unzip/unzip-6.0-fix-recmatch.patch

478 lines
18 KiB
Diff

diff -up unzip60/match.c.recmatch unzip60/match.c
--- unzip60/match.c.recmatch 2005-08-14 13:00:36.000000000 -0400
+++ unzip60/match.c 2013-05-28 10:29:57.949077543 -0400
@@ -27,16 +27,14 @@
---------------------------------------------------------------------------
- Copyright on recmatch() from Zip's util.c (although recmatch() was almost
- certainly written by Mark Adler...ask me how I can tell :-) ):
+ Copyright on recmatch() from Zip's util.c
+ Copyright (c) 1990-2005 Info-ZIP. All rights reserved.
- Copyright (C) 1990-1992 Mark Adler, Richard B. Wales, Jean-loup Gailly,
- Kai Uwe Rommel and Igor Mandrichenko.
+ See the accompanying file LICENSE, version 2004-May-22 or later
+ for terms of use.
+ If, for some reason, both of these files are missing, the Info-ZIP license
+ also may be found at: ftp://ftp.info-zip.org/pub/infozip/license.html
- Permission is granted to any individual or institution to use, copy,
- or redistribute this software so long as all of the original files are
- included unmodified, that it is not sold for profit, and that this copy-
- right notice is retained.
---------------------------------------------------------------------------
@@ -53,7 +51,7 @@
A set is composed of characters or ranges; a range looks like ``character
hyphen character'' (as in 0-9 or A-Z). [0-9a-zA-Z_] is the minimal set of
- characters allowed in the [..] pattern construct. Other characters are
+ characters ALlowed in the [..] pattern construct. Other characters are
allowed (i.e., 8-bit characters) if your system will support them.
To suppress the special syntactic significance of any of ``[]*?!^-\'', in-
@@ -101,8 +99,32 @@
# define WILDCHAR '?'
# define BEG_RANGE '['
# define END_RANGE ']'
+# define WILDCHR_SINGLE '?'
+# define DIRSEP_CHR '/'
+# define WILDCHR_MULTI '*'
#endif
+#ifdef WILD_STOP_AT_DIR
+ int wild_stop_at_dir = 1; /* default wildcards do not include / in matches */
+#else
+ int wild_stop_at_dir = 0; /* default wildcards do include / in matches */
+#endif
+
+
+
+/*
+ * case mapping functions. case_map is used to ignore case in comparisons,
+ * to_up is used to force upper case even on Unix (for dosify option).
+ */
+#ifdef USE_CASE_MAP
+# define case_map(c) upper[(c) & 0xff]
+# define to_up(c) upper[(c) & 0xff]
+#else
+# define case_map(c) (c)
+# define to_up(c) ((c) >= 'a' && (c) <= 'z' ? (c)-'a'+'A' : (c))
+#endif /* USE_CASE_MAP */
+
+
#if 0 /* GRR: add this to unzip.h someday... */
#if !(defined(MSDOS) && defined(DOSWILD))
#ifdef WILD_STOP_AT_DIR
@@ -114,8 +136,8 @@ int recmatch OF((ZCONST uch *pattern, ZC
int ignore_case __WDLPRO));
#endif
#endif /* 0 */
-static int recmatch OF((ZCONST uch *pattern, ZCONST uch *string,
- int ignore_case __WDLPRO));
+static int recmatch OF((ZCONST char *, ZCONST char *,
+ int));
static char *isshexp OF((ZCONST char *p));
static int namecmp OF((ZCONST char *s1, ZCONST char *s2));
@@ -154,192 +176,240 @@ int match(string, pattern, ignore_case _
}
dospattern[j-1] = '\0'; /* nuke the end "." */
}
- j = recmatch((uch *)dospattern, (uch *)string, ignore_case __WDL);
+ j = recmatch(dospattern, string, ignore_case);
free(dospattern);
