amxmodx/plugins/include/regex.inc

349 lines
16 KiB
SourcePawn
Raw Normal View History

// vim: set ts=4 sw=4 tw=99 noet:
//
// AMX Mod X, based on AMX Mod by Aleksander Naszko ("OLO").
// Copyright (C) The AMX Mod X Development Team.
//
// This software is licensed under the GNU General Public License, version 3 or higher.
// Additional exceptions apply. For full license details, see LICENSE.txt or visit:
// https://alliedmods.net/amxmodx-license
//
// Regular Expressions API
//
2004-10-05 12:31:42 +04:00
#if defined _regex_included
#endinput
2004-10-05 12:31:42 +04:00
#endif
#define _regex_included
#pragma reqlib regex
#if !defined AMXMODX_NOAUTOLOAD
#pragma loadlib regex
2006-05-10 14:42:49 +04:00
#endif
2005-07-15 23:05:31 +04:00
2004-10-05 12:31:42 +04:00
enum Regex
{
2014-07-17 13:20:52 +04:00
REGEX_MATCH_FAIL = -2,
REGEX_PATTERN_FAIL = -1,
REGEX_NO_MATCH = 0,
REGEX_OK = 1
2004-10-05 12:31:42 +04:00
};
/**
* Flags for compiling regex expressions.
* These come directly from the pcre library and can be used in regex_compile_ex.
*/
#define PCRE_CASELESS 0x00000001 /* Ignore Case */
#define PCRE_MULTILINE 0x00000002 /* Multilines (affects ^ and $ so that they match the start/end of a line rather than matching the start/end of the string). */
#define PCRE_DOTALL 0x00000004 /* Single line (affects . so that it matches any character, even new line characters). */
#define PCRE_EXTENDED 0x00000008 /* Pattern extension (ignore whitespace and # comments). */
#define PCRE_ANCHORED 0x00000010 /* Force pattern anchoring. */
#define PCRE_DOLLAR_ENDONLY 0x00000020 /* $ not to match newline at end. */
#define PCRE_UNGREEDY 0x00000200 /* Invert greediness of quantifiers */
#define PCRE_NOTEMPTY 0x00000400 /* An empty string is not a valid match. */
#define PCRE_UTF8 0x00000800 /* Use UTF-8 Chars */
#define PCRE_NO_UTF8_CHECK 0x00002000 /* Do not check the pattern for UTF-8 validity (only relevant if PCRE_UTF8 is set) */
2014-07-17 16:58:57 +04:00
#define PCRE_NEVER_UTF 0x00010000 /* Lock out interpretation of the pattern as UTF-8 */
#define PCRE_FIRSTLINE 0x00040000 /* Force matching to be before newline */
#define PCRE_DUPNAMES 0x00080000 /* Allow duplicate names for subpattern */
2014-07-17 16:58:57 +04:00
#define PCRE_NEWLINE_CR 0x00100000 /* Specify that a newline is indicated by a single character CR ) */
#define PCRE_NEWLINE_CRLF 0x00300000 /* specify that a newline is indicated by the two-character CRLF sequence ) Overrides the default */
#define PCRE_NEWLINE_ANY 0x00400000 /* Specify that any Unicode newline sequence should be recognized. ) newline definition (LF) */
#define PCRE_NEWLINE_ANYCRLF 0x00500000 /* Specify that any of CR, LF and CRLF sequences should be recognized ) */
#define PCRE_UCP 0x20000000 /* Change the way PCRE processes \B, \b, \D, \d, \S, \s, \W, \w etc. to use Unicode properties */
/**
* Regex expression error codes.
* This can be used with regex_compile_ex and regex_match_ex.
