Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
Currently there are quite a few ebuilds in the tree that execute dodoc or dohtml for files that do not exist. I think it would be nice to have ebuilds die if this is the case. To not break current ebuilds this would only happen with FEATURES=stricter. Sigh... There are already bugs flowing in for TEXTRELs/executable stacks checks implemented in recent portage versions. Some of these bugs are completely INVALID or CANTFIX - emulation stuff, binary-only ebuilds, etc. etc. What's the point of this breakage? Why are these QA checks fatal, causing ebuilds to bail out? How can you disable such checks per-ebuild (AFAIK - you can't) to not annoy users with QA notices and breakage one can do nothing about anyway? As Flameeyes pointed out, dodoc/dohtml is also used in eclasses. This can break many ebuilds. Users will report duplicate bugs because they will not realize that it's the eclass causing the failure, not the ebuild. Again, what's the point? How will it work with FEATURES=nodoc? Why should an ebuild ever fail just because some doc file is missing or got renamed or whatever? -- Best regards, Jakub Moc mailto:[EMAIL PROTECTED] GPG signature: http://subkeys.pgp.net:11371/pks/lookup?op=getsearch=0xCEBA3D9E Primary key fingerprint: D2D7 933C 9BA1 C95B 2C95 B30F 8717 D5FD CEBA 3D9E ... still no signature ;) pgpPOJRY0EZSR.pgp Description: PGP signature
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
Diego 'Flameeyes' Pettenò wrote: On Monday 26 December 2005 03:28, Chris White wrote: I'm not sure if we're on the same page as far as the target audience of this change. The target audience is developers/those with strict in their features. Actually stricter, and there are way too many people to put that in without knowing what that do... or is it a default nowadays, I'm not even sure. When I asked about stricter the first time, the response was that it is meant as a developer tool only so it will not be on by default any time soon. Regards, Petteri signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Jakub Moc schrieb:
|
|Currently there are quite a few ebuilds in the tree that execute dodoc or
|dohtml for files that do not exist. I think it would be nice to have
|ebuilds die if this is the case. To not break current ebuilds this would
|only happen with FEATURES=stricter.
|
|
| Sigh... There are already bugs flowing in for TEXTRELs/executable stacks
| checks implemented in recent portage versions. Some of these bugs are
| completely INVALID or CANTFIX - emulation stuff, binary-only ebuilds, etc.
| etc. What's the point of this breakage? Why are these QA checks fatal,
| causing ebuilds to bail out? How can you disable such checks per-ebuild
| (AFAIK - you can't) to not annoy users with QA notices and breakage
one can
| do nothing about anyway?
You can disable them. Have a look at dyn_install in ebuild.sh.
There are 2 categories of such QA violations:
* One category (qa_sucks_for_sure) currently only consists of ebuilds
~ that have run-paths pointing to a subdir of ${BUILDDIR}. Such bugs can
~ always be fixed (as it never affects binary packages) and thus this
~ category of bug lets the build process always die.
* The other category (qa_kinda_sucks) only causes the death of the build
~ process when the user has FEATURES=stricter and the ebuild doesn't
~ have RESTRICT=stricter.
The obvious solution for unfixable (binary) packages is to set
RESTRICT=stricter for them. On the other hand, some binary UPSTREAMs
are very kind and competent to handle such bugs if you tell them. AMD
for example, who will fix an exectuable stack problem in ACML after
the holidays.
Danny
- --
Danny van Dyk [EMAIL PROTECTED]
Gentoo/AMD64 Project, Gentoo Scientific Project
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.1 (GNU/Linux)
iD8DBQFDr+j8aVNL8NrtU6IRAq0kAJ92IHWPU/WRRzj5F807yU+89bm87gCfbbBF
lkpmuU3EgpaFHfaCaiShQxI=
=drQA
-END PGP SIGNATURE-
--
gentoo-dev@gentoo.org mailing list
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
Diego 'Flameeyes' Pettenò wrote: I'm not sure if we're on the same page as far as the target audience of this change. The target audience is developers/those with strict in their features. Actually stricter, and there are way too many people to put that in without knowing what that do... or is it a default nowadays, I'm not even sure. You're mixing up 'strict' with 'stricter'. -- Simon Stelling Gentoo/AMD64 Operational Co-Lead [EMAIL PROTECTED] -- gentoo-dev@gentoo.org mailing list
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
On Monday 26 December 2005 20:01, Jakub Moc wrote: Currently there are quite a few ebuilds in the tree that execute dodoc or dohtml for files that do not exist. I think it would be nice to have ebuilds die if this is the case. To not break current ebuilds this would only happen with FEATURES=stricter. Sigh... There are already bugs flowing in for TEXTRELs/executable stacks checks implemented in recent portage versions. Some of these bugs are completely INVALID or CANTFIX - emulation stuff, binary-only ebuilds, etc. etc. Sigh... None of these issues have made there way to dev-portage. -- Jason Stubbs -- gentoo-dev@gentoo.org mailing list
