Module Name: src
Committed By: kre
Date: Mon Dec 3 06:41:30 UTC 2018
Modified Files:
src/bin/sh: expand.c parser.c
Log Message:
Yet another foray into the mysterious world of $@ -- this time
to fix the (unusual) idiom "${1+$@}" (the quotes are part of it).
This seems to have broken about 5 or 6 years ago (somewhere
between -6 and -7), I believe.
Note this is not the same as "$@" and also not the same as ${1+"$@"}
(much more common idioms) which both worked.
Also attempt to deal with "" more correctly, especially when it
appears adjacent to "$@" (or one of the similar constructs.)
This stuff is still all as ugly and hackish (and fragile) as is
possible to imagine, but in an effort to allow some of the weirdness
to eventually go away, the parser output has been made more
regular and all quoted (parts of) words always now start with
CTLQUOTEMARK and end with CTLQUOTEEND regardless of where the
quotes appear.
This allows us to tell the difference between """$@" and "$@"
which was impossible before - yet they are required to generate
different output when there are no args (when "$@" simply vanishes).
Needless to say that change had ramifications all over the place.
To simplify any similar change in the future, there are some new
macros that can generally be used to detect the "noise" data when
processing words, rather than open coding that every time (which
meant that there would *always* be one which missed getting
updated...)
Several other bugs (of my making, and older ones) are also fixed.
The aim is that (aside from anything that is detecting the cases
that were broken before - which were all unlikely uses of sh
syntax) these changes should have no external visible impact.
Sure...
To generate a diff of this commit:
cvs rdiff -u -r1.128 -r1.129 src/bin/sh/expand.c
cvs rdiff -u -r1.156 -r1.157 src/bin/sh/parser.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
Modified files:
Index: src/bin/sh/expand.c
diff -u src/bin/sh/expand.c:1.128 src/bin/sh/expand.c:1.129
--- src/bin/sh/expand.c:1.128 Sun Nov 18 17:23:37 2018
+++ src/bin/sh/expand.c Mon Dec 3 06:41:30 2018
@@ -1,4 +1,4 @@
-/* $NetBSD: expand.c,v 1.128 2018/11/18 17:23:37 kre Exp $ */
+/* $NetBSD: expand.c,v 1.129 2018/12/03 06:41:30 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 1.128 2018/11/18 17:23:37 kre Exp $");
+__RCSID("$NetBSD: expand.c,v 1.129 2018/12/03 06:41:30 kre Exp $");
#endif
#endif /* not lint */
@@ -127,6 +127,16 @@ STATIC void rmescapes_nl(char *);
#define NULLTERM_4_TRACE(p) do { /* nothing */ } while (/*CONSTCOND*/0)
#endif
+#define IS_BORING(_ch) \
+ ((_ch) == CTLQUOTEMARK || (_ch) == CTLQUOTEEND || (_ch) == CTLNONL)
+#define SKIP_BORING(p) \
+ do { \
+ char _ch; \
+ \
+ while ((_ch = *(p)), IS_BORING(_ch)) \
+ (p)++; \
+ } while (0)
+
/*
* Expand shell variables and backquotes inside a here document.
