Date: Mon, 9 May 2011 00:04:36 +0000 (UTC) From: Gabor Kovesdan <gabor@FreeBSD.org> To: src-committers@freebsd.org, svn-src-user@freebsd.org Subject: svn commit: r221672 - user/gabor/tre-integration/include Message-ID: <201105090004.p4904ahw067625@svn.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: gabor Date: Mon May 9 00:04:36 2011 New Revision: 221672 URL: http://svn.freebsd.org/changeset/base/221672 Log: - Bring in TRE headers into include so that it can be built as a part of the base system. - It is still called tre.h for compatibility of TRE-specific applications so regex.h only contains an include to it Added: user/gabor/tre-integration/include/tre.h - copied unchanged from r221610, user/gabor/tre-integration/contrib/tre/lib/tre.h Replaced: user/gabor/tre-integration/include/regex.h - copied unchanged from r221610, user/gabor/tre-integration/contrib/tre/lib/regex.h Modified: user/gabor/tre-integration/include/Makefile Modified: user/gabor/tre-integration/include/Makefile ============================================================================== --- user/gabor/tre-integration/include/Makefile Sun May 8 21:46:12 2011 (r221671) +++ user/gabor/tre-integration/include/Makefile Mon May 9 00:04:36 2011 (r221672) @@ -22,7 +22,7 @@ INCS= a.out.h ar.h assert.h bitstring.h signal.h spawn.h stab.h \ stdbool.h stddef.h stdio.h stdlib.h string.h stringlist.h \ strings.h sysexits.h tar.h termios.h tgmath.h \ - time.h timeconv.h timers.h ttyent.h \ + time.h timeconv.h timers.h tre.h ttyent.h \ ulimit.h unistd.h utime.h utmpx.h uuid.h varargs.h vis.h \ wchar.h wctype.h wordexp.h Copied: user/gabor/tre-integration/include/regex.h (from r221610, user/gabor/tre-integration/contrib/tre/lib/regex.h) ============================================================================== --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ user/gabor/tre-integration/include/regex.h Mon May 9 00:04:36 2011 (r221672, copy of r221610, user/gabor/tre-integration/contrib/tre/lib/regex.h) @@ -0,0 +1,36 @@ +/* + regex.h - TRE legacy API + + This software is released under a BSD-style license. + See the file LICENSE for details and copyright. + + This header is for source level compatibility with old code using + the <tre/regex.h> header which defined the TRE API functions without + a prefix. New code should include <tre/tre.h> instead. + +*/ + +#ifndef TRE_REXEX_H +#define TRE_REGEX_H 1 + +#include "tre.h" + +#define regcomp tre_regcomp +#define regerror tre_regerror +#define regexec tre_regexec +#define regfree tre_regfree + +#define regacomp tre_regacomp +#define regaexec tre_regaexec +#define regancomp tre_regancomp +#define reganexec tre_reganexec +#define regawncomp tre_regawncomp +#define regawnexec tre_regawnexec +#define regncomp tre_regncomp +#define regnexec tre_regnexec +#define regwcomp tre_regwcomp +#define regwexec tre_regwexec +#define regwncomp tre_regwncomp +#define regwnexec tre_regwnexec + +#endif /* TRE_REGEX_H */ Copied: user/gabor/tre-integration/include/tre.h (from r221610, user/gabor/tre-integration/contrib/tre/lib/tre.h) ============================================================================== --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ user/gabor/tre-integration/include/tre.h Mon May 9 00:04:36 2011 (r221672, copy of r221610, user/gabor/tre-integration/contrib/tre/lib/tre.h) @@ -0,0 +1,227 @@ +/* + tre.h - TRE public API definitions + + This software is released under a BSD-style license. + See the file LICENSE for details and copyright. + +*/ + +#ifndef TRE_H +#define TRE_H 1 + +#include <sys/types.h> + +#define TRE_WCHAR 1 +#define TRE_APPROX 1 + +#ifdef __cplusplus +extern "C" { +#endif + +typedef int regoff_t; +typedef struct { + size_t re_nsub; /* Number of parenthesized subexpressions. */ + void *value; /* For internal use only. */ +} regex_t; + +typedef struct { + regoff_t rm_so; + regoff_t rm_eo; +} regmatch_t; + + +typedef enum { + REG_OK = 0, /* No error. */ + /* POSIX tre_regcomp() return error codes. (In the order listed in the + standard.) */ + REG_NOMATCH, /* No match. */ + REG_BADPAT, /* Invalid regexp. */ + REG_ECOLLATE, /* Unknown collating element. */ + REG_ECTYPE, /* Unknown character class name. */ + REG_EESCAPE, /* Trailing backslash. */ + REG_ESUBREG, /* Invalid back reference. */ + REG_EBRACK, /* "[]" imbalance */ + REG_EPAREN, /* "\(\)" or "()" imbalance */ + REG_EBRACE, /* "\{\}" or "{}" imbalance */ + REG_BADBR, /* Invalid content of {} */ + REG_ERANGE, /* Invalid use of range operator */ + REG_ESPACE, /* Out of memory. */ + REG_BADRPT /* Invalid use of repetition operators. */ +} reg_errcode_t; + +/* POSIX tre_regcomp() flags. */ +#define REG_EXTENDED 1 +#define REG_ICASE (REG_EXTENDED << 1) +#define REG_NEWLINE (REG_ICASE << 1) +#define REG_NOSUB (REG_NEWLINE << 1) + +/* Extra tre_regcomp() flags. */ +#define REG_BASIC 0 +#define REG_LITERAL (REG_NOSUB << 1) +#define REG_RIGHT_ASSOC (REG_LITERAL << 1) +#define REG_UNGREEDY (REG_RIGHT_ASSOC << 1) + +/* POSIX tre_regexec() flags. */ +#define REG_NOTBOL 1 +#define REG_NOTEOL (REG_NOTBOL << 1) + +/* Extra tre_regexec() flags. */ +#define REG_APPROX_MATCHER (REG_NOTEOL << 1) +#define REG_BACKTRACKING_MATCHER (REG_APPROX_MATCHER << 1) + +/* REG_NOSPEC and REG_LITERAL mean the same thing. */ +#if defined(REG_LITERAL) && !defined(REG_NOSPEC) +#define REG_NOSPEC REG_LITERAL +#elif defined(REG_NOSPEC) && !defined(REG_LITERAL) +#define REG_LITERAL REG_NOSPEC +#endif /* defined(REG_NOSPEC) */ + +/* The maximum number of iterations in a bound expression. */ +#undef RE_DUP_MAX +#define RE_DUP_MAX 255 + +/* The POSIX.2 regexp functions */ +extern int +tre_regcomp(regex_t *preg, const char *regex, int cflags); + +extern int +tre_regexec(const regex_t *preg, const char *string, size_t nmatch, + regmatch_t pmatch[], int eflags); + +extern size_t +tre_regerror(int errcode, const regex_t *preg, char *errbuf, + size_t errbuf_size); + +extern void +tre_regfree(regex_t *preg); + +#ifdef TRE_WCHAR +#include <wchar.h> + +/* Wide character versions (not in POSIX.2). */ +extern int +tre_regwcomp(regex_t *preg, const wchar_t *regex, int cflags); + +extern int +tre_regwexec(const regex_t *preg, const wchar_t *string, + size_t nmatch, regmatch_t pmatch[], int eflags); +#endif /* TRE_WCHAR */ + +/* Versions with a maximum length argument and therefore the capability to + handle null characters in the middle of the strings (not in POSIX.2). */ +extern int +tre_regncomp(regex_t *preg, const char *regex, size_t len, int cflags); + +extern int +tre_regnexec(const regex_t *preg, const char *string, size_t len, + size_t nmatch, regmatch_t pmatch[], int eflags); + +#ifdef TRE_WCHAR +extern int +tre_regwncomp(regex_t *preg, const wchar_t *regex, size_t len, int cflags); + +extern int +tre_regwnexec(const regex_t *preg, const wchar_t *string, size_t len, + size_t nmatch, regmatch_t pmatch[], int eflags); +#endif /* TRE_WCHAR */ + +#ifdef TRE_APPROX + +/* Approximate matching parameter struct. */ +typedef struct { + int cost_ins; /* Default cost of an inserted character. */ + int cost_del; /* Default cost of a deleted character. */ + int cost_subst; /* Default cost of a substituted character. */ + int max_cost; /* Maximum allowed cost of a match. */ + + int max_ins; /* Maximum allowed number of inserts. */ + int max_del; /* Maximum allowed number of deletes. */ + int max_subst; /* Maximum allowed number of substitutes. */ + int max_err; /* Maximum allowed number of errors total. */ +} regaparams_t; + +/* Approximate matching result struct. */ +typedef struct { + size_t nmatch; /* Length of pmatch[] array. */ + regmatch_t *pmatch; /* Submatch data. */ + int cost; /* Cost of the match. */ + int num_ins; /* Number of inserts in the match. */ + int num_del; /* Number of deletes in the match. */ + int num_subst; /* Number of substitutes in the match. */ +} regamatch_t; + + +/* Approximate matching functions. */ +extern int +tre_regaexec(const regex_t *preg, const char *string, + regamatch_t *match, regaparams_t params, int eflags); + +extern int +tre_reganexec(const regex_t *preg, const char *string, size_t len, + regamatch_t *match, regaparams_t params, int eflags); +#ifdef TRE_WCHAR +/* Wide character approximate matching. */ +extern int +tre_regawexec(const regex_t *preg, const wchar_t *string, + regamatch_t *match, regaparams_t params, int eflags); + +extern int +tre_regawnexec(const regex_t *preg, const wchar_t *string, size_t len, + regamatch_t *match, regaparams_t params, int eflags); +#endif /* TRE_WCHAR */ + +/* Sets the parameters to default values. */ +extern void +tre_regaparams_default(regaparams_t *params); +#endif /* TRE_APPROX */ + +#ifdef TRE_WCHAR +typedef wchar_t tre_char_t; +#else /* !TRE_WCHAR */ +typedef unsigned char tre_char_t; +#endif /* !TRE_WCHAR */ + +typedef struct { + int (*get_next_char)(tre_char_t *c, unsigned int *pos_add, void *context); + void (*rewind)(size_t pos, void *context); + int (*compare)(size_t pos1, size_t pos2, size_t len, void *context); + void *context; +} tre_str_source; + +extern int +tre_reguexec(const regex_t *preg, const tre_str_source *string, + size_t nmatch, regmatch_t pmatch[], int eflags); + +/* Returns the version string. The returned string is static. */ +extern char * +tre_version(void); + +/* Returns the value for a config parameter. The type to which `result' + must point to depends of the value of `query', see documentation for + more details. */ +extern int +tre_config(int query, void *result); + +enum { + TRE_CONFIG_APPROX, + TRE_CONFIG_WCHAR, + TRE_CONFIG_MULTIBYTE, + TRE_CONFIG_SYSTEM_ABI, + TRE_CONFIG_VERSION +}; + +/* Returns 1 if the compiled pattern has back references, 0 if not. */ +extern int +tre_have_backrefs(const regex_t *preg); + +/* Returns 1 if the compiled pattern uses approximate matching features, + 0 if not. */ +extern int +tre_have_approx(const regex_t *preg); + +#ifdef __cplusplus +} +#endif +#endif /* TRE_H */ + +/* EOF */
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201105090004.p4904ahw067625>