In re.c, change a bad assertion into code.HEAD master

author: Arnold D. Robbins <[email protected]> 2025-10-29 10:24:33 +0200
committer: Arnold D. Robbins <[email protected]> 2025-10-29 10:24:33 +0200
commit: a0879473da5f65b1687ed5a4af565bf400f51f88 (patch)
tree: b89e2ab5e12c82f3b043d00060105379d18a7e08
parent: 3992eafcead82df31a8daea5d25803c3e9f5e726 (diff)
download: gawk-master.tar.gz
12 files changed, 271 insertions, 7 deletions
diff --git a/ChangeLog b/ChangeLog
index c2064ccb..c032e46c 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,12 @@
+2025-10-29         Arnold D. Robbins     <[email protected]>
+
+	* re.c (make_regexp): Remove assertion that string is zero
+	terminated. It isn't always. Instead, save and restore the
+	character past the value of `len'. This required making the
+	first parameter not be const anymore. Thanks to Ben Hoyt
+	<[email protected]> for the report.
+	* awk.k (make_regexp): Adjust declaration.
+
 2025-10-28         Arnold D. Robbins     <[email protected]>
 
 	* awkgram.y: Remove never-documented support for tawk extension
diff --git a/awk.h b/awk.h
index b7f6758f..846177d1 100644
--- a/awk.h
+++ b/awk.h
@@ -1789,7 +1789,7 @@ extern bool out_of_range(NODE *n);
 extern char *format_nan_inf(NODE *n, char format);
 extern bool is_ieee_magic_val(const char *val);
 /* re.c */
-extern Regexp *make_regexp(const char *s, size_t len, bool ignorecase, bool dfa, bool canfatal);
+extern Regexp *make_regexp(char *s, size_t len, bool ignorecase, bool dfa, bool canfatal);
 extern int research(Regexp *rp, char *str, int start, size_t len, int flags);
 extern void refree(Regexp *rp);
 extern void reg_error(const char *s);
diff --git a/pc/ChangeLog b/pc/ChangeLog
index faa2d1a1..918ac82a 100644
--- a/pc/ChangeLog
+++ b/pc/ChangeLog
@@ -1,3 +1,7 @@
+2025-10-29         Arnold D. Robbins     <[email protected]>
+
+	* Makefile.tst: Regenerated.
+
 2025-10-27         Arnold D. Robbins     <[email protected]>
 
 	* Makefile.tst: Regenerated.
diff --git a/pc/Makefile.tst b/pc/Makefile.tst
index cdce2270..01d8ea40 100644
--- a/pc/Makefile.tst
+++ b/pc/Makefile.tst
@@ -165,7 +165,7 @@ BASIC_TESTS = \
 	inpref inputred intest intprec iobug1 \
 	leaddig leadnl litoct longsub longwrds \
 	manglprm match4 matchuninitialized math membug1 memleak messages \
-	matchbadarg1 matchbadarg2 \
+	matchbadarg1 matchbadarg2 match5 \
 	minusstr mmap8k \
 	nasty nasty2 negexp negrange nested nfldstr nfloop nfneg nfset \
 	nlfldsep nlinstr nlstrina noeffect nofile nofmtch noloop1 \
@@ -1969,6 +1969,11 @@ matchbadarg2:
 	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  < "$(srcdir)"/[email protected] >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/[email protected] _$@ && rm -f _$@
 
+match5:
+	@echo $@
+	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  < "$(srcdir)"/[email protected] >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) "$(srcdir)"/[email protected] _$@ && rm -f _$@
+
 minusstr:
 	@echo $@
 	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
diff --git a/re.c b/re.c
index 21b80346..f54dc244 100644
--- a/re.c
+++ b/re.c
@@ -37,7 +37,7 @@ static struct localeinfo localeinfo;
 /* make_regexp --- generate compiled regular expressions */
 
