Re: [gentoo-dev] FYI clarifications to skel.ebuild EAPI usage
On Thu, 13 Mar 2008 22:14:47 -0600 RB <[EMAIL PROTECTED]> wrote: > > non-system packages, the only thing stopping people from using > > EAPI 1 where useful is ludditism. > > While most of us appreciate your desire to move forward, ad-hominem > attacks (however subtle) really only serve to damage your point. You're aware that that's not his real name, right? > That said, this is the typical developer-wants-shiny-object, > engineering-wants-stability drama played out day after day in > corporations worldwide, and nothing ever gets solved until someone > puts up. The stability issues, or lack there-of, in EAPI 1 are well understood by those of us who were behind deciding what went into EAPI 1. There is no issue with using EAPI 1 where appropriate for non-system packages. EAPI 1 is a small, well defined, well understood set of additions to EAPI 0. > Please - for the rest of the community's sake, get over > yourselves and your high ideals and spend some of this energy doing > something positive. Like pushing for ratification/completion of > EAPI=0 so none of you have room to complain. Really, ratification of EAPI 0 doesn't affect any of this. It makes no more sense to say that we can't use EAPI 1 until EAPI 0 is ratified than it does to say that we can't use EAPI 0 until EAPI 0 is ratified. -- Ciaran McCreesh signature.asc Description: PGP signature
Re: [gentoo-dev] FYI clarifications to skel.ebuild EAPI usage
> non-system packages, the only thing stopping people from using EAPI 1 where > useful is ludditism. While most of us appreciate your desire to move forward, ad-hominem attacks (however subtle) really only serve to damage your point. That said, this is the typical developer-wants-shiny-object, engineering-wants-stability drama played out day after day in corporations worldwide, and nothing ever gets solved until someone puts up. Please - for the rest of the community's sake, get over yourselves and your high ideals and spend some of this energy doing something positive. Like pushing for ratification/completion of EAPI=0 so none of you have room to complain. Until EAPI=2. RB -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] FYI clarifications to skel.ebuild EAPI usage
On Friday 14 March 2008 03:29:26 Petteri Räty wrote: > solar reported that he had ebuild submissions blindly using EAPI=1 so we > hopefully made the text better reflect that it should not be used unless > absolutely needed. 'Absolutely' is far too strong a word. A better wording would be 'unless useful', or 'unless it would improve the quality of the ebuild'. For non-system packages, the only thing stopping people from using EAPI 1 where useful is ludditism. -- Bo Andresen Gentoo KDE Dev signature.asc Description: This is a digitally signed message part.
[gentoo-dev] FYI clarifications to skel.ebuild EAPI usage
solar reported that he had ebuild submissions blindly using EAPI=1 so we hopefully made the text better reflect that it should not be used unless absolutely needed. Regards, Petteri [EMAIL PROTECTED] /usr/portage $ cvs diff -r 1.44 skel.ebuild Index: skel.ebuild === RCS file: /var/cvsroot/gentoo-x86/skel.ebuild,v retrieving revision 1.44 retrieving revision 1.50 diff -u -r1.44 -r1.50 --- skel.ebuild 1 Jan 2008 09:27:26 - 1.44 +++ skel.ebuild 14 Mar 2008 02:22:02 - 1.50 @@ -12,11 +12,12 @@ # generated to contain the correct data. # The EAPI variable tells the ebuild format in use. -# Defaults to 0 if not specified. See current PMS draft for more details. +# Defaults to 0 if not specified. The current PMS draft contains details on +# a proposed EAPI=0 definition but is not finalized yet. # Eclasses will test for this variable if they need to use EAPI > 0 features. -# Ebuilds should not define EAPI=1 unless they need to use features added -# in that version. -#EAPI=1 +# Ebuilds should not define EAPI > 0 unless they absolutely need to use +# features added in that version. +#EAPI=0 # inherit lists eclasses to inherit functions from. Almost all ebuilds should # inherit eutils, as a large amount of important functionality has been signature.asc Description: OpenPGP digital signature
[gentoo-dev] EAPI and PMS for people who haven't been paying attention
It looks like way too many people don't understand EAPI or PMS. In the interests of reducing the amount of noise these people make whenever anyone mentions either, I give you the following guide. EAPI and PMS for People Who Haven't Been Paying Attention = This isn't a formal specification, and is not technically accurate. Most of this is massively oversimplified. It is focused at developers who aren't themselves directly involved in either package manager development or design of future EAPI features -- it's too technical for end users, who only need two sentences, and not technical enough for people who're deeply involved. In basic terms -- All ebuilds have an EAPI. The EAPI is a piece of metadata that tells the package manager what rules to follow when using that ebuild. Currently existing EAPIs are '0' and '1'. EAPI doesn't stand for anything. PMS is the Package Manager Specification. It will specify and document EAPIs '0', '1' and any future EAPIs, along with tree and profile layouts and various related requirements. EAPIs allow package managers to add new features without breaking systems using older package managers. They also allow package managers to change behaviour of existing functionality that only works the way it does by fluke, without having to support ebuilds that rely upon implementation flukes that were never supposed to be part of the ebuild API. For example: say a new Portage version comes along with a new dep resolver that's more efficient. But because it uses a different algorithm, it behaves differently with respect to blockers. Having a well defined specification for EAPIs provides a clear cut definition of whether it's ok for Portage to do that (and thus whether any ebuilds that rely upon flukes of the old behaviour must be fixed), or whether it's a Portage bug. Package managers that don't support an ebuild's EAPI mustn't attempt to do anything with that ebuild. They could, for example, display the ebuild as masked in a special way. This removes a lot of (but by no means all of) the upgrade problems and arbitrary "we must wait X months since a release before using new features" guidelines. Ebuilds have EAPI '0' unless they explicitly request EAPI '1'. The EAPI '' (empty string) is equivalent to EAPI '0'. EAPI '0' is what you all know and love. New features won't be going into it. Most, but not all existing ebuilds that aren't EAPI '1' will be EAPI '0' compliant. Some existing ebuilds won't be, since realistically they're relying upon behaviour that can neither be expected to remain the same in future Portage versions nor be independently reimplemented. EAPI '1' is EAPI '0', except for the following changes: - Slot deps are allowed. - IUSE defaults are allowed. - The default src_compile works with ECONF_SOURCE. EAPI '1' was introduced to allow developers to use functionality that was available in Portage at that time. It is defined in terms of changes to EAPI '0'. It does not contain lots of useful things that people want or need because Portage support was not available at the time of writing. It is likely that an EAPI '2' will be introduced at some point that also does not contain lots of useful things, simply because offering some of the things that people want now is better than offering all of them later. You will note how EAPI '1' can easily be specified despite EAPI '0' not being completely well defined. Profile files, etc, don't have an EAPI. You can't use slot deps anywhere in the tree except in ebuilds (see later for eclasses). EAPI doesn't specify package manager configuration, VDB or client behaviour. Ebuilds mustn't rely upon any of these. Portage doesn't, in general, do validation. In particular, it won't moan if you use EAPI '1' features in an EAPI '0' c/p-v. You have to be careful with this. EAPI in more detail --- EAPI values are strings. They may or may not be integers. They cannot be compared for anything except equality. The next EAPI might be called "larry-on-wheels". EAPI is metadata for a C/P-V. It is constant across the entire C/P-V, and is subject to the usual metadata invariancy and extraction rules. Currently EAPI is set by the EAPI variable. This is an issue with eclasses. It means that: * If EAPI is to be changed, it must be done so before any inherit. * eclasses must not modify EAPI. * eclasses must work with all EAPIs, and may not require specific EAPIs themselves. They may test for EAPI and do different things depending upon EAPI, so long as those things do not involve erroring out at any stage. So you can do this in an eclass: if [[ "$EAPI" == "1" ]] ; then DEPEND="foo/bar:2" else DEPEND="=foo/bar-2*" fi But not this: EAPI="1" DEPEND="foo/bar:2" Or this: if [[ "$EAPI" != "1" ]] ; then die "EAPI must be 1" fi Or this: foo_pkg_setup() { [[ "$EAPI" == "1" ]] || die }
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 13 Mar 2008 09:55:29 -0700 Chris Gianelloni <[EMAIL PROTECTED]> wrote: > Do people really think that a pre-compiled stage tarball is source? > How about a pre-compiled LiveCD? Anyone? No, but they do think that it's not really a distribution, since you don't use the binaries with changing dependencies or for upgrades. You won't encounter most binary-related issues if you're using them to build what is in effect a static image. -- Ciaran McCreesh signature.asc Description: PGP signature
Re: [gentoo-dev] RDEPEND=DEPEND (was: Help offered - Portage tree)
On Wed, 12 Mar 2008 21:01:09 -0700 "Alec Warner" <[EMAIL PROTECTED]> wrote: > Did we ever set RDEPEND=DEPEND or vice versa automatically? Maybe it > never went into EAPI=0. In either case I'm sure most maintainers > would appreciate feedback in this area. PMS is supposed to say something like "RDEPEND=DEPEND is done for ebuilds iff RDEPEND is unset in the ebuild. This is done before eclass values are merged in.". But I think the current master doesn't actually say what the implicit RDEPEND=DEPEND is, merely that it's done before eclass values... -- Ciaran McCreesh signature.asc Description: PGP signature
Re: [gentoo-dev] Help offered - Portage tree
