possible mass-bug filing on fc-cache-using packages
Hi folks,
I noticed the other day that there are quite a few packages like
ttf-bitstream-vera which do the following in
/var/lib/dpkg/info/ttf-bitstream-vera.postinst:
if [ $1 = configure -a -x /usr/bin/fc-cache ]
then
echo -n Regenerating fonts cache...
HOME=/root fc-cache -f -v 1/var/log/fontconfig.log 21 || \
(echo failed; see /var/log/fontconfig.log for more information.; \
exit 1)
echo done.
fi
This is wrong. Using ()'s creates a subshell, and exit then only
exits the subshell, not the postinst (which I believe it's supposed
to do). The code either needs an || exit $? appended, or {}'s
should be used. Compare
sh -c '( echo foo; exit 1; ); echo bar'
sh -c '{ echo foo; exit 1; }; echo bar'
The reason seems to be brainless cut-n-pasting.
I think this calls for a mass-bug filing on fc-cache-using packages,
severity minor. But instead of having everyone fix this, I wonder
why debhelper isn't taking care of this. Thoughts?
--
.''`. martin f. krafft [EMAIL PROTECTED]
: :' : proud Debian developer, author, administrator, and user
`. `'` http://people.debian.org/~madduck - http://debiansystem.info
`- Debian - when you have better things to do than fixing systems
even a stopped clock gives the right time twice a day.
-- orbital
digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/)
Re: possible mass-bug filing on fc-cache-using packages
On Tue, May 20, 2008 at 12:05:46PM +0100, martin f krafft wrote:
Hi folks,
I noticed the other day that there are quite a few packages like
ttf-bitstream-vera which do the following in
/var/lib/dpkg/info/ttf-bitstream-vera.postinst:
if [ $1 = configure -a -x /usr/bin/fc-cache ]
then
echo -n Regenerating fonts cache...
HOME=/root fc-cache -f -v 1/var/log/fontconfig.log 21 || \
(echo failed; see /var/log/fontconfig.log for more information.; \
exit 1)
echo done.
fi
This is wrong. Using ()'s creates a subshell, and exit then only
exits the subshell, not the postinst (which I believe it's supposed
to do). The code either needs an || exit $? appended, or {}'s
should be used. Compare
I could be wrong, but if the subshell exits 1, then the entire
expression is false, and so if you are using set -e like you're
supposed to, then the postinst fails.
For example, at the prompt, try:
(set -e; echo 1; false || (echo error; exit 1); echo 2)
which will never print 2.
ttf-bitstream-vera isn't using set -e, which is probably a bug. ISTR
that policy dictated set -e.
--
brian m. carlson / brian with sandals: Houston, Texas, US
+1 713 440 7475 | http://crustytoothpaste.ath.cx/~bmc | My opinion only
troff on top of XML: http://crustytoothpaste.ath.cx/~bmc/code/thwack
OpenPGP: RSA v4 4096b 88AC E9B2 9196 305B A994 7552 F1BA 225C 0223 B187
signature.asc
Description: Digital signature
Re: possible mass-bug filing on fc-cache-using packages
Le mardi 20 mai 2008 à 12:05 +0100, martin f krafft a écrit : if [ $1 = configure -a -x /usr/bin/fc-cache ] Note -that the $1 = configure check is wrong, see #446856. (echo failed; see /var/log/fontconfig.log for more information.; \ exit 1) This is wrong. Using ()'s creates a subshell, and exit then only exits the subshell, not the postinst (which I believe it's supposed to do). Since the postinst uses set -e, it will also exit it. But instead of having everyone fix this, I wonder why debhelper isn't taking care of this. Thoughts? This is not needed anymore. Fontconfig should manage this through a trigger, removing entirely those snippets. Cheers, -- .''`. : :' : We are debian.org. Lower your prices, surrender your code. `. `' We will add your hardware and software distinctiveness to `-our own. Resistance is futile. signature.asc Description: Ceci est une partie de message numériquement signée
Re: possible mass-bug filing on fc-cache-using packages
also sprach brian m. carlson [EMAIL PROTECTED] [2008.05.20.1321 +0100]: I could be wrong, but if the subshell exits 1, then the entire expression is false, and so if you are using set -e like you're supposed to, then the postinst fails. The policy mentions it, but it doesn't say set -e is required. ttf-bitstream-vera does not use it. ttf-bitstream-vera isn't using set -e, which is probably a bug. ISTR that policy dictated set -e. I don't think so, section 6.1: or shell scripts this means that you almost always need to use set -e (this is usually true when writing shell scripts, in fact). This, to me, is waste of space in a policy document. also sprach Josselin Mouette [EMAIL PROTECTED] [2008.05.20.1323 +0100]: Le mardi 20 mai 2008 à 12:05 +0100, martin f krafft a écrit : if [ $1 = configure -a -x /usr/bin/fc-cache ] Note -that the $1 = configure check is wrong, see #446856. Also, the -a is a bashism, isn't it? Since the postinst uses set -e, it will also exit it. It does not. But instead of having everyone fix this, I wonder why debhelper isn't taking care of this. Thoughts? This is not needed anymore. Fontconfig should manage this through a trigger, removing entirely those snippets. Okay, that should make it easier... :) -- .''`. martin f. krafft [EMAIL PROTECTED] : :' : proud Debian developer, author, administrator, and user `. `'` http://people.debian.org/~madduck - http://debiansystem.info `- Debian - when you have better things to do than fixing systems it isn't pollution that's harming the environment. it's the impurities in our air and water that are doing it. - dan quayle digital_signature_gpg.asc Description: Digital signature (see http://martin-krafft.net/gpg/)
Re: possible mass-bug filing on fc-cache-using packages
martin f krafft wrote, Tuesday, May 20, 2008 1:41 PM also sprach Josselin Mouette [EMAIL PROTECTED] [2008.05.20.1323 +0100]: Le mardi 20 mai 2008 à 12:05 +0100, martin f krafft a écrit : if [ $1 = configure -a -x /usr/bin/fc-cache ] Note -that the $1 = configure check is wrong, see #446856. Also, the -a is a bashism, isn't it? It is, but one that policy 10.4 explicitly permits. Adam -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
