ASN1_IA5STRING>). The new entry is
+added to a position determined by B<loc> and B<set>.
+
+X509_NAME_add_entry() adds a copy of B<X509_NAME_ENTRY> structure B<ne>
+to B<name>. The new entry is added to a position determined by B<loc>
+and B<set>. Since a copy of B<ne> is added B<ne> must be freed up after
+the call.
+
+X509_NAME_delete_entry() deletes an entry from B<name> at position
+B<loc>. The deleted entry is returned and must be freed up.
+
+=head1 NOTES
+
+The use of string types such as B<MBSTRING_ASC> or B<MBSTRING_UTF8>
+is strongly recommened for the B<type> parameter. This allows the
+internal code to correctly determine the type of the field and to
+apply length checks according to the relevant standards. This is
+done using ASN1_STRING_set_by_NID().
+
+If instead an ASN1 type is used no checks are performed and the
+supplied data in B<bytes> is used directly.
+
+In X509_NAME_add_entry_by_txt() the B<field> string represents
+the field name using OBJ_txt2obj(field, 0).
+
+The B<loc> and B<set> parameters determine where a new entry should
+be added. For almost all applications B<loc> can be set to -1 and B<set>
+to 0. This adds a new entry to the end of B<name> as a single valued
+RelativeDistinguishedName (RDN).
+
+B<loc> actually determines the index where the new entry is inserted:
+if it is -1 it is appended. 
+
+B<set> determines how the new type is added. If it is zero a
+new RDN is created.
+
+If B<set> is -1 or 1 it is added to the previous or next RDN
+structure respectively. This will then be a multivalued RDN:
+since multivalues RDNs are very seldom used B<set> is almost
+always set to zero.
+
+=head1 EXAMPLES
+
+Create an B<X509_NAME> structure:
+
+"C=UK, O=Disorganized Organization, CN=Joe Bloggs"
+
+ X509_NAME *nm;
+ nm = X509_NAME_new();
+ if (nm == NULL)
+	/* Some error */
+ if (!X509_NAME_add_entry_by_txt(nm, MBSTRING_ASC,
+			"C", "UK", -1, -1, 0))
+	/* Error */
+ if (!X509_NAME_add_entry_by_txt(nm, MBSTRING_ASC,
+			"O", "Disorganized Organization", -1, -1, 0))
+	/* Error */
+ if (!X509_NAME_add_entry_by_txt(nm, MBSTRING_ASC,
+			"CN", "Joe Bloggs", -1, -1, 0))
+	/* Error */
+
+=head1 RETURN VALUES
+
+X509_NAME_add_entry_by_txt(), X509_NAME_add_entry_by_OBJ(),
+X509_NAME_add_entry_by_NID() and X509_NAME_add_entry() return 1 for
+success of 0 if an error occurred.
+
+X509_NAME_delete_entry() returns either the deleted B<X509_NAME_ENTRY>
+structure of B<NULL> if an error occurred.
+
+=head1 BUGS
+
+B<type> can still be set to B<V_ASN1_APP_CHOOSE> to use a
+different algorithm to determine field types. Since this form does
+not understand multicharacter types, performs no length checks and
+can result in invalid field types its use is strongly discouraged.
+
+=head1 SEE ALSO
+
+L<ERR_get_error(3)|ERR_get_error(3)>, L<d2i_X509_NAME(3)|d2i_X509_NAME(3)>
+
+=head1 HISTORY
+
+=cut

doc/crypto/X509_NAME_print_ex.pod

@@ -0,0 +1,105 @@
+=pod
+
+=head1 NAME
+
+X509_NAME_print_ex, X509_NAME_print_ex_fp, X509_NAME_print,
+X509_NAME_oneline - X509_NAME printing routines.
+
+=head1 SYNOPSIS
+
+ #include <openssl/x509.h>
+
+ int X509_NAME_print_ex(BIO *out, X509_NAME *nm, int indent, unsigned long flags);
+ int X509_NAME_print_ex_fp(FILE *fp, X509_NAME *nm, int indent, unsigned long flags);
+ char *	X509_NAME_oneline(X509_NAME *a,char *buf,int size);
+ int X509_NAME_print(BIO *bp, X509_NAME *name, int obase);
+
+=head1 DESCRIPTION
+
+X509_NAME_print_ex() prints a human readable version of B<nm> to BIO B<out>. Each
+line (for multiline formats) is indented by B<indent> spaces. The output format
+can be extensively customised by use of the B<flags> parameter.
+
+X509_NAME_print_ex_fp() is identical to X509_NAME_print_ex() except the output is
+written to FILE pointer B<fp>.