return j == 1;
} else
#endif /* MSDOS && DOSWILD */
- return recmatch((uch *)pattern, (uch *)string, ignore_case __WDL) == 1;
+ return recmatch(pattern, string, ignore_case) == 1;
}
+#ifdef _MBCS
+
+char *___tmp_ptr;
+#endif
-static int recmatch(p, s, ic __WDL)
- ZCONST uch *p; /* sh pattern to match */
- ZCONST uch *s; /* string to which to match it */
- int ic; /* true for case insensitivity */
- __WDLDEF /* directory sepchar for WildStopAtDir mode, or 0 */
+static int recmatch(p, s, cs)
+ZCONST char *p; /* sh pattern to match */
+ZCONST char *s; /* string to match it to */
+int cs; /* flag: force case-sensitive matching */
/* Recursively compare the sh pattern p with the string s and return 1 if
- * they match, and 0 or 2 if they don't or if there is a syntax error in the
- * pattern. This routine recurses on itself no more deeply than the number
- * of characters in the pattern. */
+ they match, and 0 or 2 if they don't or if there is a syntax error in the
+ pattern. This routine recurses on itself no deeper than the number of
+ characters in the pattern. */
{
- unsigned int c; /* pattern char or start of range in [-] loop */
+ int c; /* pattern char or start of range in [-] loop */
+ /* Get first character, the pattern for new recmatch calls follows */
+ /* borrowed from Zip's global.c */
+ int no_wild = 0;
+ int allow_regex=1;
+ /* This fix provided by akt@m5.dion.ne.jp for Japanese.
+ See 21 July 2006 mail.
+ It only applies when p is pointing to a doublebyte character and
+ things like / and wildcards are not doublebyte. This probably
+ should not be needed. */
- /* Get first character, the pattern for new recmatch calls follows */
- c = *p; INCSTR(p);
+#ifdef _MBCS
+ if (CLEN(p) == 2) {
+ if (CLEN(s) == 2) {
+ return (*p == *s && *(p+1) == *(s+1)) ?
+ recmatch(p + 2, s + 2, cs) : 0;
+ } else {
+ return 0;
+ }
+ }
+#endif /* ?_MBCS */
- /* If that was the end of the pattern, match if string empty too */
- if (c == 0)
- return *s == 0;
+ c = *POSTINCSTR(p);
- /* '?' (or '%') matches any character (but not an empty string). */
- if (c == WILDCHAR)
-#ifdef WILD_STOP_AT_DIR
- /* If uO.W_flag is non-zero, it won't match '/' */
- return (*s && (!sepc || *s != (uch)sepc))
- ? recmatch(p, s + CLEN(s), ic, sepc) : 0;
-#else
- return *s ? recmatch(p, s + CLEN(s), ic) : 0;
-#endif
+ /* If that was the end of the pattern, match if string empty too */
+ if (c == 0)
+ return *s == 0;
+
+ /* '?' (or '%' or '#') matches any character (but not an empty string) */
+ if (c == WILDCHR_SINGLE) {
+ if (wild_stop_at_dir)
+ return (*s && *s != DIRSEP_CHR) ? recmatch(p, s + CLEN(s), cs) : 0;
+ else
+ return *s ? recmatch(p, s + CLEN(s), cs) : 0;
+ }
- /* '*' matches any number of characters, including zero */
+ /* WILDCHR_MULTI ('*') matches any number of characters, including zero */
#ifdef AMIGA
- if (c == '#' && *p == '?') /* "#?" is Amiga-ese for "*" */
- c = '*', p++;
+ if (!no_wild && c == '#' && *p == '?') /* "#?" is Amiga-ese for "*" */
+ c = WILDCHR_MULTI, p++;
#endif /* AMIGA */
- if (c == '*') {
-#ifdef WILD_STOP_AT_DIR
- if (sepc) {
- /* check for single "*" or double "**" */
-# ifdef AMIGA
- if ((c = p[0]) == '#' && p[1] == '?') /* "#?" is Amiga-ese for "*" */