Re[2]: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
26.12.2005, 14:28:12, Jason Stubbs wrote: On Monday 26 December 2005 20:01, Jakub Moc wrote: Currently there are quite a few ebuilds in the tree that execute dodoc or dohtml for files that do not exist. I think it would be nice to have ebuilds die if this is the case. To not break current ebuilds this would only happen with FEATURES=stricter. Sigh... There are already bugs flowing in for TEXTRELs/executable stacks checks implemented in recent portage versions. Some of these bugs are completely INVALID or CANTFIX - emulation stuff, binary-only ebuilds, etc. etc. Sigh... None of these issues have made there way to dev-portage. -- Jason Stubbs Well, then assign Bug 116499 or Bug 116602 to yourself (qemu), there're textrels in openoffice-bin, mozilla-firefox-bin (upstream, don't hold your breath to get this fixed), acroread (cantfix really), this for sure will be an issue for many games binaries, etc. While it's often upstream/cantfix, I don't see much sense in making these QA checks fatal really. -- Best regards, Jakub Moc mailto:[EMAIL PROTECTED] GPG signature: http://subkeys.pgp.net:11371/pks/lookup?op=getsearch=0xCEBA3D9E Primary key fingerprint: D2D7 933C 9BA1 C95B 2C95 B30F 8717 D5FD CEBA 3D9E ... still no signature ;) pgpmuHTx23Adq.pgp Description: PGP signature
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
On Monday 26 December 2005 13:59, Simon Stelling wrote: Actually stricter, and there are way too many people to put that in without knowing what that do... or is it a default nowadays, I'm not even sure. You're mixing up 'strict' with 'stricter'. Well if I'm mixing up, someone moved the QA checks from stricter to strict lately ;) I don't run strict as I usually have modified ebuilds if I'm working on them; I don't run stricter as lot of packages that fails are not mine, I usually use it only when I'm testing my packages or my changes. -- Diego Flameeyes Pettenò - http://dev.gentoo.org/~flameeyes/ Gentoo/ALT lead, Gentoo/FreeBSD, Video, AMD64, Sound, PAM, KDE pgpsZyuWhdqN2.pgp Description: PGP signature
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
On 12/26/05, Diego 'Flameeyes' Pettenò [EMAIL PROTECTED] wrote: On Monday 26 December 2005 13:59, Simon Stelling wrote: Actually stricter, and there are way too many people to put that in without knowing what that do... or is it a default nowadays, I'm not even sure. You're mixing up 'strict' with 'stricter'. Well if I'm mixing up, someone moved the QA checks from stricter to strict lately ;) I don't run strict as I usually have modified ebuilds if I'm working on them; I don't run stricter as lot of packages that fails are not mine, I usually use it only when I'm testing my packages or my changes. strict is in make.defaults... This causes packages with executable stacks to die, and fairly arbitrarily imo (with portage 2.1_pre2 that is) (see bug 116611). IMUO, portage should never die when an issue of questionable merit comes up and features are simply those set by default. -- Diego Flameeyes Pettenò - http://dev.gentoo.org/~flameeyes/ Gentoo/ALT lead, Gentoo/FreeBSD, Video, AMD64, Sound, PAM, KDE -- gentoo-dev@gentoo.org mailing list
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
and my bad. I am not yet awake. It died cause of runpaths on strict, it just showed both, and I wasn't thinking when I sent earlier email... On 12/26/05, Dan Meltzer [EMAIL PROTECTED] wrote: On 12/26/05, Diego 'Flameeyes' Pettenò [EMAIL PROTECTED] wrote: On Monday 26 December 2005 13:59, Simon Stelling wrote: Actually stricter, and there are way too many people to put that in without knowing what that do... or is it a default nowadays, I'm not even sure. You're mixing up 'strict' with 'stricter'. Well if I'm mixing up, someone moved the QA checks from stricter to strict lately ;) I don't run strict as I usually have modified ebuilds if I'm working on them; I don't run stricter as lot of packages that fails are not mine, I usually use it only when I'm testing my packages or my changes. strict is in make.defaults... This causes packages with executable stacks to die, and fairly arbitrarily imo (with portage 2.1_pre2 that is) (see bug 116611). IMUO, portage should never die when an issue of questionable merit comes up and features are simply those set by default. -- Diego Flameeyes Pettenò - http://dev.gentoo.org/~flameeyes/ Gentoo/ALT lead, Gentoo/FreeBSD, Video, AMD64, Sound, PAM, KDE -- gentoo-dev@gentoo.org mailing list
[gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
Moving this to gentoo-dev then.
