summaryrefslogtreecommitdiff
path: root/libc/regexp/patch.2
diff options
context:
space:
mode:
Diffstat (limited to 'libc/regexp/patch.2')
-rw-r--r--libc/regexp/patch.227
1 files changed, 27 insertions, 0 deletions
diff --git a/libc/regexp/patch.2 b/libc/regexp/patch.2
new file mode 100644
index 0000000..fcf4778
--- /dev/null
+++ b/libc/regexp/patch.2
@@ -0,0 +1,27 @@
+Subject: 2nd bug fix for regexp(3), in regsub()
+Newsgroups: mod.sources
+Approved: jpn@panda.UUCP
+
+Mod.sources: Volume 4, Issue 33
+Submitted by: genrad!decvax!utzoo!henry
+
+I cringe with embarrassment! Another bug in regexp! In regsub.c, the
+following line (line 72 in the original distribution):
+
+ if (*(dst-1) == '\0') { /* strncpy hit NUL. */
+
+should read
+
+ if (len != 0 && *(dst-1) == '\0') { /* strncpy hit NUL. */
+
+This shows up as an occasional spurious complaint from regsub about
+"damaged match string"... *if* the program invoking it, and the compiler
+compiling it, happen to interact just right. It didn't show in any of
+my tests because mine don't.
+
+My thanks to H}vard Eidnes for finding this.
+
+ Henry Spencer @ U of Toronto Zoology
+ {allegra,ihnp4,linus,decvax}!utzoo!henry
+
+