*/
enum /*RegexError*/
{
REGEX_ERROR_NONE = 0, /* No error */
REGEX_ERROR_NOMATCH = -1, /* No match was found */
REGEX_ERROR_NULL = -2,
REGEX_ERROR_BADOPTION = -3,
REGEX_ERROR_BADMAGIC = -4,
REGEX_ERROR_UNKNOWN_OPCODE = -5,
REGEX_ERROR_NOMEMORY = -6,
REGEX_ERROR_NOSUBSTRING = -7,
REGEX_ERROR_MATCHLIMIT = -8,
REGEX_ERROR_CALLOUT = -9, /* Never used by PCRE itself */
REGEX_ERROR_BADUTF8 = -10,
REGEX_ERROR_BADUTF8_OFFSET = -11,
REGEX_ERROR_PARTIAL = -12,
REGEX_ERROR_BADPARTIAL = -13,
REGEX_ERROR_INTERNAL = -14,
REGEX_ERROR_BADCOUNT = -15,
REGEX_ERROR_DFA_UITEM = -16,
REGEX_ERROR_DFA_UCOND = -17,
REGEX_ERROR_DFA_UMLIMIT = -18,
REGEX_ERROR_DFA_WSSIZE = -19,
REGEX_ERROR_DFA_RECURSE = -20,
REGEX_ERROR_RECURSIONLIMIT = -21,
REGEX_ERROR_NULLWSLIMIT = -22, /* No longer actually used */
REGEX_ERROR_BADNEWLINE = -23,
REGEX_ERROR_BADOFFSET = -24,
REGEX_ERROR_SHORTUTF8 = -25,
REGEX_ERROR_RECURSELOOP = -26,
REGEX_ERROR_JIT_STACKLIMIT = -27,
REGEX_ERROR_BADMODE = -28,
REGEX_ERROR_BADENDIANNESS = -29,
REGEX_ERROR_DFA_BADRESTART = -30,
REGEX_ERROR_JIT_BADOPTION = -31,
REGEX_ERROR_BADLENGTH = -32,
REGEX_ERROR_UNSET = -33
};
/**
* Precompile a regular expression.
*
2014-07-18 19:58:37 +04:00
* @note Use this if you intend on using the same expression multiple times.
* Pass the regex handle returned here to regex_match_c to check for matches.
*
* @note This handle is automatically freed on map change. However,
* if you are completely done with it before then, you should
* call regex_free on this handle.
*
2014-07-18 19:58:37 +04:00
* @note Consider using regex_compile_ex instead if you want to use PCRE_* flags.
2014-07-17 16:58:57 +04:00
*
* @param pattern The regular expression pattern.
* @param ret Error code encountered, if applicable.
* @param error Error message encountered, if applicable.
* @param maxLen Maximum string length of the error buffer.
* @param flags General flags for the regular expression.
* i = Ignore case
* m = Multilines (affects ^ and $ so that they match
* the start/end of a line rather than matching the
* start/end of the string).
* s = Single line (affects . so that it matches any character,
* even new line characters).
* x = Pattern extension (ignore whitespace and # comments).
*
* @return -1 on error in the pattern, > valid regex handle (> 0) on success.
*/
native Regex:regex_compile(const pattern[], &ret = 0, error[] = "", maxLen = 0, const flags[]="");
/**
* Matches a string against a pre-compiled regular expression pattern.
*
* @note You should free the returned handle with regex_free()
* when you are done with this pattern.
*
* @note Use the regex handle passed to this function to extract
* matches with regex_substr().
*
* @param string The string to check.
* @param pattern The regular expression pattern.
* @param ret Error code, if applicable, or number of results on success. See REGEX_ERROR_* defines.
2014-07-17 16:58:57 +04:00
*
* @return -2 = Matching error (error code is stored in ret)
* 0 = No match.
* >1 = Number of results.
*/
native regex_match_c(const string[], Regex:pattern, &ret = 0);
/**
* Matches a string against a regular expression pattern.
*
* @note If you intend on using the same regular expression pattern
* multiple times, consider using regex_compile and regex_match_ex
* instead of making this function reparse the expression each time.
*
* @note Flags only exist in amxmodx 1.8 and later.
*
* @note You should free the returned handle with regex_free()
* when you are done extracting all of the substrings.
*
* @param string The string to check.
* @param pattern The regular expression pattern.
* @param ret Error code, or result state of the match.
* @param error Error message, if applicable.
* @param maxLen Maximum length of the error buffer.
* @param flags General flags for the regular expression.
* i = Ignore case
* m = Multilines (affects ^ and $ so that they match
* the start/end of a line rather than matching the
* start/end of the string).