 Regexp *
-make_regexp(const char *s, size_t len, bool ignorecase, bool dfa, bool canfatal)
+make_regexp(char *s, size_t len, bool ignorecase, bool dfa, bool canfatal)
 {
 	static char metas[] = ".*+(){}[]|?^$\\";
 	Regexp *rp;
@@ -53,8 +53,22 @@ make_regexp(const char *s, size_t len, bool ignorecase, bool dfa, bool canfatal)
 	int i;
 	static struct dfa* dfaregs[2] = { NULL, NULL };
 	static bool nul_warned = false;
+	char save;
+	size_t savelen;
 
-	assert(s[len] == '\0');
+	/*
+	 * 10/2025: We used to have:
+	 *
+	 *	assert(s[len] == '\0');
+	 *
+	 * here, but data can come in, by way of re_update(), that is from $0 or
+	 * elsewhere where there is no final '\0'. So we save and restore
+	 * the character at s[len] and force a '\0' into position there.
+	 * It needs to be a C string for use in error messages.
+	 */
+	savelen = len;
+	save = s[len];
+	s[len] = '\0';
 
 	if (do_lint && ! nul_warned && memchr(s, '\0', len) != NULL) {
 		nul_warned = true;
@@ -275,7 +289,7 @@ make_regexp(const char *s, size_t len, bool ignorecase, bool dfa, bool canfatal)
 	 * character sets only.
 	 *
 	 * On the other hand, if we do have a single-byte character set,
-	 * using the casetable should give  a performance improvement, since
+	 * using the casetable should give a performance improvement, since
 	 * it's computed only once, not each time a regex is compiled.  We
 	 * also think it's probably better for portability.  See the
 	 * discussion by the definition of casetable[] in eval.c.
@@ -311,6 +325,7 @@ make_regexp(const char *s, size_t len, bool ignorecase, bool dfa, bool canfatal)
 		if (! canfatal) {
 			/* rerr already gettextized inside regex routines */
 			error("%s: /%s/", rerr, s);
+			s[savelen] = save;
  			return NULL;
 		}
 		fatal("invalid regexp: %s: /%s/", rerr, s);
@@ -340,6 +355,7 @@ make_regexp(const char *s, size_t len, bool ignorecase, bool dfa, bool canfatal)
 		}
 	}
 
+	s[savelen] = save;
 	return rp;
 }
 
diff --git a/test/ChangeLog b/test/ChangeLog
index b341b3ba..e7a8d773 100644
--- a/test/ChangeLog
+++ b/test/ChangeLog
@@ -1,3 +1,8 @@
+2025-10-29         Arnold D. Robbins     <[email protected]>
+
+	* Makefile.am (EXTRADIST): New test, match5.
+	* match5.awk, match5.in, match5.ok: New files.
+
 2025-10-28         Arnold D. Robbins     <[email protected]>
 
 	* lintwarn.awk, lintwarn.ok: Adjust after code changes.
diff --git a/test/Makefile.am b/test/Makefile.am
index 3f72d117..d6b8a8c3 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -792,6 +792,9 @@ EXTRA_DIST = \
 	match3.ok \
 	match4.awk \
 	match4.ok \
+	match5.awk \
+	match5.in \
+	match5.ok \
 	matchbadarg1.awk \
 	matchbadarg1.in \
 	matchbadarg1.ok \
@@ -1581,7 +1584,7 @@ BASIC_TESTS = \
 	inpref inputred intest intprec iobug1 \
 	leaddig leadnl litoct longsub longwrds \
 	manglprm match4 matchuninitialized math membug1 memleak messages \
-	matchbadarg1 matchbadarg2 \
+	matchbadarg1 matchbadarg2 match5 \
 	minusstr mmap8k \
 	nasty nasty2 negexp negrange nested nfldstr nfloop nfneg nfset \
 	nlfldsep nlinstr nlstrina noeffect nofile nofmtch noloop1 \
diff --git a/test/Makefile.in b/test/Makefile.in
index a5793aae..a6bc128e 100644
--- a/test/Makefile.in
+++ b/test/Makefile.in
@@ -1056,6 +1056,9 @@ EXTRA_DIST = \
 	match3.ok \
 	match4.awk \
 	match4.ok \
+	match5.awk \
+	match5.in \
+	match5.ok \
 	matchbadarg1.awk \
 	matchbadarg1.in \
 	matchbadarg1.ok \
@@ -1845,7 +1848,7 @@ BASIC_TESTS = \
 	inpref inputred intest intprec iobug1 \
 	leaddig leadnl litoct longsub longwrds \
 	manglprm match4 matchuninitialized math membug1 memleak messages \
-	matchbadarg1 matchbadarg2 \
+	matchbadarg1 matchbadarg2 match5 \
 	minusstr mmap8k \
 	nasty nasty2 negexp negrange nested nfldstr nfloop nfneg nfset \
 	nlfldsep nlinstr nlstrina noeffect nofile nofmtch noloop1 \
@@ -3836,6 +3839,11 @@ matchbadarg2:
 	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  < "$(srcdir)"/[email protected] >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/[email protected] _$@ && rm -f _$@
 