*/
@@ -271,6 +281,8 @@ argstr(const char *p, int flag)
STPUTC('\n', expdest); /* no line_number++ */
break;
case CTLQUOTEEND:
+ if ((flag & EXP_SPLIT) != 0)
+ STPUTC(c, expdest);
ifs_split = EXP_IFS_SPLIT;
break;
case CTLESC:
@@ -284,11 +296,13 @@ argstr(const char *p, int flag)
case CTLVAR: {
#ifdef DEBUG
unsigned int pos = expdest - stackblock();
+ NULLTERM_4_TRACE(expdest);
#endif
p = evalvar(p, (flag & ~EXP_IFS_SPLIT) | (flag & ifs_split));
NULLTERM_4_TRACE(expdest);
VTRACE(DBG_EXPAND, ("argstr evalvar "
- "added \"%s\" to expdest\n",
+ "added %zd \"%s\" to expdest\n",
+ (size_t)(expdest - (stackblock() + pos)),
stackblock() + pos));
break;
}
@@ -810,7 +824,7 @@ subevalvar_trim(const char *p, int strlo
abort();
}
-recordleft:
+ recordleft:
*loc = c;
amount = ((str - 1) - (loc - startp)) - expdest;
STADJUST(amount, expdest);
@@ -818,7 +832,7 @@ recordleft:
*startp++ = *loc++;
return 1;
-recordright:
+ recordright:
amount = loc - expdest;
STADJUST(amount, expdest);
STPUTC('\0', expdest);
@@ -897,6 +911,10 @@ evalvar(const char *p, int flag)
}
}
+ if (!set && subtype != VSPLUS && special && *var == '@')
+ if (startloc > 0 && expdest[-1] == CTLQUOTEMARK)
+ expdest--, startloc--;
+
if (set && subtype != VSPLUS) {
/* insert the value of the variable */
if (special) {
@@ -936,9 +954,7 @@ evalvar(const char *p, int flag)
}
- if (flag & EXP_IN_QUOTES)
- apply_ifs = 0;
- else if (varflags & VSQUOTE) {
+ if (varflags & VSQUOTE) {
if (*var == '@' && shellparam.nparam != 1)
apply_ifs = 1;
else {
@@ -949,6 +965,8 @@ evalvar(const char *p, int flag)
flag |= EXP_IN_QUOTES;
apply_ifs = 0;
}
+ } else if (flag & EXP_IN_QUOTES) {
+ apply_ifs = 0;
} else
apply_ifs = 1;
@@ -1111,6 +1129,12 @@ varvalue(const char *name, int quoted, i
int i;
int sep;
char **ap;
+#ifdef DEBUG
+ char *start = expdest;
+#endif
+
+ VTRACE(DBG_EXPAND, ("varvalue(%c%s, sub=%d, fl=%#x)", *name,
+ quoted ? ", quoted" : "", subtype, flag));
if (subtype == VSLENGTH) /* no magic required ... */
flag &= ~EXP_FULL;
@@ -1147,9 +1171,17 @@ varvalue(const char *name, int quoted, i
if (optlist[optorder[i]].val)
STPUTC(optlist[optorder[i]].letter, expdest);
}
+ VTRACE(DBG_EXPAND, (": %.*s\n", expdest-start, start));
return;
case '@':
if (flag & EXP_SPLIT && quoted) {
+ VTRACE(DBG_EXPAND, (": $@ split (%d)\n",
+ shellparam.nparam));
+ /* GROSS HACK */
+ if (shellparam.nparam == 0 &&
+ expdest[-1] == CTLQUOTEMARK)
+ expdest--;
+ /* KCAH SSORG */
for (ap = shellparam.p ; (p = *ap++) != NULL ; ) {
STRTODEST(p);
if (*ap)
@@ -1175,6 +1207,7 @@ varvalue(const char *name, int quoted, i
&& !quoted && **ap != '\0')
STPUTC('\0', expdest);
}
+ VTRACE(DBG_EXPAND, (": %.*s\n", expdest-start, start));
return;
default:
if (is_digit(*name)) {
@@ -1193,12 +1226,15 @@ varvalue(const char *name, int quoted, i
return;
STRTODEST(p);
}
+ VTRACE(DBG_EXPAND, (": %.*s\n", expdest-start, start));
return;
}
/*
* only the specials with an int value arrive here
*/
+ VTRACE(DBG_EXPAND, ("(%d)", num));
expdest = cvtnum(num, expdest);
+ VTRACE(DBG_EXPAND, (": %.*s\n", expdest-start, start));
}
@@ -1274,7 +1310,7 @@ ifsbreakup(char *string, struct arglist
while (p < string + ifsp->endoff) {
had_param_ch = 1;
q = p;
- if (*p == CTLNONL) {
+ if (IS_BORING(*p)) {
p++;
continue;
}
@@ -1335,6 +1371,9 @@ ifsbreakup(char *string, struct arglist
}
}
+ while (*start == CTLQUOTEEND)
+ start++;
+
/*
* Save anything left as an argument.