Original Message
Subject: Re: [gentoo-portage-dev] making dodoc and dohtml die when they
fail and stricter is on
Date: Sun, 25 Dec 2005 15:14:46 -0800
From: Brian Harring [EMAIL PROTECTED]
Reply-To: gentoo-portage-dev@lists.gentoo.org
To: gentoo-portage-dev@lists.gentoo.org
References: [EMAIL PROTECTED]
On Mon, Dec 26, 2005 at 12:54:04AM +0200, Petteri Räty wrote:
Currently there are quite a few ebuilds in the tree that execute dodoc
or dohtml for files that do not exist. I think it would be nice to have
ebuilds die if this is the case. To not break current ebuilds this would
only happen with FEATURES=stricter. This is what I currently do in my
bashrc. Obviously when integreted to portage one can use helper
functions like hasq which are not available in bashrc.
if [[ ${FEATURES/stricter} != ${FEATURES} ]]; then
_makefail() {
bin=/usr/lib/portage/bin/${1}
shift 1
${bin} [EMAIL PROTECTED] || die ${bin} [EMAIL PROTECTED] failed
}
dodoc() { _makefail ${FUNCNAME} [EMAIL PROTECTED]; }
dohtml() {_makefail ${FUNCNAME} [EMAIL PROTECTED]; }
Seems like more of a -dev discussion imo, since they're the ones
affected by it (for us it's just an api change).
~harring
signature.asc
Description: OpenPGP digital signature
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
On Monday 26 December 2005 08:14, Brian Harring wrote:
On Mon, Dec 26, 2005 at 12:54:04AM +0200, Petteri Räty wrote:
Currently there are quite a few ebuilds in the tree that execute dodoc
or dohtml for files that do not exist. I think it would be nice to have
ebuilds die if this is the case. To not break current ebuilds this would
only happen with FEATURES=stricter. This is what I currently do in my
bashrc. Obviously when integreted to portage one can use helper
functions like hasq which are not available in bashrc.
if [[ ${FEATURES/stricter} != ${FEATURES} ]]; then
_makefail() {
bin=/usr/lib/portage/bin/${1}
shift 1
${bin} [EMAIL PROTECTED] || die ${bin} [EMAIL PROTECTED] failed
}
dodoc() { _makefail ${FUNCNAME} [EMAIL PROTECTED]; }
dohtml() { _makefail ${FUNCNAME} [EMAIL PROTECTED]; }
Seems like more of a -dev discussion imo, since they're the ones
affected by it (for us it's just an api change).
As a side note, dodoc didn't return non-zero when specified files don't exist
up until a month or two ago. dohtml was updated yesterday. Hence, up until
now the above was not possible.
--
Jason Stubbs
--
gentoo-dev@gentoo.org mailing list
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
On Monday 26 December 2005 00:32, Petteri Räty wrote: Currently there are quite a few ebuilds in the tree that execute dodoc or dohtml for files that do not exist. I think it would be nice to have ebuilds die if this is the case. I wouldn't like this. The reason is, they are also used in eclasses that might be generic; while for example kde eclass checks for the presence of files before dodoc-ing them, I would rather see it ignore the actually presence or less of the files, and just dodoc the one that exists, without failing if some does not exists. One can be reasonably safe that it will find AUTHORS ChangeLog README NEWS and TODO files in generic packages, if they follow GNUs style for example, but sometimes they can be missing. I'd rather not see the change. -- Diego Flameeyes Pettenò - http://dev.gentoo.org/~flameeyes/ Gentoo/ALT lead, Gentoo/FreeBSD, Video, AMD64, Sound, PAM, KDE pgpms11GjfT7b.pgp Description: PGP signature
Re: [gentoo-dev] making dodoc and dohtml die when they fail and stricter is on
On Monday 26 December 2005 10:26, Diego 'Flameeyes' Pettenò wrote: I wouldn't like this. The reason is, they are also used in eclasses that might be generic; while for example kde eclass checks for the presence of files before dodoc-ing them, I would rather see it ignore the actually presence or less of the files, and just dodoc the one that exists, without failing if some does not exists. I'm not sure if we're on the same page as far as the target audience of this change. The target audience is developers/those with strict in their features. The whole reason it's done is to be a bit more strict (hence the FEATURE phrasing) on general QA. For users that don't need this level of strict checking, they simply disable the feature. Generalized assumptions of code finding files simply isn't clean. Checks should be done to verify files before attempting to install, that's just the way it should be out of general practice. One can be reasonably safe that it will find AUTHORS ChangeLog README NEWS and TODO files in generic packages, if they follow GNUs style for example, but sometimes they can be missing. Yes, but while GNU style is indeed the more popular of build sytems, others still do exist, and will continue to exist. I've always found dodoc should be checked anyways, and if we're assuming the documentation consists of the formentioned items, then we're also having the situation of missing other important documentation as well. This all should be checked the first time a package is imported/version bumped for consistancy. I'd rather not see the change. Chris White pgpPl3WvLt9yi.pgp Description: PGP signature
