Hi,
I have a very interesing problem with sed in FreeBSD.
Lets take the following sed command (from the ncurses MKlib_gen.sh script):
sed -e '/^\([a-z_][a-z_]*\) /s//\1 gen_/'
This command alters the input:
blah something -> blah gen_something
This works, but I have found a specific pattern, where it does not work:
int something -> int something
Does anybody have any idea, what would be the cause of the problem and
how to fix it.
I also have several other FreeBSD systems, where the sed behaves
correctly. I also copied the sed over to the broken system, but no luck.
The broken system used to be 4.6-STABLE and I managed to upgrade it to
4.11-RELEASE-p11 (without working sed). I hoped that upgrade procedure
would elliminate the problem, but apparently not.
"which sed" points to /usr/bin/sed
"ident /usr/bin/sed" output:
# ident /usr/bin/sed
/usr/bin/sed:
$FreeBSD: src/usr.bin/sed/compile.c,v 1.13.2.8 2002/08/17 05:47:06
tjr Exp $
$FreeBSD: src/usr.bin/sed/main.c,v 1.9.2.7 2002/08/06 10:03:29 fanf
Exp $
$FreeBSD: src/usr.bin/sed/misc.c,v 1.3.2.2 2002/07/17 09:35:56 tjr
Exp $
$FreeBSD: src/usr.bin/sed/process.c,v 1.10.2.11 2004/01/10 06:30:37
tjr Exp $
I suspect, that it is some sort of a regex library issue as sed does not
contain its own regex engine.
-- Rein
_______________________________________________
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