Joshua, I know that draft quite well, I used as reference for writing Entropy, our binary package manager which only uses {R,P}DEPEND and not DEPEND. So here comes the issue, when *DEPEND are not declared properly Entropy pulls in unneeded packaged. What you are saying is something I am already aware of :) zmedico has been really helpful :) On 3/14/08, joshua jackson <[EMAIL PROTECTED]> wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Fabio Erculiani wrote: > > | Hi Joshua, > | I never had issues with my emails. So I don't really know what to > | answer you regarding to your issues :) > | SPLIT: Although I think it can be a suboptimal thing for us, I can > | understand your policy. Let me add that, to me, the biggest issue is > | about (R)DEPEND. Splitting packages and maintaining in an overlay it's > | not that hard. > | > | > | > > I personally have no desire to follow the redhat/debian/other binary > packaging systems which split up infinitesimally small packages. It > causes a lot more busywork in my opinion then any potential benefits > that it gains you. > > As far as the depend issue you mentioned: Having both Rdepends and > Depends isn't as far as I'm aware part of any EAPI currently (Correct me > if I'm wrong people). Rdepends are needed for the builds so you will > often see either RDEPENDS=${DEPEND} or vice versa. If its not there then > its more of a matter of accounting then anything. I would think, and > correct me if I'm wrong again, that it would make sense that if you only > have RDEPENDS or DEPEND, then those same applications are required in > the runtime of the application. Does it need to be explicitly stated? So > far the three package manager that I'm aware of all manage this fine. > Those being portage, paludis, and pkgcore. If there are other package > managers out there that might have issues Its a perfect example of a > reason to be involved in the EAPI discussions to help define what is > needed and where. > > So what I suggest to you is perhaps looking over the EAPI=0 draft > documentation and proposing some additions and or modifications that > benefit everyone (not just one person), as its designed to be a standard > for anyone who makes use of ebuilds and beyond. > > http://dev.gentoo.org/~spb/pms.pdf > > Is the current form, but halcy0n is working on an updated version of it > for the next council meeting. > > -BEGIN PGP SIGNATURE- > Version: GnuPG v2.0.7 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > > iD8DBQFH2bL22ZWR0Jhg/EsRAkduAJsGBKKl5HgR5YXziPn9yOLbi5F5MwCfacIC > b/aqsokP3A6JFJ7hO4LGNXY= > =BGqi > > -END PGP SIGNATURE- > -- > gentoo-dev@lists.gentoo.org mailing list > > -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
[EMAIL PROTECTED] kirjoitti: Well I'm a newcomer to Gentoo and never heard of Sabayon (great project btw). Knowing no one here or there, nor any history: Never heard of and you say it's a great project? This conversation reminds me of Human Resources. They always have 'procedures' and 'career tracks.' Gentoo's chitchat about earning gold stars and brownie points is giving me HR sickness. You're asking Michael Jordan to prove himself on the high school team. > You don't know Sabayon and can evaluate the skills of lxnay? I've read Gentoo's new dev announcements about monkeys and paper weights. People with a couple of small open-source projects. The monkeys and paper weights get CVS rights. Then the chief architect of Sabayon is scotched over bugzilla output? Please. That smells like bad fish. The new dev announcements are traditionally written in a funky way. Bugzillas are mostly good for non-devs to report bugs. I know zero developers who first think to themselves, "ok, I need a project bugzilla...then I can begin writing code." That isn't how development works. > But I don't know any major project not having a bug reporting tool. "So you don't have time to file bugs but you would have time to fix them" is rhetoric. The issue is ROI. Why file bugzillas that some "dev" authority figure may or may not fix in two years, when you can fix the code yourself? > The perception of things rottening in bugzilla hasn't been proven. Here are the response times to bug filed by lxnay in 2007 and 2008: https://bugs.gentoo.org/show_bug.cgi?id=160333 * first response in two days (and marked as WONTFIX but later reopened) * interactions could be better on either sides https://bugs.gentoo.org/show_bug.cgi?id=188024 * Fixed in two days https://bugs.gentoo.org/show_bug.cgi?id=192786 * Fixed in 7 hours after patch submission https://bugs.gentoo.org/show_bug.cgi?id=193337 * Your "expert" is missing basic ebuild knowledge --> WORKSFORME * 9 minutes https://bugs.gentoo.org/show_bug.cgi?id=199886 * little under two months (yes this is too long) https://bugs.gentoo.org/show_bug.cgi?id=211324 * Been open a couple of weeks waiting for other changes (borderline) We do have lots of stuff that rot in bugzilla in many cases they are requests for new ebuilds and not problems with existing ebuilds. If you want to call him a Gentoo developer, then do so ASAP, and give him CVS. He knows what he is doing and filing bugzillas is a waste of talent. If you let him fix his own bugzillas he might go for that. There is nothing stopping him from taking the quizzes as any other new dev. If he is as good say you he is, answering the quizzes won't take long. You need to attract people like this and if bugzilla isn't working, think up something new. Bugzilla isn't the only way new people come in. > If you dislike his CVS mods you can always revert, take votes, etc. But I say +1 let him have at it. Very useful after it has already been synced to tons of machines and users are hitting our IRC channels with torches. Regards, Petteri signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 13 Mar 2008 09:25:17 -0700 Chris Gianelloni <[EMAIL PROTECTED]> wrote: > On Thu, 2008-03-13 at 13:53 +0100, Fabio Erculiani wrote: > > What I need is _basic_ respect on #gentoo-dev > > I guess you don't understand that respect has to be earned. Mmm, funny, when I said that, certain people disagreed very loudly... -- Ciaran McCreesh signature.asc Description: PGP signature
Re: [gentoo-dev] Help offered - Portage tree
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Fabio Erculiani wrote: | Hi Joshua, | I never had issues with my emails. So I don't really know what to | answer you regarding to your issues :) | SPLIT: Although I think it can be a suboptimal thing for us, I can | understand your policy. Let me add that, to me, the biggest issue is | about (R)DEPEND. Splitting packages and maintaining in an overlay it's | not that hard. | | | I personally have no desire to follow the redhat/debian/other binary packaging systems which split up infinitesimally small packages. It causes a lot more busywork in my opinion then any potential benefits that it gains you. As far as the depend issue you mentioned: Having both Rdepends and Depends isn't as far as I'm aware part of any EAPI currently (Correct me if I'm wrong people). Rdepends are needed for the builds so you will often see either RDEPENDS=${DEPEND} or vice versa. If its not there then its more of a matter of accounting then anything. I would think, and correct me if I'm wrong again, that it would make sense that if you only have RDEPENDS or DEPEND, then those same applications are required in the runtime of the application. Does it need to be explicitly stated? So far the three package manager that I'm aware of all manage this fine. Those being portage, paludis, and pkgcore. If there are other package managers out there that might have issues Its a perfect example of a reason to be involved in the EAPI discussions to help define what is needed and where. So what I suggest to you is perhaps looking over the EAPI=0 draft documentation and proposing some additions and or modifications that benefit everyone (not just one person), as its designed to be a standard for anyone who makes use of ebuilds and beyond. http://dev.gentoo.org/~spb/pms.pdf Is the current form, but halcy0n is working on an updated version of it for the next council meeting. -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFH2bL22ZWR0Jhg/EsRAkduAJsGBKKl5HgR5YXziPn9yOLbi5F5MwCfacIC b/aqsokP3A6JFJ7hO4LGNXY= =BGqi -END PGP SIGNATURE- -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
On 3/14/08, Pierre-Yves Rofes <[EMAIL PROTECTED]> wrote: > > You seem to know him pretty well it seems... > Come on lxnay, who are you trying to fool here? You think that just by > opening an anonymous mail account, we would be dumb enough to not > recognize you? You could at least have been a little more subtle and a > little less self congratuling, maybe it would have worked, who knows :) > > - -- > Pierre-Yves Rofes > Gentoo Linux Security Team I hope you were joking :))) -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Hi Joshua, I never had issues with my emails. So I don't really know what to answer you regarding to your issues :) SPLIT: Although I think it can be a suboptimal thing for us, I can understand your policy. Let me add that, to me, the biggest issue is about (R)DEPEND. Splitting packages and maintaining in an overlay it's not that hard. On 3/13/08, joshua jackson <[EMAIL PROTECTED]> wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Fabio Erculiani wrote: > | Hi all, > | > | Cheers > > interestingly enough ixnay...I've tried contacting you about working > together with Gentoo and on things related to eapi as sabayon is one of > the more popular distributions that has somewhat of a basis on Gentoo > (I've tried approximately 3-4 times in the last year or so) . Every time > I tried from 4 different domain accounts including my Gentoo one I was > denied the ability to send you an email. > > While I'm sure many comments are going to be a bit harsh if realistic > please do feel free to talk to any of the developers. > > Splitting isn't really realistic as that is getting away from upstream. > As an organization we try to maintain the same way as upstream intends. > If they say that mysql is not a collection of server, client then its > just mysql. Xorg is a perfect example. It was a huge package, that got > split up. It took Donnie and the rest of the X team a while to get > everything ready for the tree but we followed upstream in having > individual packages for the different aspects of the larger project. > > Please feel free to contact me directly if you wish > -BEGIN PGP SIGNATURE- > Version: GnuPG v2.0.7 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org > > iD8DBQFH2aop2ZWR0Jhg/EsRAkppAJ0e5u5LEfrdHP/FpsgghMm0kd07mQCfRmZP > 3rMibnJCkKJih3bsz/VYGpY= > =c41u > -END PGP SIGNATURE- > > > -- > gentoo-dev@lists.gentoo.org mailing list > > -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