* s = Single line (affects . so that it matches any character,
* even new line characters).
* x = Pattern extension (ignore whitespace and # comments).
*
* @return -2 = Matching error (error code is stored in ret)
* -1 = Error in pattern (error message and offset # in error and ret)
* 0 = No match.
* >1 = Handle for getting more information (via regex_substr)
*/
native Regex:regex_match(const string[], const pattern[], &ret = 0, error[] = "", maxLen = 0, const flags[] = "");
/**
* Returns a matched substring from a regex handle.
*
* @note Substring ids start at 0 and end at ret - 1, where ret is from the corresponding
2014-07-17 16:58:57 +04:00
* regex_match* function call.
*
* @param id The regex handle to extract data from.
* @param str_id The index of the expression to get - starts at 0, and ends at ret - 1.
* @param buffer The buffer to set to the matching substring.
* @param maxLen The maximum string length of the buffer.
*
* @return 1 on success, otherwise 0 on failure.
*/
native regex_substr(Regex:id, str_id, buffer[], maxLen);
/**
* Frees the memory associated with a regex result, and sets the handle to 0.
*
* @note This must be called on all results from regex_match() when you are done extracting
* the results with regex_substr().
*
2014-07-17 16:58:57 +04:00
* @note The results of regex_compile() or regex_compile_ex() (and subsequently, regex_match_c())
* only need to be freed when you are done using the pattern.
*
* @note Do not use the handle again after freeing it!
*
* @param id The regex handle to free.
* @noreturn
*/
native regex_free(&Regex:id);
/**
* The following natives are only available in 1.8.3 and above.
*/
/**
* Precompile a regular expression.
*
2014-07-18 19:58:37 +04:00
* @note Use this if you intend on using the same expression multiple times.
2014-07-17 16:58:57 +04:00
* Pass the regex handle returned here to regex_match_c() to check for matches.
*
2014-07-18 19:58:37 +04:00
* @note Unlike regex_compile(), this allows you to use PCRE flags directly.
*
* @param pattern The regular expression pattern.
* @param flags General flags for the regular expression, see PCRE_* defines.
* @param error Error message encountered, if applicable.
* @param maxLen Maximum string length of the error buffer.
* @param errcode Regex type error code encountered, if applicable. See REGEX_ERROR_* defines.
*
* @return Valid regex handle (> 0) on success, or -1 on failure.
*/
native Regex:regex_compile_ex(const pattern[], flags = 0, error[]= "", maxLen = 0, &errcode = 0);
/**
2014-07-17 16:58:57 +04:00
* Matches a string against a pre-compiled regular expression pattern, matching all
2014-07-18 19:58:37 +04:00
* occurrences of the pattern inside the string. This is similar to using the "g" flag
* in perl regex.
*
2014-07-17 16:58:57 +04:00
* @note You should free the returned handle (with regex_free())
* when you are done with this pattern.
*
* @note Use the regex handle passed to this function to extract
* matches with regex_substr().
*
* @param pattern The regular expression pattern.
* @param string The string to check.
2014-07-17 16:58:57 +04:00
* @param ret Error code, if applicable, or number of results on success.
* See REGEX_ERROR_* defines.
*
* @return -2 = Matching error (error code is stored in ret)
* 0 = No match.
* >1 = Number of results.
*/
native regex_match_all_c(const string[], Regex:pattern, &ret = 0);
/**
2014-07-18 19:58:37 +04:00
* Matches a string against a regular expression pattern, matching all occurrences of the
* pattern inside the string. This is similar to using the "g" flag in perl regex.
*
* @note If you intend on using the same regular expression pattern
* multiple times, consider using regex_compile and regex_match_ex
* instead of making this function reparse the expression each time.
*
* @note Flags only exist in amxmodx 1.8 and later.
*
* @note You should free the returned handle with regex_free()
* when you are done extracting all of the substrings.
*
* @param string The string to check.
* @param pattern The regular expression pattern.
* @param flags General flags for the regular expression, see PCRE_* defines.
* @param error Error message encountered, if applicable.