- c = '*', p++;
- if (c != '*') {
-# else /* !AMIGA */
- if (*p != '*') {
-# endif /* ?AMIGA */
- /* single "*": this doesn't match the dirsep character */
- for (; *s && *s != (uch)sepc; INCSTR(s))
- if ((c = recmatch(p, s, ic, sepc)) != 0)
- return (int)c;
- /* end of pattern: matched if at end of string, else continue */
- if (*p == '\0')
- return (*s == 0);
- /* continue to match if at sepc in pattern, else give up */
- return (*p == (uch)sepc || (*p == '\\' && p[1] == (uch)sepc))
- ? recmatch(p, s, ic, sepc) : 2;
- }
- /* "**": this matches slashes */
- ++p; /* move p behind the second '*' */
- /* and continue with the non-W_flag code variant */
- }
-#endif /* WILD_STOP_AT_DIR */
+ if (!no_wild && c == WILDCHR_MULTI)
+ {
+ if (wild_stop_at_dir) {
+ /* Check for an immediately following WILDCHR_MULTI */
+# ifdef AMIGA
+ if ((c = p[0]) == '#' && p[1] == '?') /* "#?" is Amiga-ese for "*" */
+ c = WILDCHR_MULTI, p++;
+ if (c != WILDCHR_MULTI) {
+# else /* !AMIGA */
+ if (*p != WILDCHR_MULTI) {
+# endif /* ?AMIGA */
+ /* Single WILDCHR_MULTI ('*'): this doesn't match slashes */
+ for (; *s && *s != DIRSEP_CHR; INCSTR(s))
+ if ((c = recmatch(p, s, cs)) != 0)
+ return c;
+ /* end of pattern: matched if at end of string, else continue */
if (*p == 0)
- return 1;
- if (isshexp((ZCONST char *)p) == NULL) {
- /* Optimization for rest of pattern being a literal string:
- * If there are no other shell expression chars in the rest
- * of the pattern behind the multi-char wildcard, then just
- * compare the literal string tail.
- */
- ZCONST uch *srest;
-
- srest = s + (strlen((ZCONST char *)s) - strlen((ZCONST char *)p));
- if (srest - s < 0)
- /* remaining literal string from pattern is longer than rest
- * of test string, there can't be a match
- */
- return 0;
- else
- /* compare the remaining literal pattern string with the last
- * bytes of the test string to check for a match
- */
+ return (*s == 0);
+ /* continue to match if at DIRSEP_CHR in pattern, else give up */
+ return (*p == DIRSEP_CHR || (*p == '\\' && p[1] == DIRSEP_CHR))
+ ? recmatch(p, s, cs) : 2;
+ }
+ /* Two consecutive WILDCHR_MULTI ("**"): this matches DIRSEP_CHR ('/') */
+ p++; /* move p past the second WILDCHR_MULTI */
+ /* continue with the normal non-WILD_STOP_AT_DIR code */
+ } /* wild_stop_at_dir */
+
+ /* Not wild_stop_at_dir */
+ if (*p == 0)
+ return 1;
+ if (!isshexp((char *)p))
+ {
+ /* optimization for rest of pattern being a literal string */
+
+ /* optimization to handle patterns like *.txt */
+ /* if the first char in the pattern is '*' and there */
+ /* are no other shell expression chars, i.e. a literal string */
+ /* then just compare the literal string at the end */
+
+ ZCONST char *srest;
+
+ srest = s + (strlen(s) - strlen(p));
+ if (srest - s < 0)
+ /* remaining literal string from pattern is longer than rest of
+ test string, there can't be a match
+ */
+ return 0;
+ else
+ /* compare the remaining literal pattern string with the last bytes
+ of the test string to check for a match */
#ifdef _MBCS
- {
- ZCONST uch *q = s;
+ {
+ ZCONST char *q = s;
- /* MBCS-aware code must not scan backwards into a string from
- * the end.