This conversation reminds me of Human Resources. They always have 'procedures' and 'career tracks.' Gentoo's chitchat about earning gold stars and brownie points is giving me HR sickness. You're asking Michael Jordan to prove himself on the high school team. You must be American. I am President of Sabayon. Give me rights to clobber Gentoo NOW! Except, s/Sabayon/United States s/Gentoo/Iraq -- Anant -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [EMAIL PROTECTED] a écrit : > Well I'm a newcomer to Gentoo and never heard of Sabayon (great project > btw). [...] > > When someone as expert as this [...] > He is worth ten bugzillas. [...] > He knows what he is doing and filing bugzillas is a waste of > talent. [...] > There needs to be a red carpet for people like > this. [...] You seem to know him pretty well it seems... Come on lxnay, who are you trying to fool here? You think that just by opening an anonymous mail account, we would be dumb enough to not recognize you? You could at least have been a little more subtle and a little less self congratuling, maybe it would have worked, who knows :) - -- Pierre-Yves Rofes Gentoo Linux Security Team -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFH2bjAuhJ+ozIKI5gRAkSCAJ4wojl14fhpJSkHUVpFhJ2tDxIZGwCeNVJT lfKC4xa+C2vbO8mYEZar0RE= =LnRw -END PGP SIGNATURE- -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Fabio Erculiani wrote: | Hi all, | | Cheers interestingly enough ixnay...I've tried contacting you about working together with Gentoo and on things related to eapi as sabayon is one of the more popular distributions that has somewhat of a basis on Gentoo (I've tried approximately 3-4 times in the last year or so) . Every time I tried from 4 different domain accounts including my Gentoo one I was denied the ability to send you an email. While I'm sure many comments are going to be a bit harsh if realistic please do feel free to talk to any of the developers. Splitting isn't really realistic as that is getting away from upstream. As an organization we try to maintain the same way as upstream intends. If they say that mysql is not a collection of server, client then its just mysql. Xorg is a perfect example. It was a huge package, that got split up. It took Donnie and the rest of the X team a while to get everything ready for the tree but we followed upstream in having individual packages for the different aspects of the larger project. Please feel free to contact me directly if you wish -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFH2aop2ZWR0Jhg/EsRAkppAJ0e5u5LEfrdHP/FpsgghMm0kd07mQCfRmZP 3rMibnJCkKJih3bsz/VYGpY= =c41u -END PGP SIGNATURE- -- gentoo-dev@lists.gentoo.org mailing list
[gentoo-dev] Re: Help offered - Portage tree
[EMAIL PROTECTED] writes: > Then the chief architect of > Sabayon is scotched over bugzilla output? Can I declare myself Chief Architect of unieject? Does that give me right to take and fix others' software without passing through the usual ways? Counting titles seems more an HR thing to me than looking at behaviour. [and I didn't even want to write to this topic] -- Diego "Flameeyes" Pettenò http://blog.flameeyes.eu/ pgpLV5Q12WHWy.pgp Description: PGP signature
[gentoo-dev] Re: Help offered - Portage tree
Well I'm a newcomer to Gentoo and never heard of Sabayon (great project btw). Knowing no one here or there, nor any history: This conversation reminds me of Human Resources. They always have 'procedures' and 'career tracks.' Gentoo's chitchat about earning gold stars and brownie points is giving me HR sickness. You're asking Michael Jordan to prove himself on the high school team. I've read Gentoo's new dev announcements about monkeys and paper weights. People with a couple of small open-source projects. The monkeys and paper weights get CVS rights. Then the chief architect of Sabayon is scotched over bugzilla output? Please. That smells like bad fish. When someone as expert as this offers help, take it and make him a fast lane. He is worth ten bugzillas. Like a scientist once told me - it would be inefficient for him to clean his office, they have janitors for that. Bugzillas are broken and most Linux people know it. Ubuntu has hundreds of bugs sitting around for years and years. Personally: I have stopped filing bugzillas at various places. Projects organized around bugzillas are inefficient. Bugzillas are mostly good for non-devs to report bugs. I know zero developers who first think to themselves, "ok, I need a project bugzilla...then I can begin writing code." That isn't how development works. "So you don't have time to file bugs but you would have time to fix them" is rhetoric. The issue is ROI. Why file bugzillas that some "dev" authority figure may or may not fix in two years, when you can fix the code yourself? If you want to call him a Gentoo developer, then do so ASAP, and give him CVS. He knows what he is doing and filing bugzillas is a waste of talent. If you let him fix his own bugzillas he might go for that. But wasting his life trying to earn brownie points and respect? Sad. Gentoo needs the manpower and blowing it off with HR excuses is really, really dumb. I can hardly believe what I'm reading. It makes me want to cry. Maybe I should help Sabayon deploy on PowerPC instead of writing to you guys. I don't really care who misunderstood whom, or who has an attitude problem. There needs to be a red carpet for people like this. I would not care if he had a 666 on his head. You need to attract people like this and if bugzilla isn't working, think up something new. If you dislike his CVS mods you can always revert, take votes, etc. But I say +1 let him have at it. -- http://www.fastmail.fm - A fast, anti-spam email service. -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, Mar 13, 2008 at 8:44 PM, Natanael Copa <[EMAIL PROTECTED]> wrote: > Regarding the RDPEND's, there is nothing in the framework protecting the > RDEPENDS from be wrong. If its wrong, package still compiles and > installs and (almost) everyone is happy. Just because it compiles and installs, doesn't mean it runs. > It pulls in unnecessary stuff but who cares? Disk space is cheap. Not always the case. -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Bo Ørsted Andresen wrote: We can all agree that Gentoo's binary package support leaves a lot of room for improvement. But to fix that you need to come up with real solutions and be prepared to discuss them here and through our bugzilla. And accept the fact that it will take time... On top of that you need to be able to accept that we might not agree with your proposed solution for some of the issues... fair enough -nc -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Chris Gianelloni wrote: On Thu, 2008-03-13 at 14:34 +0100, Natanael Copa wrote: On Thu, 2008-03-13 at 00:35 +0100, Fabio Erculiani wrote: I offer my help to fix DEPEND/RDEPEND split issues which is causing me a lot of headaches (along with localizations). For reference, please have a look here: http://planet.sabayonlinux.org/?p=105 I'm another distro builder that uses the Gentoo framework. I can only agree. I had to roll my own binary package format and after a short while I had to do the dependencies myself and just ignore RDEPEND since it was close to useless. I'm a distro builder, too, and I haven't been hitting any of these problems. Would you care to point out the actual problems, or will the "close to useless" comment be our only indication of the perceived problems? Regarding the RDPEND's, there is nothing in the framework protecting the RDEPENDS from be wrong. If its wrong, package still compiles and installs and (almost) everyone is happy. It pulls in unnecessary stuff but who cares? Disk space is cheap. So since I build a distro where size does matter (uclibc) I realised that even if I submit bugs for broken RDEPEND, there will never be an end to those bug reports. Looking at this thread, it seems i was right. That doesn't mean i dont submit bugreports. I do and I very often submit a patch. But there is a limit on how much you can fix in upstream before you need to go other ways. (That applies to fixing package splitting upstream as well btw...) -nc -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Rémi Cardona wrote: Natanael Copa a écrit : But somethimes you just need to accept we don't live in a perfect world. I understood early that nobody cares that much about binpkgs anyway and moved on. I'd say you're mistaken. A lot of people care about binpkg. It's not because a majority of devs don't _use_ them that they are not willing to fix bugs for other use cases. If it would be practically possible I would still use RDEPEND. It worked alot better to build stuff around NEEDED. Same thing can be said about other arches or other OSs Gentoo can run on. and hardened. As for the URL you've provided, I don't know what you were trying to demonstrate because about 3/4th of the bugs were marked as fixed, showing me that your patches got accepted. :) Which is good... isn't it? I wanted to demonstrate that I do submit bugs and very often I submit a patch with it. (so I kinda know what bugs.g.o is) I, for one, encourage you to keep opening bug reports. Don't worry, I will. Cheers -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
On Thu, Mar 13, 2008 at 05:43:35PM +0100, Fabio Erculiani wrote: > What I just need is respect. > I might found around 150-200 bugs on (R)DEPEND. Take 200 on about 6500 > packages we have in our repository, if I take 5 minutes each, I'd end > up to take 16 hours. As the others said, you can either group them together, or as I noted previously, you can use Bugzilla templates. I think the templates are going to be a better option for you, based on not being able to see what you meant about lzo, x264-svn and attr. Hopefully this URL won't break, as I'm deliberately not shortening it to show you a template: http://bugs.gentoo.org/enter_bug.cgi?product=Gentoo%20Linux&version=2007.0&component=Applications&rep_platform=All&op_sys=Linux&priority=P2&bug_severity=normal&bug_status=NEW&alias=&bug_file_loc=&short_desc=Please%20stabilize%20&comment=Target%20keywords%3A%20%0D%0A%0D%0ATest%20procedure%3A%20FEATURES%3Dtest%20emerge%20%0D%0A%0D%0ATest%20procedure%3A%20FEATURES%3D-test%20emerge%0D%0A%28compile-test%20only%29&commentprivacy=0&keywords=STABLEREQ&dependson=&blocked=&maketemplate=Remember%20values%20as%20bookmarkable%20template&form_name=enter_bug&assigned_to=robbat2%40gentoo.org From the advanced bug filing page, write in a lot of the content that you want, and then use the button on the button 'Remember values as bookmarkable template'. If it takes you 5 minutes to file a bunch of near identical bugs, something is wrong. The above is my core URL for filing stable bugs. With a little bit of magic, you can pre-populate every field (eg look at the KEYWORDS+metadata and work out 'CC' automatically), and then just fill in a little bit of unique description if needed for each one. The only time it takes me more than 30 seconds to file a bug requesting stabilization on one of my packages is if I want to write a detailed testplan. -- Robin Hugh Johnson Gentoo Linux Developer & Infra Guy E-Mail : [EMAIL PROTECTED] GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 pgpwKjDy3UKmI.pgp Description: PGP signature