+
+X509_NAME_oneline() prints an ASCII version of B<a> to B<buf>. At most B<size>
+bytes will be written. If B<buf> is B<NULL> then a buffer is dynamically allocated
+and returned, otherwise B<buf> is returned.
+
+X509_NAME_print() prints out B<name> to B<bp> indenting each line by B<obase> 
+characters. Multiple lines are used if the output (including indent) exceeds
+80 characters.
+
+=head1 NOTES
+
+The functions X509_NAME_oneline() and X509_NAME_print() are legacy functions which
+produce a non standard output form, they don't handle multi character fields and
+have various quirks and inconsistencies. Their use is strongly discouraged in new
+applications.
+
+Although there are a large number of possible flags for most purposes
+B<XN_FLAG_ONELINE>, B<XN_FLAG_MULTILINE> or B<XN_FLAG_RFC2253> will suffice.
+As noted on the L<ASN1_STRING_print_ex(3)|ASN1_STRING_print_ex(3)> manual page
+for UTF8 terminals the B<ASN1_STRFLAGS_ESC_MSB> should be unset: so for example
+B<XN_FLAG_ONELINE & ~ASN1_STRFLAGS_ESC_MSB> would be used.
+
+The complete set of the flags supported by X509_NAME_print_ex() is listed below.
+
+Several options can be ored together.
+
+The options B<XN_FLAG_SEP_COMMA_PLUS>, B<XN_FLAG_SEP_CPLUS_SPC>,
+B<XN_FLAG_SEP_SPLUS_SPC> and B<XN_FLAG_SEP_MULTILINE> determine the field separators
+to use. Two distinct separators are used between distinct RelativeDistinguishedName
+components and separate values in the same RDN for a multi-valued RDN. Multi-valued
+RDNs are currently very rare so the second separator will hardly ever be used.
+
+B<XN_FLAG_SEP_COMMA_PLUS> uses comma and plus as separators. B<XN_FLAG_SEP_CPLUS_SPC>
+uses comma and plus with spaces: this is more readable that plain comma and plus.
+B<XN_FLAG_SEP_SPLUS_SPC> uses spaced semicolon and plus. B<XN_FLAG_SEP_MULTILINE> uses
+spaced newline and plus respectively.
+
+If B<XN_FLAG_DN_REV> is set the whole DN is printed in reversed order.
+
+The fields B<XN_FLAG_FN_SN>, B<XN_FLAG_FN_LN>, B<XN_FLAG_FN_OID>,
+B<XN_FLAG_FN_NONE> determine how a field name is displayed. It will
+use the short name (e.g. CN) the long name (e.g. commonName) always
+use OID numerical form (normally OIDs are only used if the field name is not
+recognised) and no field name respectively.
+
+If B<XN_FLAG_SPC_EQ> is set then spaces will be placed around the '=' character
+separating field names and values.
+
+If B<XN_FLAG_DUMP_UNKNOWN_FIELDS> is set then the encoding of unknown fields is
+printed instead of the values.
+
+If B<XN_FLAG_FN_ALIGN> is set then field names are padded to 20 characters: this
+is only of use for multiline format.
+
+Additionally all the options supported by ASN1_STRING_print_ex() can be used to 
+control how each field value is displayed.
+
+In addition a number options can be set for commonly used formats.
+
+B<XN_FLAG_RFC2253> sets options which produce an output compatible with RFC2253 it
+is equivalent to:
+ B<ASN1_STRFLGS_RFC2253 | XN_FLAG_SEP_COMMA_PLUS | XN_FLAG_DN_REV | XN_FLAG_FN_SN | XN_FLAG_DUMP_UNKNOWN_FIELDS>
+
+
+B<XN_FLAG_ONELINE> is a more readable one line format it is the same as:
+ B<ASN1_STRFLGS_RFC2253 | ASN1_STRFLGS_ESC_QUOTE | XN_FLAG_SEP_CPLUS_SPC | XN_FLAG_SPC_EQ | XN_FLAG_FN_SN>
+
+B<XN_FLAG_MULTILINE> is a multiline format is is the same as:
+ B<ASN1_STRFLGS_ESC_CTRL | ASN1_STRFLGS_ESC_MSB | XN_FLAG_SEP_MULTILINE | XN_FLAG_SPC_EQ | XN_FLAG_FN_LN | XN_FLAG_FN_ALIGN>
+
+B<XN_FLAG_COMPAT> uses a format identical to X509_NAME_print(): in fact it calls X509_NAME_print() internally.
+
+=head1 SEE ALSO
+
+L<ASN1_STRING_print_ex(3)|ASN1_STRING_print_ex(3)>
+
+=head1 HISTORY
+
+TBA
+
+=cut