+match5:
+	@echo $@
+	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  < "$(srcdir)"/[email protected] >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) "$(srcdir)"/[email protected] _$@ && rm -f _$@
+
 minusstr:
 	@echo $@
 	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
diff --git a/test/Maketests b/test/Maketests
index a782659e..276eccdb 100644
--- a/test/Maketests
+++ b/test/Maketests
@@ -629,6 +629,11 @@ matchbadarg2:
 	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  < "$(srcdir)"/[email protected] >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) "$(srcdir)"/[email protected] _$@ && rm -f _$@
 
+match5:
+	@echo $@
+	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  < "$(srcdir)"/[email protected] >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) "$(srcdir)"/[email protected] _$@ && rm -f _$@
+
 minusstr:
 	@echo $@
 	@-AWKPATH="$(srcdir)" $(AWK) -f [email protected]  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
diff --git a/test/match5.awk b/test/match5.awk
new file mode 100644
index 00000000..0b238a29
--- /dev/null
+++ b/test/match5.awk
@@ -0,0 +1,6 @@
+NF > 0 && match($NF, $1) {
+	print $0, RSTART, RLENGTH
+	if (RLENGTH != length($1))
+		printf "match error at %d: %d %d\n",
+			NR, RLENGTH, RSTART >"/dev/tty"
+}
diff --git a/test/match5.in b/test/match5.in
new file mode 100644
index 00000000..be3dbf61
--- /dev/null
+++ b/test/match5.in
@@ -0,0 +1,199 @@
+/dev/rrp3:
+
+17379	mel
+16693	bwk	me
+16116	ken	him	someone else
+15713	srb
+11895	lem
+10409	scj
+10252	rhm
+ 9853	shen
+ 9748	a68
+ 9492	sif
+ 9190	pjw
+ 8912	nls
+ 8895	dmr
+ 8491	cda
+ 8372	bs
+ 8252	llc
+ 7450	mb
+ 7360	ava
+ 7273	jrv
+ 7080	bin
+ 7063	greg
+ 6567	dict
+ 6462	lck
+ 6291	rje
+ 6211	lwf
+ 5671	dave
+ 5373	jhc
+ 5220	agf
+ 5167	doug
+ 5007	valerie
+ 3963	jca
+ 3895	bbs
+ 3796	moh
+ 3481	xchar
+ 3200	tbl
+ 2845	s
+ 2774	tgs
+ 2641	met
+ 2566	jck
+ 2511	port
+ 2479	sue
+ 2127	root
+ 1989	bsb
+ 1989	jeg
+ 1933	eag
+ 1801	pdj
+ 1590	tpc
+ 1385	cvw
+ 1370	rwm
+ 1316	avg
+ 1205	eg
+ 1194	jam
+ 1153	dl
+ 1150	lgm
+ 1031	cmb
+ 1018	jwr
+  950	gdb
+  931	marc
+  898	usg
+  865	ggr
+  822	daemon
+  803	mihalis
+  700	honey
+  624	tad
+  559	acs
+  541	uucp
+  523	raf
+  495	adh
+  456	kec
+  414	craig
+  386	donmac
+  375	jj
+  348	ravi
+  344	drw
+  327	stars
+  288	mrg
+  272	jcb