Re: [gentoo-dev] Help offered - Portage tree
On Thu, Mar 13, 2008 at 01:53:34PM +0100, Fabio Erculiani wrote: > Hi Robin, > first of all. > What I need is _basic_ respect on #gentoo-dev > You here seem all polite, but there you like playing me. > This is not a good start. Excuse me? I have never spoken to you on the #gentoo-dev IRC channel, and thus I cannot be 'playing you' there. The only places I have thus communicated with you are this mailing list, and a private IRC discussion. You still haven't responded either to the private IRC, or here, as to what you see about media-libs/x264-svn, dev-libs/lzo or sys-apps/attr is wrong, and I'd really like to know. P.S. Please don't top post. -- Robin Hugh Johnson Gentoo Linux Developer & Infra Guy E-Mail : [EMAIL PROTECTED] GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 pgpLBhTMG6t9f.pgp Description: PGP signature
[gentoo-dev] Re: Help offered - Portage tree
René 'Necoro' Neumann schrieb: I'm not a gentoo-dev - and I did not read the whole thread, because it was too political for me (do I really have to read all these IRC quotes?). +1, this stuff belongs to the -project mailinglist But I just had an idea for this topic (don't know if anyone had this already - or if it is not applicable here), that I want to share: Why not try to find someone, who does all the bug filing? - So lxnay can find and fix the bugs - and someone else files the bugs and does the discussing with the gentoo-devs. Then both sides have what they want. Of course, it still takes time to get things into the tree, but this shouldn't be a problem :) (I think). maybe finding someone who works with a bunch of people would do the trick ;) Just an idea - please don't eat me, if it's a silly one ^^ /me eats portatoes today ;) Greetz -Jokey signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Help offered - Portage tree
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I'm not a gentoo-dev - and I did not read the whole thread, because it was too political for me (do I really have to read all these IRC quotes?). But I just had an idea for this topic (don't know if anyone had this already - or if it is not applicable here), that I want to share: Why not try to find someone, who does all the bug filing? - So lxnay can find and fix the bugs - and someone else files the bugs and does the discussing with the gentoo-devs. Then both sides have what they want. Of course, it still takes time to get things into the tree, but this shouldn't be a problem :) (I think). Just an idea - please don't eat me, if it's a silly one ^^ Regards, Necoro -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFH2X6a4UOg/zhYFuARAhiWAJ0WzGC6jzRODv9pjezsygRBAUoTWQCfQZro eCQ/dsAY+OZsMvg+ffLGCAc= =NqLb -END PGP SIGNATURE- -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 14:48 -0400, Caleb Tennis wrote: > > As much as I hate to say it, your example was rather bunk, because > > openssl changed SONAME during that time. Keeping the package > > You're right here. After review, the problem was the difference between > 0.9.8e and > 0.9.8g, the latter of which provided some form of newer symbol that wasn't in > e. > But the concept is the same. Correct. That would not have been caught and would be an issue, still. > > Uhh... >= in RDEPEND does that, already... Also, this wouldn't have > > resolved your openssl issue, at all. Your machine scenario above would > > have still failed, since the minimum version was 0.9.7 on your build > > host. > > I'm not talking about meeting the minimum required by the ebuild, I'm talking > about > the minimum that were installed at the time of the emerge. > > > Well, I sincerely hope that you do not file such a bug, as it would > > royally screw over the one team in Gentoo that *does* consistently use > > our binary package support. > > I don't plan on filing the bug, but if it was an optional emerge option to > use the > actual version deps vs. the DEPEND of the ebuild, it wouldn't affect you > would it? If it were optional, it wouldn't affect us. I'd have no issue with some kind of optional support for this sort of thing. > > I would definitely like to see the support improved, but not at the > > expense of doing very stupid things like locking to specific > > versions/revisions of packages. No offense, but that screams of RPM > > hell. > > I'm not trying to lock to any specific version. I'm trying to reproduce on > machine > 2 the same state of packages that package A was compiled against on machine > 1. And > even make it optional to do so, via an emerge flag. This is likely usually done by controlling the binrepo. At least, that's how I do it. -- Chris Gianelloni Release Engineering Strategic Lead Games Developer signature.asc Description: This is a digitally signed message part
Re: [gentoo-dev] Help offered - Portage tree
>> I use this example because it's actually hit me before, but it extends to >> lots of >> other scenarios. The obvious fix is to either use --deep, or just make sure >> you >> need machine 2 up to date with machine 1, though that's difficult to do when >> you're >> talking about machine 301 and machine 559. > > As much as I hate to say it, your example was rather bunk, because > openssl changed SONAME during that time. Keeping the package You're right here. After review, the problem was the difference between 0.9.8e and 0.9.8g, the latter of which provided some form of newer symbol that wasn't in e. But the concept is the same. > information isn't *nearly* as important and doing some checking on the > package. It sounds more like we need to keep some additional > information around, so checks on things like NEEDED can be done. > Perhaps some new "LIBRARIES" file which lists libraries installed by the > package. Then, prior to merge, $package_manager could check NEEDED > versus RDEPEND versus LIBRARIES and bail if something's not > right/missing. In this case, even if the RDEPEND was >>=dev-libs/openssl-0.9.7 and you have 0.9.8, it would fail because > NEEDED would list libssl.so.0.9.7, but LIBRARIES would only have > libssl.so.0.9.8 in it. This seems perfectly acceptable to me. > Uhh... >= in RDEPEND does that, already... Also, this wouldn't have > resolved your openssl issue, at all. Your machine scenario above would > have still failed, since the minimum version was 0.9.7 on your build > host. I'm not talking about meeting the minimum required by the ebuild, I'm talking about the minimum that were installed at the time of the emerge. > Well, I sincerely hope that you do not file such a bug, as it would > royally screw over the one team in Gentoo that *does* consistently use > our binary package support. I don't plan on filing the bug, but if it was an optional emerge option to use the actual version deps vs. the DEPEND of the ebuild, it wouldn't affect you would it? > I would definitely like to see the support improved, but not at the > expense of doing very stupid things like locking to specific > versions/revisions of packages. No offense, but that screams of RPM > hell. I'm not trying to lock to any specific version. I'm trying to reproduce on machine 2 the same state of packages that package A was compiled against on machine 1. And even make it optional to do so, via an emerge flag. -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
Fabio Erculiani a écrit : I might found around 150-200 bugs on (R)DEPEND. Take 200 on about 6500 packages we have in our repository, if I take 5 minutes each, I'd end up to take 16 hours. Then open a reduced number of bugs, say one per portage category that has over 20 bugs and group the rest in a one or 2 other bug reports. Then bug wranglers can start sifting through the list to see if you've been smoking crack or not and add relevant herds/arch teams/maintainers as CC to those bugs to get them fixed. I've done this before on a much smaller scale (20+ packages) and others have too for other issues (eg. Flameeyes with WANT_AUTO* which started out as a huge list of 100+ packages) and this process has always worked fine so far. > To build my previous list, I took about 30 minutes, it's not that big, but even that small. So, what I just wanted to try to build up is a fast lane. Bugzilla is still the fastest lane around. We're not the LKML, bugzilla _is_ our primary tool. When I say "I don't have time", it means that I can't waste my time fighting with some of you just because you have the knife in your hand and like to make fun of me. Honestly, this is borderline paranoia. Most of us get bug reports from people we've never heard of, living in countries we've never been to, with ages and social backgrounds that we don't even know (or care) about. And I have yet to see a Gentoo dev closing a bug as WONTFIXBECAUSEIDONTLIKEYOU. :) BTW, It's funny to see the difference of attitudes from here and IRC, let me underline that :) So this is a neutral ground. I'll add that Bugzilla is also very neutral, even more so than mailing lists AFAICS. Cheers, Rémi -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
On 3/13/08, Chris Gianelloni <[EMAIL PROTECTED]> wrote: > Thanks for reminding me once again how you like to interact with the > people that you're trying to "help" out. You wonder why people respond > negatively to your demands and this is how you react to people. > > > -- > > Chris Gianelloni > Release Engineering Strategic Lead > Games Developer Oh so the stupid is me. True true... -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 17:48 +0100, Fabio Erculiani wrote: > On 3/13/08, Chris Gianelloni <[EMAIL PROTECTED]> wrote: > > I'm a distro builder, too, and I haven't been hitting any of these > > problems. Would you care to point out the actual problems, or will the > > "close to useless" comment be our only indication of the perceived > > problems? > Yeah, but IIRC you are a SOURCE distro builder. Arent't you? (I am just > asking!) No. I build binary packages. Hell, catalyst uses the binary package support *heavily* for its caching. Do people really think that a pre-compiled stage tarball is source? How about a pre-compiled LiveCD? Anyone? -- Chris Gianelloni Release Engineering Strategic Lead Games Developer signature.asc Description: This is a digitally signed message part
