First pass at fixing some of the more arcane pattern matching
possibilities that we do not currently handle all that well. This mostly means (for now) making sure that quoted pattern magic characters (as well as quoted sh syntax magic chars) are properly marked, so they remain known as being quoted, and do not turn into pattern magic. Also, make sure that an unquoted \ in a pattern always quotes whatever comes next (which, unlike in regular expressions, includes inside [] matches),
This commit is contained in:
kre
parent
da66a31a53
commit
c83568a7dc
|
|
@ -1,4 +1,4 @@
|
|||
/* $NetBSD: expand.c,v 1.123 2018/06/22 18:19:41 kre Exp $ */
|
||||
/* $NetBSD: expand.c,v 1.124 2018/07/20 22:47:26 kre Exp $ */
|
||||
|
||||
/*-
|
||||
* Copyright (c) 1991, 1993
|
||||
|
|
@ -37,7 +37,7 @@
|
|||
#if 0
|
||||
static char sccsid[] = "@(#)expand.c 8.5 (Berkeley) 5/15/95";
|
||||
#else
|
||||
__RCSID("$NetBSD: expand.c,v 1.123 2018/06/22 18:19:41 kre Exp $");
|
||||
__RCSID("$NetBSD: expand.c,v 1.124 2018/07/20 22:47:26 kre Exp $");
|
||||
#endif
|
||||
#endif /* not lint */
|
||||
|
||||
|
|
@ -1109,7 +1109,7 @@ varvalue(const char *name, int quoted, int subtype, int flag)
|
|||
int num;
|
||||
char *p;
|
||||
int i;
|
||||
char sep;
|
||||
int sep;
|
||||
char **ap;
|
||||
char const *syntax;
|
||||
|
||||
|
|
@ -1167,10 +1167,14 @@ varvalue(const char *name, int quoted, int subtype, int flag)
|
|||
STRTODEST(p);
|
||||
if (!*ap)
|
||||
break;
|
||||
if (sep)
|
||||
if (sep) {
|
||||
if (quoted && (flag & (EXP_GLOB|EXP_CASE)) &&
|
||||
(SQSYNTAX[sep] == CCTL || SQSYNTAX[sep] == CSBACK))
|
||||
STPUTC(CTLESC, expdest);
|
||||
STPUTC(sep, expdest);
|
||||
else if ((flag & (EXP_SPLIT|EXP_IN_QUOTES)) == EXP_SPLIT
|
||||
&& !quoted && **ap != '\0')
|
||||
} else
|
||||
if ((flag & (EXP_SPLIT|EXP_IN_QUOTES)) == EXP_SPLIT
|
||||
&& !quoted && **ap != '\0')
|
||||
STPUTC('\0', expdest);
|
||||
}
|
||||
return;
|
||||
|
|
@ -1749,19 +1753,31 @@ patmatch(const char *pattern, const char *string, int squoted)
|
|||
int invert, found;
|
||||
unsigned char chr;
|
||||
|
||||
/*
|
||||
* First quick check to see if there is a
|
||||
* possible matching ']' - if not, then this
|
||||
* is not a char class, and the '[' is just
|
||||
* a literal '['.
|
||||
*
|
||||
* This check will not detect all non classes, but
|
||||
* that's OK - It just means that we execute the
|
||||
* harder code sometimes when it it cannot succeed.
|
||||
*/
|
||||
endp = p;
|
||||
if (*endp == '!')
|
||||
if (*endp == '!' || *endp == '^')
|
||||
endp++;
|
||||
for (;;) {
|
||||
while (*endp == CTLQUOTEMARK || *endp==CTLNONL)
|
||||
endp++;
|
||||
if (*endp == '\0')
|
||||
goto dft; /* no matching ] */
|
||||
goto dft; /* no matching ] */
|
||||
if (*endp == CTLESC)
|
||||
endp++;
|
||||
if (*++endp == ']')
|
||||
break;
|
||||
}
|
||||
/* end shortcut */
|
||||
|
||||
invert = 0;
|
||||
savep = p, saveq = q;
|
||||
invert = 0;
|
||||
|
|
@ -1789,12 +1805,12 @@ patmatch(const char *pattern, const char *string, int squoted)
|
|||
continue;
|
||||
}
|
||||
}
|
||||
if (c == CTLESC)
|
||||
if (c == CTLESC || c == '\\')
|
||||
c = *p++;
|
||||
wc = (unsigned char)c;
|
||||
if (*p == '-' && p[1] != ']') {
|
||||
p++;
|
||||
if (*p == CTLESC)
|
||||
if (*p == CTLESC || *p == '\\')
|