+  263	ralph
+  253	tom
+  251	sjb
+  248	haight
+  224	sharon
+  222	chuck
+  213	dsj
+  201	bill
+  184	god
+  176	sys
+  166	meh
+  163	jon
+  144	dan
+  143	fox
+  123	dale
+  116	kab
+   95	buz
+   80	asc
+   79	jas
+   79	trt
+   64	wsb
+   62	dwh
+   56	ktf
+   54	lr
+   47	dlc
+   45	dls
+   45	jwf
+   44	mash
+   43	ars
+   43	vgl
+   37	jfo
+   32	rab
+   31	pd
+   29	jns
+   25	spm
+   22	rob
+   15	egb
+   10	hm
+   10	mhb
+    6	aed
+    6	cpb
+    5	evp
+    4	ber
+    4	men
+    4	mitch
+    3	ast
+    3	jfr
+    3	lax
+    3	nel
+    2	blue
+    2	jfk
+    2	njas
+    1	122sec
+    1	ddwar
+    1	gopi
+    1	jk
+    1	learn
+    1	low
+    1	nac
+    1	sidor
+1root:EMpNB8Zp56:0:0:Super-User,,,,,,,:/:/bin/sh
+2roottcsh:*:0:0:Super-User running tcsh [cbm]:/:/bin/tcsh
+3sysadm:*:0:0:System V Administration:/usr/admin:/bin/sh
+4diag:*:0:996:Hardware Diagnostics:/usr/diags:/bin/csh
+5daemon:*:1:1:daemons:/:/bin/sh
+6bin:*:2:2:System Tools Owner:/bin:/dev/null
+7nuucp:BJnuQbAo:6:10:UUCP.Admin:/usr/spool/uucppublic:/usr/lib/uucp/uucico
+8uucp:*:3:5:UUCP.Admin:/usr/lib/uucp:
+9sys:*:4:0:System Activity Owner:/usr/adm:/bin/sh
+10adm:*:5:3:Accounting Files Owner:/usr/adm:/bin/sh
+11lp:*:9:9:Print Spooler Owner:/var/spool/lp:/bin/sh
+12auditor:*:11:0:Audit Activity Owner:/auditor:/bin/sh
+13dbadmin:*:12:0:Security Database Owner:/dbadmin:/bin/sh
+14bootes:dcon:50:1:Tom Killian (DO NOT REMOVE):/tmp:
+15cdjuke:dcon:51:1:Tom Killian (DO NOT REMOVE):/tmp:
+16rfindd:*:66:1:Rfind Daemon and Fsdump:/var/rfindd:/bin/sh
+17EZsetup:*:992:998:System Setup:/var/sysadmdesktop/EZsetup:/bin/csh
+18demos:*:993:997:Demonstration User:/usr/demos:/bin/csh
+19tutor:*:994:997:Tutorial User:/usr/tutor:/bin/csh
+20tour:*:995:997:IRIS Space Tour:/usr/people/tour:/bin/csh
+21guest:nfP4/Wpvio/Rw:998:998:Guest Account:/usr/people/guest:/bin/csh
+224Dgifts:0nWRTZsOMt.:999:998:4Dgifts Account:/usr/people/4Dgifts:/bin/csh
+23nobody:*:60001:60001:SVR4 nobody uid:/dev/null:/dev/null
+24noaccess:*:60002:60002:uid no access:/dev/null:/dev/null
+25nobody:*:-2:-2:original nobody uid:/dev/null:/dev/null
+26rje:*:8:8:RJE Owner:/usr/spool/rje:
+27changes:*:11:11:system change log:/:
+28dist:sorry:9999:4:file distributions:/v/adm/dist:/v/bin/sh
+29man:*:99:995:On-line Manual Owner:/:
+30phoneca:*:991:991:phone call log [tom]:/v/adm/log:/v/bin/sh