Re: [gentoo-dev] Re: Help offered - Portage tree
On Thu, 2008-03-13 at 17:22 +0100, Fabio Erculiani wrote: > [gio mar 13 2008] [02:54:56] agaffney: I don't think it's > worth it wasting my time insulting you, I've something better to do > [gio mar 13 2008] [02:55:07]lxnay: oh, burn! > [gio mar 13 2008] [02:55:11] * agaffney cries in the corner > [gio mar 13 2008] [02:55:40] what stupid are you Thanks for reminding me once again how you like to interact with the people that you're trying to "help" out. You wonder why people respond negatively to your demands and this is how you react to people. -- Chris Gianelloni Release Engineering Strategic Lead Games Developer signature.asc Description: This is a digitally signed message part
Re: [gentoo-dev] Help offered - Portage tree
Fabio Erculiani wrote: ^^ This is a really stupid sentence. It seems some of you don't even realize how many users we brought to Gentoo, and this is really sad. I'm not sure I understand how exactly you bring people to Gentoo. You bring people to your distribution which is a binary rebuilt of Gentoo, AFAIK. Or do you have a steady stream of users drifting away from Sabayon to Gentoo? You see, people like Halcy0n, agaffney, zlin keep us away from interacting with you. "Us" being you or who exactly? Halcy0n was just trying to understand what exactly are you going to improve. If you want to stop bad press, you (all) should firstly become more gentle with users and external contributors.> I am not talking to you directly Robin, but to whom are quite annoying and provocative. I know that the majority of you have been always kind, but I will never hang on #gentoo-dev anymore just to be played around giving me voice until I annoy someone with my POV. Well, I was trying pretty hard last night to hear interesting suggestions from you which could be actually implemented. I have even asked the same questions as Halcy0n did, yet you call him a "bad guy" and not me. That's strange. Anyway, please take your time to read the following and think about it. Perhaps you'll find out that we aren't a group of lazy and angry morons, but a group of people that respect each other and wants to get technical issues solved, but with limited time at hand. All you said yesterday was "I don't have time to wait till my bugs are fixed, gimme access so that I can fix them myself". As we have been trying to tell you in more than two hours, this is not how things work. In Gentoo, we respect other developers' work, so if we see a flaw in their code, we speak to them about it and don't go blindly fixing stuff without prior chat with maintainers. Having more than 13k packages in the three, no single person can be expected to know the whole tree well. That's why we are organized into groups and generally talk to each other before fixing bugs. A change you make might have huge impact on packages you haven't ever heard of. During the chat, you proposed various things like "having a mailing lists where child distributions could send bugreports they find". This is not the way to go. We already have a support channel, the Bugzilla. There is really no way to speed up maintainers' reactions. That doesn't depend on how they get the reports, but entirely on their spare time and motivation. If you don't like working with bugzilla's web interface, you've been already offered another access vectors to the bugzilla database. But let me repeat it once again -- if you are worried about maintainers taking "long time" to respond (where "long time" is, by your definition, at about "more than two hours", if I understand you correctly), there's no way I'm aware of that this could be changed. We are just humans who have to sleep, eat, work, date beautiful girls and drink beer. We are not going to abandon any of these just to make the "child distributions" happy, sorry. I have quite a mixed feelings about your offer, too -- you said you're willing to fix stuff, yet you refuse to file bugs, giving a reason that it "takes time". That doesn't make much sense to me, sorry. If you don't file the bug, the same error will stay in the package, it will propagate to each and every next release and you'll have to fix it over and over again in your code. This is not a democratic way, let's talk publicly here, without hiding in a development channel, we probably get more visibility, don't we? I'm afraid I don't fully understand your point here -- Gentoo is not about democracy as in "what majority wants, that happens". If it was such kind of "democracy", we'd have reiser4 as a default filesystem for three years now. In Gentoo, things that happen are things that developers want. If you're bored with that, hey, become a developer and change stuff. Asking us to change the way we work, the process that has worked for many years and that we are happy with, just because it might give some benefits to your distribution, while also causing more work for us, that simply won't happen. Please, try to think about our reasons. Cheers, -jkt -- cd /local/pub && more beer > /dev/mouth signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Help offered - Portage tree
On 3/13/08, Chris Gianelloni <[EMAIL PROTECTED]> wrote: > I'm a distro builder, too, and I haven't been hitting any of these > problems. Would you care to point out the actual problems, or will the > "close to useless" comment be our only indication of the perceived > problems? > > > -- > Chris Gianelloni > Release Engineering Strategic Lead > Games Developer > > Yeah, but IIRC you are a SOURCE distro builder. Arent't you? (I am just asking!) -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 10:15 -0400, Caleb Tennis wrote: > > +1 on that and if people who use binary pkgs don't tell us what breaks, > > we won't know. > > The binpkg format needs some way to store the actual versions of the > dependencies as > they were on the machine the package was compiled on. Then, when emerging the > binpkg, someway to force those dependencies on the new install machine would > be > nice. Please... God... no... > I use this example because it's actually hit me before, but it extends to > lots of > other scenarios. The obvious fix is to either use --deep, or just make sure > you > need machine 2 up to date with machine 1, though that's difficult to do when > you're > talking about machine 301 and machine 559. As much as I hate to say it, your example was rather bunk, because openssl changed SONAME during that time. Keeping the package information isn't *nearly* as important and doing some checking on the package. It sounds more like we need to keep some additional information around, so checks on things like NEEDED can be done. Perhaps some new "LIBRARIES" file which lists libraries installed by the package. Then, prior to merge, $package_manager could check NEEDED versus RDEPEND versus LIBRARIES and bail if something's not right/missing. In this case, even if the RDEPEND was >=dev-libs/openssl-0.9.7 and you have 0.9.8, it would fail because NEEDED would list libssl.so.0.9.7, but LIBRARIES would only have libssl.so.0.9.8 in it. > If there was a way to tell the bin package installer to make sure you met all > of the > same minimum verisons of the deps as they were on the original compiling > machine, > that would be fantastic. Uhh... >= in RDEPEND does that, already... Also, this wouldn't have resolved your openssl issue, at all. Your machine scenario above would have still failed, since the minimum version was 0.9.7 on your build host. > Now, I'm happy to file a bug and assign it (to the portage team?), but I view > this > really as a wishlist item, and since admittedly very few devs use the binpkg > stuff, > I didn't see it as something that would probably get acted upon anyway. I'm > not > complaining about that either, just merely stating a fact. Well, I sincerely hope that you do not file such a bug, as it would royally screw over the one team in Gentoo that *does* consistently use our binary package support. I would definitely like to see the support improved, but not at the expense of doing very stupid things like locking to specific versions/revisions of packages. No offense, but that screams of RPM hell. -- Chris Gianelloni Release Engineering Strategic Lead Games Developer signature.asc Description: This is a digitally signed message part
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 14:34 +0100, Natanael Copa wrote: > On Thu, 2008-03-13 at 00:35 +0100, Fabio Erculiani wrote: > > > I offer my help to fix DEPEND/RDEPEND split issues which is causing me > > a lot of headaches (along with localizations). > > For reference, please have a look here: > > http://planet.sabayonlinux.org/?p=105 > > I'm another distro builder that uses the Gentoo framework. I can only > agree. I had to roll my own binary package format and after a short > while I had to do the dependencies myself and just ignore RDEPEND since > it was close to useless. I'm a distro builder, too, and I haven't been hitting any of these problems. Would you care to point out the actual problems, or will the "close to useless" comment be our only indication of the perceived problems? -- Chris Gianelloni Release Engineering Strategic Lead Games Developer signature.asc Description: This is a digitally signed message part
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 13:53 +0100, Fabio Erculiani wrote: > What I need is _basic_ respect on #gentoo-dev I guess you don't understand that respect has to be earned. There's a major difference between courtesy and respect. You should be given courtesy. You have to earn respect. Demanding respect tends to give the exact opposite response than the demanding individual intends. Try doing something worthy of respecting, and you'll likely gain some respect. -- Chris Gianelloni Release Engineering Strategic Lead Games Developer signature.asc Description: This is a digitally signed message part
Re: [gentoo-dev] Re: Help offered - Portage tree
