bin/143365: [patch] incorrect regexp matching in awk(1)
Mikolaj Golub
to.my.trociny at gmail.com
Sat Jan 30 10:00:08 UTC 2010
>Number: 143365
>Category: bin
>Synopsis: [patch] incorrect regexp matching in awk(1)
>Confidential: no
>Severity: non-critical
>Priority: low
>Responsible: freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Sat Jan 30 10:00:07 UTC 2010
>Closed-Date:
>Last-Modified:
>Originator: Mikolaj Golub
>Release: 8.0-STABLE, 7.2-STABLE
>Organization:
>Environment:
FreeBSD zhuzha.ua1 8.0-STABLE FreeBSD 8.0-STABLE #6: Sun Jan 24 21:36:17 EET 2010 root at zhuzha.ua1:/usr/obj/usr/src/sys/GENERIC i386
>Description:
This problem with awk(1) regexp working incorrectly was reported to NetBSD by Aleksey Cheusov and it was fixed there.
http://www.netbsd.org/cgi-bin/query-pr-single.pl?number=38737
FreeBSD version still has this bug:
kopusha:~% printf '\t\n' | LC_CTYPE=C awk '/^[[:cntrl:]]/ {print "It is ok"}'
It is ok
kopusha:~% printf '\t\n' | LC_CTYPE=C awk '/^[[:cntrl:]]$/ {print "It is ok"}'
kopusha:~%
>How-To-Repeat:
printf '\t\n' | LC_CTYPE=C awk '/^[[:cntrl:]]$/ {print "It is ok"}'
>Fix:
See the attached patch adopted from NetBSD (don't build the character class table
starting at 0, because will always be treated as the empty string).
Patch attached with submission follows:
diff -ru contrib/one-true-awk.orig/b.c contrib/one-true-awk/b.c
--- contrib/one-true-awk.orig/b.c 2007-06-05 18:33:51.000000000 +0300
+++ contrib/one-true-awk/b.c 2010-01-30 11:37:15.000000000 +0200
@@ -818,7 +818,7 @@
if (cc->cc_name != NULL && prestr[1 + cc->cc_namelen] == ':' &&
prestr[2 + cc->cc_namelen] == ']') {
prestr += cc->cc_namelen + 3;
- for (i = 0; i < NCHARS; i++) {
+ for (i = 1; i < NCHARS; i++) {
if (!adjbuf((char **) &buf, &bufsz, bp-buf+1, 100, (char **) &bp, "relex2"))
FATAL("out of space for reg expr %.10s...", lastre);
if (cc->cc_func(i)) {
>Release-Note:
>Audit-Trail:
>Unformatted:
More information about the freebsd-bugs
mailing list