+1r oot EMpNB8Zp56 0 0 Super-User,,,,,,, / /bin/sh
+2r oottcsh * 0 0 Super-User running tcsh [cbm] / /bin/tcsh
+3s ysadm * 0 0 System V Administration /usr/admin /bin/sh
+4d iag * 0 996 Hardware Diagnostics /usr/diags /bin/csh
+5d aemon * 1 1 daemons / /bin/sh
+6b in * 2 2 System Tools Owner /bin /dev/null
+7n uucp BJnuQbAo 6 10 UUCP.Admin /usr/spool/uucppublic /usr/lib/uucp/uucico
+8u ucp * 3 5 UUCP.Admin /usr/lib/uucp 
+9s ys * 4 0 System Activity Owner /usr/adm /bin/sh
+10 adm * 5 3 Accounting Files Owner /usr/adm /bin/sh
+11 lp * 9 9 Print Spooler Owner /var/spool/lp /bin/sh
+12 auditor * 11 0 Audit Activity Owner /auditor /bin/sh
+13 dbadmin * 12 0 Security Database Owner /dbadmin /bin/sh
+14 bootes dcon 50 1 Tom Killian (DO NOT REMOVE) /tmp 
+15 cdjuke dcon 51 1 Tom Killian (DO NOT REMOVE) /tmp 
+16 rfindd * 66 1 Rfind Daemon and Fsdump /var/rfindd /bin/sh
+17 EZsetup * 992 998 System Setup /var/sysadmdesktop/EZsetup /bin/csh
+18 demos * 993 997 Demonstration User /usr/demos /bin/csh
+19 tutor * 994 997 Tutorial User /usr/tutor /bin/csh
+20 tour * 995 997 IRIS Space Tour /usr/people/tour /bin/csh
+21 guest nfP4/Wpvio/Rw 998 998 Guest Account /usr/people/guest /bin/csh
+22 4Dgifts 0nWRTZsOMt. 999 998 4Dgifts Account /usr/people/4Dgifts /bin/csh
+23 nobody * 60001 60001 SVR4 nobody uid /dev/null /dev/null
+24 noaccess * 60002 60002 uid no access /dev/null /dev/null
+25 nobody * -2 -2 original nobody uid /dev/null /dev/null
+26 rje * 8 8 RJE Owner /usr/spool/rje 
+27 changes * 11 11 system change log / 
+28 dist sorry 9999 4 file distributions /v/adm/dist /v/bin/sh
+29 man * 99 995 On-line Manual Owner / 
+30 phoneca * 991 991 phone call log [tom] /v/adm/log /v/bin/sh
diff --git a/test/match5.ok b/test/match5.ok
new file mode 100644
index 00000000..8b0731d9
--- /dev/null
+++ b/test/match5.ok
@@ -0,0 +1,4 @@
+/dev/rrp3: 1 10
+    1	122sec 1 1
+1root:EMpNB8Zp56:0:0:Super-User,,,,,,,:/:/bin/sh 1 48
+7nuucp:BJnuQbAo:6:10:UUCP.Admin:/usr/spool/uucppublic:/usr/lib/uucp/uucico 1 74
author	Arnold D. Robbins <[email protected]>	2025-10-29 10:24:33 +0200
committer	Arnold D. Robbins <[email protected]>	2025-10-29 10:24:33 +0200
commit	a0879473da5f65b1687ed5a4af565bf400f51f88 (patch)
tree	b89e2ab5e12c82f3b043d00060105379d18a7e08
parent	3992eafcead82df31a8daea5d25803c3e9f5e726 (diff)
download	gawk-master.tar.gz