What I just need is respect. I might found around 150-200 bugs on (R)DEPEND. Take 200 on about 6500 packages we have in our repository, if I take 5 minutes each, I'd end up to take 16 hours. To build my previous list, I took about 30 minutes, it's not that big, but even that small. So, what I just wanted to try to build up is a fast lane. I'm sure there's something we could do to better Gentoo. When I say "I don't have time", it means that I can't waste my time fighting with some of you just because you have the knife in your hand and like to make fun of me. I really admire the commitment of some of you and it's the only thing that led me coming here to talk. BTW, It's funny to see the difference of attitudes from here and IRC, let me underline that :) So this is a neutral ground. -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Re: Help offered - Portage tree
[gio mar 13 2008] [02:50:43] if you want access to the tree, find a mentor and go through the recruitment process [gio mar 13 2008] [02:50:45]lxnay: Other people have to live with the changes. no-one can maintain the whole tree single handed. [gio mar 13 2008] [02:50:51] Halcy0n: because I know where to stop [gio mar 13 2008] [02:51:02] bullshit [gio mar 13 2008] [02:51:05] Betelgeuse: Apparently lxnay can [gio mar 13 2008] [02:51:11] Betelgeuse: mighty mouse could. [gio mar 13 2008] [02:51:17] agaffney: that's what I was trying to do [gio mar 13 2008] [02:51:24] Entra Tommy[D] è entrato nel canale ([EMAIL PROTECTED]/user/TommyD). [gio mar 13 2008] [02:51:47] ferringb: hah [gio mar 13 2008] [02:52:01] lxnay: again, you are working in a community environment. It is not all about you, and there are people that have much more in depth knowledge about the packages they maintain than you could. You want to compare knowledge of GCC with myself or vapier? I can tell you right now if you touch glibc or gcc, we would file a devrel bug immediately. [gio mar 13 2008] [02:52:09] lxnay == mighty mouse? [gio mar 13 2008] [02:52:18] lxnay: If you don't have time to file bugs, howdahell will you have time for recruitment? [gio mar 13 2008] [02:52:31] Entra Maxi è entrato nel canale ([EMAIL PROTECTED]). [gio mar 13 2008] [02:52:35] agaffney: you like a cow [gio mar 13 2008] [02:52:40] agaffney: happy now? [gio mar 13 2008] [02:52:50] Entra idl0r è entrato nel canale ([EMAIL PROTECTED]/idl0r). [gio mar 13 2008] [02:52:59] hparker: I could find time if this allows me to reduce time [gio mar 13 2008] [02:53:04] * Halcy0n thinks he needs to go have another shot of whiskey because this conversation is not making any sense anymore. [gio mar 13 2008] [02:53:14]Time to go to bed. [gio mar 13 2008] [02:53:17] Halcy0n: Get me one too please, I'm out [gio mar 13 2008] [02:53:20] Entra dmwaters è entrato nel canale ([EMAIL PROTECTED]/staff/gentoo.dmwaters). [gio mar 13 2008] [02:53:20] Modalità ChanServ ha dato privilegi di operatore del canale a dmwaters. [gio mar 13 2008] [02:53:24] hparker: Makers work for you? [gio mar 13 2008] [02:53:29] yup! [gio mar 13 2008] [02:53:32] Halcy0n: dibs; still haven't found my bug [gio mar 13 2008] [02:53:32] GOt a new bottle I got to crack open :) [gio mar 13 2008] [02:53:39] ;) [gio mar 13 2008] [02:53:44] Entra cilly è entrato nel canale ([EMAIL PROTECTED]/cilly). [gio mar 13 2008] [02:53:56] I alternate between Knob and Makers. [gio mar 13 2008] [02:53:57] flamefest still going on? [gio mar 13 2008] [02:54:15] dwindling [gio mar 13 2008] [02:54:16] KingTaco: I wouldn't call it a flamefest. [gio mar 13 2008] [02:54:25] lxnay: moo [gio mar 13 2008] [02:54:32] lxnay: do you really think you can insult me? [gio mar 13 2008] [02:54:39] KingTaco: The popcorn didn't get burnt, it was brought to the perfect temperature. [gio mar 13 2008] [02:54:41] Halcy0n: when did it make sense? [gio mar 13 2008] [02:54:56] agaffney: I don't think it's worth it wasting my time insulting you, I've something better to do [gio mar 13 2008] [02:54:57] agaffney: I'm not sure, but I'm thinking if I drink more, it might [gio mar 13 2008] [02:55:07] lxnay: oh, burn! [gio mar 13 2008] [02:55:11] * agaffney cries in the corner [gio mar 13 2008] [02:55:15] agaffney: you should go die alone. [gio mar 13 2008] [02:55:23] oh noes1! [gio mar 13 2008] [02:55:40] what stupid are you On 3/13/08, Ryan Hill <[EMAIL PROTECTED]> wrote: > For those of you playing along at home, the conversation went something like > this: > > hey there, i found a whole bunch of broken stuff in your tree > cool, can you file some reports in bugzilla so we can fix it? > no, i'm too busy and you guys are slow. give me cvs access. > uh no? > you're stupid. > > > > -- > fonts, gcc-porting, by design, by neglect > mips, treecleaner,for a fact or just for effect > wxwidgets @ gentoo EFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662 > > > -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Natanael Copa a écrit : But somethimes you just need to accept we don't live in a perfect world. I understood early that nobody cares that much about binpkgs anyway and moved on. I'd say you're mistaken. A lot of people care about binpkg. It's not because a majority of devs don't _use_ them that they are not willing to fix bugs for other use cases. Same thing can be said about other arches or other OSs Gentoo can run on. As for the URL you've provided, I don't know what you were trying to demonstrate because about 3/4th of the bugs were marked as fixed, showing me that your patches got accepted. :) Which is good... isn't it? I, for one, encourage you to keep opening bug reports. Cheers -- Rémi Cardona LRI, INRIA [EMAIL PROTECTED] [EMAIL PROTECTED] -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Since you wanted to bring this onto this mailing list, I will address all of the points you are making here, though I don't believe this to be the correct place. To preface the whole thing, I was really trying to understand your viewpoint and help make things better, but you were either not understanding what I was saying because of some language issue or something else. Instead of trying to understand me, you took it as some sort of personal attack for some reason. I don't understand why you did, or why you believe I was being "annoying" or "boring" (those were your words last night). So, now onto addressing what you have said. Fabio Erculiani <[EMAIL PROTECTED]> said: > [02:31] lxnay: we offer all of our work that you base your > distribution off, and you don't contribute back at all, in any way. > > ^^ This is a really stupid sentence. It seems some of you don't even > realize how many users we brought to Gentoo, and this is really sad. I wish you would realize that the work all of the Gentoo developers (and users and outside contributors) has done for your distribution. It is entirely based upon our work, and we would really appreciate it if when you find a problem, you report it to us. We want to provide quality packages, but if no one tells us of when there is a problem, we can't do anything. Also, the one statement of mine that you decided to pull out should be put into a little bit of context: 21:26:46 <+lxnay> Halcy0n: I entered #gentoo-portage, found zmedico, talked about the issue, he wrote a patch in less than 1 hour and committed 21:27:06 <+lxnay> Halcy0n: I got the issue solved in less than 3 hours 21:27:17 <+lxnay> if I had to use bugzilla 21:27:26 <+lxnay> it'd have took me days, if not dozen of days 21:27:31 <@hparker> He woulda seen the bug and fixed it... 21:27:41 <@Betelgeuse> yeah 21:27:42 <@Halcy0n> lxnay: I'm not understanding what you are trying to show me. Sometimes you can get in touch with someone on IRC, othertimes you need to use bugzilla. 21:27:55 <@Halcy0n> lxnay: and it would have taken Zac just about the same amount of time. He's quick to respond to things like that anyway. 21:28:19 <@Halcy0n> lxnay: What if zac didn't reply to you iin #g-portage? What would you have done then? 21:28:40 <+lxnay> Halcy0n: I just wanted to tell you, I don't have time to wait (dozen of days) and I always end up forking. and you will never see my fixes, that's a pity for gentoo 21:29:03 <@jkt|> lxnay: but that's entirely your choice, not our flaw 21:29:10 <+lxnay> Halcy0n: he always replied me in a matter of hours 21:29:23 <+lxnay> jkt|: it is if your dev take that much to fix a silly bug 21:29:24 <@Halcy0n> lxnay: so, if you couldn't find Zac on IRC? What would you have done then? Just gone and fixed it on your side? 21:30:01 <@jkt|> lxnay: are you to blame us if we don't fix bugs in less than one week? now this is funny 21:30:07 <+lxnay> sometimes my bugs got closed just because the maintainer didn't find them valid 21:30:09 <@Halcy0n> jkt|: shhh, I'm trying to find out something. 21:30:20 <+lxnay> if someone comes to you and say, I have a problem with X 21:30:31 <@Halcy0n> lxnay: please...if you can't find a dev on IRC, what do you do? 21:30:36 <+lxnay> you should answer, I'll get this fixed asap, as I always do :) 21:30:47 <+lxnay> Halcy0n: I fork the ebuild 21:30:53 <@Halcy0n> lxnay: that's very sad. 21:30:59 <+lxnay> Halcy0n: it's not my fault 21:31:04 <@Halcy0n> lxnay: yes it is completely your fault. 21:31:08 <@jkt|> lxnay: well, it is. 21:31:12 <+lxnay> Halcy0n: no it isn't 21:31:15 <+lxnay> why? 21:31:21 <+lxnay> you make mistakes 21:31:21 <@Halcy0n> lxnay: we offer all of our work that you base your distribution off, and you don't contribute back at all, in any way. 21:31:23 <+lxnay> then it's my fault? 21:31:35 <+lxnay> Halcy0n: yeah of course... 21:31:43 <@jkt|> nope, your fault is not reporting them to us in a reasonable way 21:31:46 <@Halcy0n> lxnay: the fact that you don't contribute it back upstream is, and if that's the attitude you are going to take, then I am done with this conversation as I don't see it going in a positive direction. Just so everyone can see the what led up to my comment, and hopefully understand exactly where it is coming from. Maybe it could have been phrased differently, but I find it very sad when someone is completely unwilling to use our bug tracking system to even try to get a problem fixed. Not all Gentoo devs are on IRC, and we all have other jobs and families that we must take care of. We can't fix bugs within hours all of the time. If you start paying us, then maybe that would change :) > You see, people like Halcy0n, agaffney, zlin keep us away from > interacting with you. Who is "us"? If you have instances where I have made it difficult for any user to approach me and have me help them with a problem, I would love to know. The same goes for everyone else, please contact me privately if you have a problem with m