* @param maxLen Maximum string length of the error buffer.
* @param errcode Regex type error code encountered, if applicable. See REGEX_ERROR_* defines.
*
* @return -2 = Matching error (error code is stored in ret)
* -1 = Error in pattern (error message and offset # in error and ret)
* 0 = No match.
* >1 = Handle for getting more information (via regex_substr)
*/
native Regex:regex_match_all(const string[], const pattern[], flags = 0, error[]= "", maxLen = 0, &errcode = 0);
/**
* Matches a string against a regular expression pattern.
*
* @note If you intend on using the same regular expression pattern
2014-07-17 16:58:57 +04:00
* multiple times, consider using compile regex_compile_ex and regex_match*
* instead of making this function reparse the expression each time.
*
* @param str The string to check.
* @param pattern The regular expression pattern.
* @param flags General flags for the regular expression.
* @param error Error message, if applicable.
* @param maxLen Maximum length of the error buffer.
2014-07-17 13:20:52 +04:00
* @param errcode Regex type error code encountered, if applicable. See REGEX_ERROR_* defines.
*
* @return -2 = Matching error (error code is stored in ret)
* -1 = Pattern error (error code is stored in ret)
* 0 = No match.
* >1 = Number of results.
2004-10-05 12:31:42 +04:00
*/
stock regex_match_simple(const str[], const pattern[], flags = 0, error[]= "", maxLen = 0, &errcode = 0)
{
new Regex:regex = regex_compile_ex(pattern, flags, error, maxLen, errcode);
2014-11-07 16:11:55 +03:00
if (regex < REGEX_OK)
{
return -1;
}
new substrings = regex_match_c(str, regex);
regex_free(regex);
return substrings;
2014-07-17 13:20:52 +04:00
}
/**
2014-07-18 19:58:37 +04:00
* Flags used with regex_replace to control the replacement behavior.
2014-07-17 13:20:52 +04:00
*/
#define REGEX_FORMAT_DEFAULT 0 /* Uses the standard formatting rules to replace matches */
2014-07-17 16:58:57 +04:00
#define REGEX_FORMAT_NOCOPY (1<<0) /* The sections that do not match the regular expression are not copied when replacing matches. */
#define REGEX_FORMAT_FIRSTONLY (1<<1) /* Only the first occurrence of a regular expression is replaced. */
2014-07-17 13:20:52 +04:00
/**
* Perform a regular expression search and replace.
*
2014-07-18 19:58:37 +04:00
* An optional parameter, flags, allows you to specify options on how the replacement is performed.
2014-07-17 13:20:52 +04:00
* Supported format specifiers for replace parameter:
2014-07-17 16:58:57 +04:00
* $number : Substitutes the substring matched by group number.
2014-07-17 13:20:52 +04:00
* n must be an integer value designating a valid backreference, greater than 0, and of two digits at most.
* ${name} : Substitutes the substring matched by the named group name (a maximum of 32 characters).
* $& : Substitutes a copy of the whole match.
* $` : Substitutes all the text of the input string before the match.
* $' : Substitutes all the text of the input string after the match.
* $+ : Substitutes the last group that was captured.
2014-07-17 16:58:57 +04:00
* $_ : Substitutes the entire input string.
2014-07-17 13:20:52 +04:00
* $$ : Substitutes a literal "$".
* As note, the character \ can be also used with format specifier, this is same hehavior as $.
*
* @param pattern The regular expression pattern.
* @param string The string to check.
* @param error Error message, if applicable.
* @param maxLen Maximum length of the error buffer.
* @param replace The string will be used to replace any matches. See above for format specifiers.
2014-07-18 19:58:37 +04:00
* @param flags General flags to control how the string is replaced. See REGEX_FORMAT_* defines.
2014-07-17 13:20:52 +04:00
* @param errcode Regex type error code encountered, if applicable. See REGEX_ERROR_* defines.
*
* @return -2 = Matching error (error code is stored in ret)
* 0 = No match.
* >1 = Number of matches.
*/
native regex_replace(Regex:pattern, string[], maxLen, const replace[], flags = REGEX_FORMAT_DEFAULT, &errcode = 0);