- * So, we have to move forward by character from our well-known
- * character position s in the test string until we have
- * advanced to the srest position.
- */
- while (q < srest)
- INCSTR(q);
- /* In case the byte *srest is a trailing byte of a multibyte
- * character in the test string s, we have actually advanced
- * past the position (srest).
- * For this case, the match has failed!
- */
- if (q != srest)
- return 0;
- return ((ic
- ? namecmp((ZCONST char *)p, (ZCONST char *)q)
- : strcmp((ZCONST char *)p, (ZCONST char *)q)
- ) == 0);
- }
+ /* MBCS-aware code must not scan backwards into a string from
+ * the end.
+ * So, we have to move forward by character from our well-known
+ * character position s in the test string until we have advanced
+ * to the srest position.
+ */
+ while (q < srest)
+ INCSTR(q);
+ /* In case the byte *srest is a trailing byte of a multibyte
+ * character, we have actually advanced past the position (srest).
+ * For this case, the match has failed!
+ */
+ if (q != srest)
+ return 0;
+ return ((cs ? strcmp(p, q) : namecmp(p, q)) == 0);
+ }
#else /* !_MBCS */
- return ((ic
- ? namecmp((ZCONST char *)p, (ZCONST char *)srest)
- : strcmp((ZCONST char *)p, (ZCONST char *)srest)
- ) == 0);
+ return ((cs ? strcmp(p, srest) : namecmp(p, srest)) == 0);
#endif /* ?_MBCS */
- } else {
- /* pattern contains more wildcards, continue with recursion... */
- for (; *s; INCSTR(s))
- if ((c = recmatch(p, s, ic __WDL)) != 0)
- return (int)c;
- return 2; /* 2 means give up--match will return false */
- }
}
-
- /* Parse and process the list of characters and ranges in brackets */
- if (c == BEG_RANGE) {
- int e; /* flag true if next char to be taken literally */
- ZCONST uch *q; /* pointer to end of [-] group */
- int r; /* flag true to match anything but the range */
-
- if (*s == 0) /* need a character to match */
- return 0;
- p += (r = (*p == '!' || *p == '^')); /* see if reverse */
- for (q = p, e = 0; *q; INCSTR(q)) /* find closing bracket */
- if (e)
- e = 0;
- else
- if (*q == '\\') /* GRR: change to ^ for MS-DOS, OS/2? */
- e = 1;
- else if (*q == END_RANGE)
- break;
- if (*q != END_RANGE) /* nothing matches if bad syntax */
- return 0;
- for (c = 0, e = (*p == '-'); p < q; INCSTR(p)) {
- /* go through the list */
- if (!e && *p == '\\') /* set escape flag if \ */
- e = 1;
- else if (!e && *p == '-') /* set start of range if - */
- c = *(p-1);
- else {
- unsigned int cc = Case(*s);
-
- if (*(p+1) != '-')
- for (c = c ? c : *p; c <= *p; c++) /* compare range */
- if ((unsigned)Case(c) == cc) /* typecast for MSC bug */
- return r ? 0 : recmatch(q + 1, s + 1, ic __WDL);
- c = e = 0; /* clear range, escape flags */
- }
- }
- return r ? recmatch(q + CLEN(q), s + CLEN(s), ic __WDL) : 0;
- /* bracket match failed */
+ else
+ {
+ /* pattern contains more wildcards, continue with recursion... */
+ for (; *s; INCSTR(s))
+ if ((c = recmatch(p, s, cs)) != 0)
+ return c;
+ return 2; /* 2 means give up--shmatch will return false */
}
+ }
- /* if escape ('\\'), just compare next character */