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 14:48 +0100, Rémi Cardona wrote: > Natanael and Fabio, > But a proper bug report is the way to go if you things to move in any > direction. Happens once in while i post a patch there http://tinyurl.com/399gnr But somethimes you just need to accept we don't live in a perfect world. I understood early that nobody cares that much about binpkgs anyway and moved on. -nc -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 15:42 +0200, Petteri Räty wrote: > Natanael Copa kirjoitti: > > While Gentoo is fantasitc to build stuff, the binary packagement has > > some serious issues. It would be really nice if Gentoo could be better > > on supporting other binary only package managers. > Expected as devs rarely use bin pkgs at all and the Portage support is > what it is. Thats what i found out. -nc -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
> I think remi was more speaking about incorrect deps (say misplaced in > RDEPEND) than problems concerning the package manager. > > In any case, openssl is the perfect example of what can go wrong because > of upstream's behavior. The problem is that program A compiled against > version X of openssl won't work with version Y>X. Currently we need to > keep X's libs around and run revdep-rebuild to fix this. Right, my example was mildly contrived. But I've run into this same issue with packages like ruby and glibc, where the build system had newer versions and you run into symbol issues (or errors like "invalid binary format") because you need to upgrade underlying libraries. > Most librairies don't cause this problem though so I don't really see > this as a bug on the gentoo side even if it's annoying. It's not really Gentoo's fault, no, but it's a problem that could be somewhat fixed. > Anyway, to keep machines using binary in sync without much headache, my > current solution is to use a squashfsed portage tree with --deep. It > works pretty well. Agreed, but the problem is (at least in my case) we're talking about production machines that are actively running, and the customer needs an upgrade of a package but we don't want to take a chance at ruining something else by upgrading --deep if we can help it. >From their perspective, they just want it to work (and don't care about what >has to be upgraded), but from a sysadmin perspective it's a difficult problem to solve over time, especially when you have 10+ other sysadmins, who all may not know that when you upgrade package X be sure to remember to also upgrade packages Y and Z at the same time or you'll run into problems. -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
> Isn't that stored in the NEEDED file? It very well might be, I'm not much of an expert here :) > I think binpkgs store more information than you think. It's just that > Portage doesn't fully use it (yet). This is good information to know. Thanks! Caleb -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Le jeudi 13 mars 2008 à 10:15 -0400, Caleb Tennis a écrit : > > +1 on that and if people who use binary pkgs don't tell us what breaks, > > we won't know. > > I'll kick it off, then. > > The binpkg format needs some way to store the actual versions of the > dependencies as > they were on the machine the package was compiled on. Then, when emerging the > binpkg, someway to force those dependencies on the new install machine would > be > nice. > > I'll give an example. Package A was built on machine 1, and has a dep on > >=openssl-0.9.7. Machine 1 has openssl-0.9.8 already installed. Binary > >package > built, no problem. > > Now, we attempt to install binary package A on machine 2, which has > openssl-0.9.7. > It installs fine, deps met. But, whoops, there's some symbols missing when > we go to > use package A on machine 2. After some time, we finally realize it's because > we > need new openssl. > > I use this example because it's actually hit me before, but it extends to > lots of > other scenarios. The obvious fix is to either use --deep, or just make sure > you > need machine 2 up to date with machine 1, though that's difficult to do when > you're > talking about machine 301 and machine 559. > > If there was a way to tell the bin package installer to make sure you met all > of the > same minimum verisons of the deps as they were on the original compiling > machine, > that would be fantastic. > > Now, I'm happy to file a bug and assign it (to the portage team?), but I view > this > really as a wishlist item, and since admittedly very few devs use the binpkg > stuff, > I didn't see it as something that would probably get acted upon anyway. I'm > not > complaining about that either, just merely stating a fact. I think remi was more speaking about incorrect deps (say misplaced in RDEPEND) than problems concerning the package manager. In any case, openssl is the perfect example of what can go wrong because of upstream's behavior. The problem is that program A compiled against version X of openssl won't work with version Y>X. Currently we need to keep X's libs around and run revdep-rebuild to fix this. Most librairies don't cause this problem though so I don't really see this as a bug on the gentoo side even if it's annoying. Anyway, to keep machines using binary in sync without much headache, my current solution is to use a squashfsed portage tree with --deep. It works pretty well. -- Gilles Dartiguelongue <[EMAIL PROTECTED]> Gentoo -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
(I experimented with binpkgs a little while ago in Prefix) On 13-03-2008 10:15:33 -0400, Caleb Tennis wrote: > > +1 on that and if people who use binary pkgs don't tell us what breaks, > > we won't know. > > I'll kick it off, then. > > The binpkg format needs some way to store the actual versions of the > dependencies as they were on the machine the package was compiled on. > Then, when emerging the binpkg, someway to force those dependencies on > the new install machine would be nice. > > I'll give an example. Package A was built on machine 1, and has a dep on > >=openssl-0.9.7. Machine 1 has openssl-0.9.8 already installed. Binary > >package > built, no problem. > > Now, we attempt to install binary package A on machine 2, which has > openssl-0.9.7. It installs fine, deps met. But, whoops, there's some > symbols missing when we go to use package A on machine 2. After some > time, we finally realize it's because we need new openssl. Isn't that stored in the NEEDED file? > I use this example because it's actually hit me before, but it extends > to lots of other scenarios. The obvious fix is to either use --deep, > or just make sure you need machine 2 up to date with machine 1, though > that's difficult to do when you're talking about machine 301 and > machine 559. > > If there was a way to tell the bin package installer to make sure you > met all of the same minimum verisons of the deps as they were on the > original compiling machine, that would be fantastic. I guess ideally the SLOTs should match, as for instance libpcre 7.5 and 7.6 work fine as long as libpcre.so.0 is there. (No guarantees) But even, for platforms that need libgcc_s.so.1, any gcc that provides it should be fine. Though luckily gcc is almost never in DEPEND/RDEPEND. > Now, I'm happy to file a bug and assign it (to the portage team?), but > I view this really as a wishlist item, and since admittedly very few > devs use the binpkg stuff, I didn't see it as something that would > probably get acted upon anyway. I'm not complaining about that > either, just merely stating a fact. I think binpkgs store more information than you think. It's just that Portage doesn't fully use it (yet). -- Fabian Groffen Gentoo on a different level -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
> +1 on that and if people who use binary pkgs don't tell us what breaks, > we won't know. I'll kick it off, then. The binpkg format needs some way to store the actual versions of the dependencies as they were on the machine the package was compiled on. Then, when emerging the binpkg, someway to force those dependencies on the new install machine would be nice. I'll give an example. Package A was built on machine 1, and has a dep on >=openssl-0.9.7. Machine 1 has openssl-0.9.8 already installed. Binary package built, no problem. Now, we attempt to install binary package A on machine 2, which has openssl-0.9.7. It installs fine, deps met. But, whoops, there's some symbols missing when we go to use package A on machine 2. After some time, we finally realize it's because we need new openssl. I use this example because it's actually hit me before, but it extends to lots of other scenarios. The obvious fix is to either use --deep, or just make sure you need machine 2 up to date with machine 1, though that's difficult to do when you're talking about machine 301 and machine 559. If there was a way to tell the bin package installer to make sure you met all of the same minimum verisons of the deps as they were on the original compiling machine, that would be fantastic. Now, I'm happy to file a bug and assign it (to the portage team?), but I view this really as a wishlist item, and since admittedly very few devs use the binpkg stuff, I didn't see it as something that would probably get acted upon anyway. I'm not complaining about that either, just merely stating a fact. Caleb -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thursday 13 March 2008 14:34:50 Natanael Copa wrote: > On Thu, 2008-03-13 at 00:35 +0100, Fabio Erculiani wrote: > > I offer my help to fix DEPEND/RDEPEND split issues which is causing me > > a lot of headaches (along with localizations). > > For reference, please have a look here: > > http://planet.sabayonlinux.org/?p=105 > > I'm another distro builder that uses the Gentoo framework. I can only > agree. I had to roll my own binary package format and after a short > while I had to do the dependencies myself and just ignore RDEPEND since > it was close to useless. > > While Gentoo is fantasitc to build stuff, the binary packagement has > some serious issues. It would be really nice if Gentoo could be better > on supporting other binary only package managers. We can all agree that Gentoo's binary package support leaves a lot of room for improvement. But to fix that you need to come up with real solutions and be prepared to discuss them here and through our bugzilla. And accept the fact that it will take time... On top of that you need to be able to accept that we might not agree with your proposed solution for some of the issues... -- Bo Andresen Gentoo KDE Dev signature.asc Description: This is a digitally signed message part.