* Traditionally we have treated 'IFS=':'; set -- x$IFS' as
@@ -1391,7 +1430,7 @@ expandmeta(struct strlist *str, int flag
for (;;) { /* fast check for meta chars */
if ((c = *p++) == '\0')
goto nometa;
- if (c == '*' || c == '?' || c == '[' || c == '!')
+ if (c == '*' || c == '?' || c == '[' /* || c == '!' */)
break;
}
savelastp = exparg.lastp;
@@ -1466,7 +1505,7 @@ expmeta(char *enddir, char *name)
if (*q == '!' || *q == '^')
q++;
for (;;) {
- while (*q == CTLQUOTEMARK || *q == CTLNONL)
+ while (IS_BORING(*q))
q++;
if (*q == ']') {
q++;
@@ -1518,7 +1557,7 @@ expmeta(char *enddir, char *name)
}
} else if (*p == '\0')
break;
- else if (*p == CTLQUOTEMARK || *p == CTLNONL)
+ else if (IS_BORING(*p))
continue;
else if (*p == CTLESC)
p++;
@@ -1532,7 +1571,7 @@ expmeta(char *enddir, char *name)
if (enddir != expdir)
metaflag++;
for (p = name ; ; p++) {
- if (*p == CTLQUOTEMARK || *p == CTLNONL)
+ if (IS_BORING(*p))
continue;
if (*p == CTLESC)
p++;
@@ -1548,7 +1587,7 @@ expmeta(char *enddir, char *name)
if (start != name) {
p = name;
while (p < start) {
- while (*p == CTLQUOTEMARK || *p == CTLNONL)
+ while (IS_BORING(*p))
p++;
if (*p == CTLESC)
p++;
@@ -1575,7 +1614,7 @@ expmeta(char *enddir, char *name)
}
matchdot = 0;
p = start;
- while (*p == CTLQUOTEMARK || *p == CTLNONL)
+ while (IS_BORING(*p))
p++;
if (*p == CTLESC)
p++;
@@ -1767,6 +1806,7 @@ patmatch(const char *pattern, const char
goto backtrack;
break;
case CTLQUOTEMARK:
+ case CTLQUOTEEND:
case CTLNONL:
continue;
case '?':
@@ -1781,7 +1821,7 @@ patmatch(const char *pattern, const char
c = *p;
while (c == CTLQUOTEMARK || c == '*')
c = *++p;
- if (c != CTLESC && c != CTLQUOTEMARK && c != CTLNONL &&
+ if (c != CTLESC && !IS_BORING(c) &&
c != '?' && c != '*' && c != '[') {
while (*q != c) {
if (squoted && *q == CTLESC &&
@@ -1828,13 +1868,11 @@ patmatch(const char *pattern, const char
if (*endp == '!' || *endp == '^')
endp++;
for (;;) {
- while (*endp == CTLQUOTEMARK || *endp==CTLNONL)
+ while (IS_BORING(*endp))
endp++;
if (*endp == '\0')
goto dft; /* no matching ] */
- if (*endp == CTLESC)
- endp++;
- if (*++endp == ']')
+ if (*endp++ == ']')
break;
}
/* end shortcut */
@@ -1856,7 +1894,7 @@ patmatch(const char *pattern, const char
chr = (unsigned char)*q++;
c = *p++;
do {
- if (c == CTLQUOTEMARK || c == CTLNONL)
+ if (IS_BORING(c))
continue;
if (c == '\0') {
p = savep, q = saveq;
@@ -1930,13 +1968,13 @@ rmescapes(char *str)
char *p, *q;
p = str;
- while (*p != CTLESC && *p != CTLQUOTEMARK && *p != CTLNONL) {
+ while (*p != CTLESC && !IS_BORING(*p)) {
if (*p++ == '\0')