||||
p++;
|
||||
wc2 = (unsigned char)*p++;
|
||||
if ( collate_range_cmp(chr, wc) >= 0
|
||||
|
|
|
|||
|
|
@ -1,4 +1,4 @@
|
|||
/* $NetBSD: parser.c,v 1.147 2018/07/13 22:43:44 kre Exp $ */
|
||||
/* $NetBSD: parser.c,v 1.148 2018/07/20 22:47:26 kre Exp $ */
|
||||
|
||||
/*-
|
||||
* Copyright (c) 1991, 1993
|
||||
|
|
@ -37,7 +37,7 @@
|
|||
#if 0
|
||||
static char sccsid[] = "@(#)parser.c 8.7 (Berkeley) 5/16/95";
|
||||
#else
|
||||
__RCSID("$NetBSD: parser.c,v 1.147 2018/07/13 22:43:44 kre Exp $");
|
||||
__RCSID("$NetBSD: parser.c,v 1.148 2018/07/20 22:47:26 kre Exp $");
|
||||
#endif
|
||||
#endif /* not lint */
|
||||
|
||||
|
|
@ -1770,7 +1770,7 @@ readtoken1(int firstc, char const *syn, int magicq)
|
|||
for (c = firstc ;; c = pgetc_macro()) { /* until of token */
|
||||
if (syntax == ARISYNTAX)
|
||||
out = insert_elided_nl(out);
|
||||
CHECKSTRSPACE(4, out); /* permit 4 calls to USTPUTC */
|
||||
CHECKSTRSPACE(6, out); /* permit 6 calls to USTPUTC */
|
||||
switch (syntax[c]) {
|
||||
case CNL: /* '\n' */
|
||||
if (syntax == BASESYNTAX && varnest == 0)
|
||||
|
|
@ -1788,6 +1788,7 @@ readtoken1(int firstc, char const *syn, int magicq)
|
|||
out = readcstyleesc(out);
|
||||
continue;
|
||||
}
|
||||
USTPUTC(CTLESC, out);
|
||||
/* FALLTHROUGH */
|
||||
case CWORD:
|
||||
USTPUTC(c, out);
|
||||
|
|
@ -1818,7 +1819,7 @@ readtoken1(int firstc, char const *syn, int magicq)
|
|||
if (ISDBLQUOTE() && c != '\\' && c != '`' &&
|
||||
c != '$' && (c != '"' || magicq))
|
||||
USTPUTC('\\', out);
|
||||
if (SQSYNTAX[c] == CCTL)
|
||||
if (SQSYNTAX[c] == CCTL || SQSYNTAX[c] == CSBACK)
|
||||
USTPUTC(CTLESC, out);
|
||||
else if (!magicq) {
|
||||
USTPUTC(CTLQUOTEMARK, out);
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
/* $NetBSD: syntax.c,v 1.5 2017/08/21 13:20:49 kre Exp $ */
|
||||
/* $NetBSD: syntax.c,v 1.6 2018/07/20 22:47:26 kre Exp $ */
|
||||
|
||||
#include <sys/cdefs.h>
|
||||
__RCSID("$NetBSD: syntax.c,v 1.5 2017/08/21 13:20:49 kre Exp $");
|
||||
__RCSID("$NetBSD: syntax.c,v 1.6 2018/07/20 22:47:26 kre Exp $");
|
||||
|
||||
#include <limits.h>
|
||||
#include "shell.h"
|
||||
|
|
@ -46,7 +46,7 @@ const char dqsyntax[257] = { CEOF,
|
|||
set('`', CBQUOTE)
|
||||
set('$', CVAR)
|
||||
set('}', CENDVAR)
|
||||
/* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
|
||||
/* ':/' for tilde expansion, '-]' for [a\-x] pattern ranges */
|
||||
set('!', CCTL)
|
||||
set('*', CCTL)
|
||||
set('?', CCTL)
|
||||
|
|
@ -56,6 +56,7 @@ const char dqsyntax[257] = { CEOF,
|
|||
set(':', CCTL)
|
||||
set('/', CCTL)
|
||||
set('-', CCTL)
|
||||
set(']', CCTL)
|
||||
};
|
||||
|
||||
/* syntax table used when in single quotes */
|
||||
|
|
@ -64,7 +65,7 @@ const char sqsyntax[257] = { CEOF,
|
|||
set('\n', CNL)
|
||||
set('\'', CSQUOTE)
|
||||
set('\\', CSBACK)
|
||||
/* ':/' for tilde expansion, '-' for [a\-x] pattern ranges */
|
||||
/* ':/' for tilde expansion, '-]' for [a\-x] pattern ranges */
|
||||
set('!', CCTL)
|
||||
set('*', CCTL)
|
||||
set('?', CCTL)
|
||||
|
|
@ -74,6 +75,7 @@ const char sqsyntax[257] = { CEOF,
|
|||
set(':', CCTL)
|
||||
set('/', CCTL)
|
||||
set('-', CCTL)
|
||||
set(']', CCTL)
|
||||
};
|
||||
|
||||
/* syntax table used when in arithmetic */
|
||||
|
|
|
|||
Loading…
Reference in New Issue