- if (c == '\\' && (c = *p++) == 0) /* if \ at end, then syntax error */
- return 0;
+#ifndef VMS /* No bracket matching in VMS */
+ /* Parse and process the list of characters and ranges in brackets */
+ if (!no_wild && allow_regex && c == '[')
+ {
+ int e; /* flag true if next char to be taken literally */
+ ZCONST char *q; /* pointer to end of [-] group */
+ int r; /* flag true to match anything but the range */
+
+ if (*s == 0) /* need a character to match */
+ return 0;
+ p += (r = (*p == '!' || *p == '^')); /* see if reverse */
+ for (q = p, e = 0; *q; q++) /* find closing bracket */
+ if (e)
+ e = 0;
+ else
+ if (*q == '\\')
+ e = 1;
+ else if (*q == ']')
+ break;
+ if (*q != ']') /* nothing matches if bad syntax */
+ return 0;
+ for (c = 0, e = *p == '-'; p < q; p++) /* go through the list */
+ {
+ if (e == 0 && *p == '\\') /* set escape flag if \ */
+ e = 1;
+ else if (e == 0 && *p == '-') /* set start of range if - */
+ c = *(p-1);
+ else
+ {
+ uch cc = (cs ? (uch)*s : case_map((uch)*s));
+ uch uc = (uch) c;
+ if (*(p+1) != '-')
+ for (uc = uc ? uc : (uch)*p; uc <= (uch)*p; uc++)
+ /* compare range */
+ if ((cs ? uc : case_map(uc)) == cc)
+ return r ? 0 : recmatch(q + CLEN(q), s + CLEN(s), cs);
+ c = e = 0; /* clear range, escape flags */
+ }
+ }
+ return r ? recmatch(q + CLEN(q), s + CLEN(s), cs) : 0;
+ /* bracket match failed */
+ }
+#endif /* !VMS */
- /* just a character--compare it */
-#ifdef QDOS
- return QMatch(Case((uch)c), Case(*s)) ?
- recmatch(p, s + CLEN(s), ic __WDL) : 0;
-#else
- return Case((uch)c) == Case(*s) ?
- recmatch(p, s + CLEN(s), ic __WDL) : 0;
-#endif
+ /* If escape ('\'), just compare next character */
+ if (!no_wild && c == '\\')
+ if ((c = *p++) == '\0') /* if \ at end, then syntax error */
+ return 0;
+
+#ifdef VMS
+ /* 2005-11-06 SMS.
+ Handle "..." wildcard in p with "." or "]" in s.
+ */
+ if ((c == '.') && (*p == '.') && (*(p+ CLEN( p)) == '.') &&
+ ((*s == '.') || (*s == ']')))
+ {
+ /* Match "...]" with "]". Continue after "]" in both. */
+ if ((*(p+ 2* CLEN( p)) == ']') && (*s == ']'))
+ return recmatch( (p+ 3* CLEN( p)), (s+ CLEN( s)), cs);
+
+ /* Else, look for a reduced match in s, until "]" in or end of s. */
+ for (; *s && (*s != ']'); INCSTR(s))
+ if (*s == '.')
+ /* If reduced match, then continue after "..." in p, "." in s. */
+ if ((c = recmatch( (p+ CLEN( p)), s, cs)) != 0)
+ return (int)c;
+
+ /* Match "...]" with "]". Continue after "]" in both. */
+ if ((*(p+ 2* CLEN( p)) == ']') && (*s == ']'))
+ return recmatch( (p+ 3* CLEN( p)), (s+ CLEN( s)), cs);
+
+ /* No reduced match. Quit. */
+ return 2;
+ }
+
+#endif /* def VMS */
+
+ /* Just a character--compare it */
+ return (cs ? c == *s : case_map((uch)c) == case_map((uch)*s)) ?
+ recmatch(p, s + CLEN(s), cs) : 0;
+}
-} /* end function recmatch() */
+/*************************************************************************************************/
static char *isshexp(p)
ZCONST char *p;
/* If p is a sh expression, a pointer to the first special character is