return;
}
q = p;
while (*p) {
- if (*p == CTLQUOTEMARK || *p == CTLNONL) {
+ if (IS_BORING(*p)) {
p++;
continue;
}
@@ -1969,7 +2007,7 @@ rmescapes_nl(char *str)
int nls = 0, holdnl = 0, holdlast;
p = str;
- while (*p != CTLESC && *p != CTLQUOTEMARK && *p != CTLNONL) {
+ while (*p != CTLESC && !IS_BORING(*p)) {
if (*p++ == '\0')
return;
}
@@ -1977,7 +2015,7 @@ rmescapes_nl(char *str)
--p; /* so we do not place a \n badly */
q = p;
while (*p) {
- if (*p == CTLQUOTEMARK) {
+ if (*p == CTLQUOTEMARK || *p == CTLQUOTEEND) {
p++;
continue;
}
Index: src/bin/sh/parser.c
diff -u src/bin/sh/parser.c:1.156 src/bin/sh/parser.c:1.157
--- src/bin/sh/parser.c:1.156 Mon Dec 3 06:40:26 2018
+++ src/bin/sh/parser.c Mon Dec 3 06:41:30 2018
@@ -1,4 +1,4 @@
-/* $NetBSD: parser.c,v 1.156 2018/12/03 06:40:26 kre Exp $ */
+/* $NetBSD: parser.c,v 1.157 2018/12/03 06:41:30 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 1.156 2018/12/03 06:40:26 kre Exp $");
+__RCSID("$NetBSD: parser.c,v 1.157 2018/12/03 06:41:30 kre Exp $");
#endif
#endif /* not lint */
@@ -1834,10 +1834,8 @@ readtoken1(int firstc, char const *syn,
if (NEEDESC(c))
USTPUTC(CTLESC, out);
else if (!magicq) {
- USTPUTC(CTLQUOTEMARK, out);
+ USTPUTC(CTLESC, out);
USTPUTC(c, out);
- if (varnest != 0)
- USTPUTC(CTLQUOTEEND, out);
continue;
}
USTPUTC(c, out);
@@ -1859,7 +1857,7 @@ readtoken1(int firstc, char const *syn,
}
/* End of single quotes... */
TS_POP();
- if (syntax == BASESYNTAX && varnest != 0)
+ if (syntax == BASESYNTAX)
USTPUTC(CTLQUOTEEND, out);
continue;
case CDQUOTE:
@@ -1871,6 +1869,7 @@ readtoken1(int firstc, char const *syn,
quotef = 1;
if (arinest) {
if (ISDBLQUOTE()) {
+ USTPUTC(CTLQUOTEEND, out);
TS_POP();
} else {
TS_PUSH();
@@ -1884,8 +1883,7 @@ readtoken1(int firstc, char const *syn,
continue;
if (ISDBLQUOTE()) {
TS_POP();
- if (varnest != 0)
- USTPUTC(CTLQUOTEEND, out);
+ USTPUTC(CTLQUOTEEND, out);
} else {
TS_PUSH();
syntax = DQSYNTAX;
@@ -2137,7 +2135,7 @@ parsesub: {
synerror("no modifiers allowed with ${#var}");
pungetc();
}
- if (ISDBLQUOTE() || arinest)
+ if (quoted || arinest)
flags |= VSQUOTE;
if (subtype >= VSTRIMLEFT && subtype <= VSTRIMRIGHTMAX)
flags |= VSPATQ;
@@ -2148,7 +2146,7 @@ parsesub: {
arinest = 0;
if (subtype > VSASSIGN) { /* # ## % %% */
syntax = BASESYNTAX;
- CLRDBLQUOTE();
+ quoted = 0;
}
}
} else if (c == '\'' && syntax == BASESYNTAX) {
@@ -2232,7 +2230,7 @@ noexpand(char *text)
p = text;
while ((c = *p++) != '\0') {
- if (c == CTLQUOTEMARK)
+ if (c == CTLQUOTEMARK || c == CTLQUOTEEND)
continue;
if (c == CTLESC)
p++;