Re: [gentoo-dev] Help offered - Portage tree
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Fabio Erculiani wrote: | [02:31] lxnay: we offer all of our work that you base your | distribution off, and you don't contribute back at all, in any way. | | ^^ This is a really stupid sentence. It seems some of you don't even | realize how many users we brought to Gentoo, and this is really sad. Saying 'this is stupid' won't convince anyone and moreover damage your attempt to build any positive relationship. If the statement is false, simply show with a few examples that it isn't true. To react to a (perceived) personal attack with anger (even if justified) and (more) mudslinging is not going to achieve anything than a further breakdown of communication. Try a more political approach, if you want to get your point across. Personally, I admire your attempt to contribute back and get bugs fixed. I am all for better relationships between Gentoo and Sabayon, because I believe cooperation will be to our mutual benefit. So I think the discussion should concentrate on: ~- what are the best ways to contribute ~- what are the best channels to get bugs fixed ~- how could we streamline these processes I can't believe there is any serious dev that would not want improvement. On the other hand, we are all volunteers and have limited time. Also, there is the natural tendency in most people to stick with the known, to resist changes. And in a big organisation such as Gentoo there will always be various differing points of view. The resulting bureaucracy and politics is something we have to live with, improving it where we can. Therefore, I think that if you really want to improve things, you will have to work with the current infrastructure. Keep filing bugs in bugzilla, keep contributing patches, keep trying to work with other devs, and you _will_ gain respect. That way you will also be in a much better position to suggest changes and to propose ways of streamlining processes. It may take a fair bit of investing time and even swallowing pride. But if you do the work, and let personal attacks and annoyances slide off your thick skin, you _will_ create a win-win situation. Honestly wishing you the best, Ben -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.7 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFH2TVDi+u7I1rvkiYRAm9xAKCTQWLeZ7n6q8K2c+Ugohy+WbbhvgCgh4xo 3Od/xgRiAMyK5Wu1WG6niQc= =N4R3 -END PGP SIGNATURE- -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, Mar 13, 2008 at 2:10 PM, Fabio Erculiani <[EMAIL PROTECTED]> wrote: > [02:31] lxnay: we offer all of our work that you base your > distribution off, and you don't contribute back at all, in any way. > > ^^ This is a really stupid sentence. While I would agree Halcy0n's statement is slightly exaggerated, it's somewhat true. There are exactly 26 non-duplicates in our bugzilla that you either filed or commented on. You have to admit that for somebody who's been a user for 7 years and who's been "architecting" (your word) a distribution based on Gentoo for 3 years (or more ? can't remember) this is a ridiculously low number. How do you want us to help you if you don't give us any feedback on what you need ? We're not very good at communicating but we have at least set up some tools for you to use, and the most important of them in your particular case is I believe bugzilla. And don't ask us to read you blog, we can't possibly read everybody's blog. Feedback is part of the game in the open source world. Gentoo itself gives a lot of feedback to upstream projects using upstream's communications tools. If you don't play the game, one thing is sure is that you'll never win. > It seems some of you don't even > realize how many users we brought to Gentoo, and this is really sad. I'm not sure we should thank you for this. Not that we don't care, but our aim isn't really to compete against say Ubuntu in the users department. Some of us are really happy of the success of Sabayon, there's a place for everybody. We don't make any money out of Gentoo, and do not intend to. We're only a bunch of volunteers who waste their free time doing something they think may be needed. It seems you don't even realize how many users we brought to Sabayon, and this is really sad. > You see, people like Halcy0n, agaffney, zlin keep us away from > interacting with you. I wouldn't agree with you here. But even if you were right, we live in stupid world and Gentoo doesn't claim to be better than what it's made of. Note that we're trying to though. But in the end there are always going to be obnoxious people everywhere. If you give up on the nice guys because of the bad guys, you lose and the bad guys win. That's life. > If you want to stop bad press, you (all) should firstly become more > gentle with users and external contributors. I'm trying to believe you are not threatening here because *that* would be stupid. We'd love to be gentle with you. You've just got to make it happen. Denis. -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Fabio Erculiani wrote: media-libs/x264-svn -> dev-lang/yasm dev-libs/lzo -> dev-lang/nasm sys-apps/attr -> sys-devel/autoconf *snip* Some of those aren't broken, and I just fixed a few media ones in the tree, but that list is similiar to what I was asking for earlier, and a good way to contribute. Thanks Steve -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Natanael and Fabio, Petteri Räty a écrit : Natanael Copa kirjoitti: I'm another distro builder that uses the Gentoo framework. I can only agree. I had to roll my own binary package format and after a short while I had to do the dependencies myself and just ignore RDEPEND since it was close to useless. > http://bugs.gentoo.org I know that we (in the Gnome Herd) will try to fix things when they are reported and I have no doubt other devs will do so as well. But a proper bug report is the way to go if you things to move in any direction. Expected as devs rarely use bin pkgs at all and the Portage support is what it is. +1 on that and if people who use binary pkgs don't tell us what breaks, we won't know. Cheers -- Rémi Cardona LRI, INRIA [EMAIL PROTECTED] [EMAIL PROTECTED] -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Fabio Erculiani kirjoitti: [02:31] lxnay: we offer all of our work that you base your distribution off, and you don't contribute back at all, in any way. ^^ This is a really stupid sentence. It seems some of you don't even realize how many users we brought to Gentoo, and this is really sad. > Nope it's not. I already told you on IRC that you weren't understanding Halcy0n properly, at least from my POV. You say you have to maintain many local changes in your overlay and that you don't have the time to send them back upstream (via the official contribution method, bugs.gentoo.org). For me that means that you aren't contributing back upstream. Regards, Petteri signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Help offered - Portage tree
Natanael Copa kirjoitti: On Thu, 2008-03-13 at 00:35 +0100, Fabio Erculiani wrote: I offer my help to fix DEPEND/RDEPEND split issues which is causing me a lot of headaches (along with localizations). For reference, please have a look here: http://planet.sabayonlinux.org/?p=105 I'm another distro builder that uses the Gentoo framework. I can only agree. I had to roll my own binary package format and after a short while I had to do the dependencies myself and just ignore RDEPEND since it was close to useless. > http://bugs.gentoo.org While Gentoo is fantasitc to build stuff, the binary packagement has some serious issues. It would be really nice if Gentoo could be better on supporting other binary only package managers. -nc Expected as devs rarely use bin pkgs at all and the Portage support is what it is. Regards, Petteri signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Help offered - Portage tree
On Thu, 2008-03-13 at 00:35 +0100, Fabio Erculiani wrote: > I offer my help to fix DEPEND/RDEPEND split issues which is causing me > a lot of headaches (along with localizations). > For reference, please have a look here: http://planet.sabayonlinux.org/?p=105 I'm another distro builder that uses the Gentoo framework. I can only agree. I had to roll my own binary package format and after a short while I had to do the dependencies myself and just ignore RDEPEND since it was close to useless. While Gentoo is fantasitc to build stuff, the binary packagement has some serious issues. It would be really nice if Gentoo could be better on supporting other binary only package managers. -nc -- gentoo-dev@lists.gentoo.org mailing list
[gentoo-dev] Re: Help offered - Portage tree
Fabio Erculiani wrote: [02:31] lxnay: we offer all of our work that you base your distribution off, and you don't contribute back at all, in any way. ^^ This is a really stupid sentence. It seems some of you don't even realize how many users we brought to Gentoo, and this is really sad. For those of you playing along at home, the conversation went something like this: hey there, i found a whole bunch of broken stuff in your tree cool, can you file some reports in bugzilla so we can fix it? no, i'm too busy and you guys are slow. give me cvs access. uh no? you're stupid. -- fonts, gcc-porting, by design, by neglect mips, treecleaner,for a fact or just for effect wxwidgets @ gentoo EFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662 signature.asc Description: OpenPGP digital signature
Re: [gentoo-dev] Help offered - Portage tree
[02:31] lxnay: we offer all of our work that you base your distribution off, and you don't contribute back at all, in any way. ^^ This is a really stupid sentence. It seems some of you don't even realize how many users we brought to Gentoo, and this is really sad. You see, people like Halcy0n, agaffney, zlin keep us away from interacting with you. What we do is just trying to do our best, on the desktop, aggregating new technologies and bringing them to users. If you want to stop bad press, you (all) should firstly become more gentle with users and external contributors. I am not talking to you directly Robin, but to whom are quite annoying and provocative. I know that the majority of you have been always kind, but I will never hang on #gentoo-dev anymore just to be played around giving me voice until I annoy someone with my POV. This is not a democratic way, let's talk publicly here, without hiding in a development channel, we probably get more visibility, don't we? I will review your stuff on lzo probably tomorrow, hope won't be a problem. On 3/13/08, Fabio Erculiani <[EMAIL PROTECTED]> wrote: > Hi Robin, > first of all. > What I need is _basic_ respect on #gentoo-dev > You here seem all polite, but there you like playing me. > This is not a good start. > > > On 3/13/08, Robin H. Johnson <[EMAIL PROTECTED]> wrote: > > On Thu, Mar 13, 2008 at 10:24:23AM +0100, Fabio Erculiani wrote: > > > media-libs/x264-svn -> dev-lang/yasm > > > dev-libs/lzo -> dev-lang/nasm > > > > I responded to you on IRC about these two, please see my message there, > > as from everything I can see, the DEPs are actually correct. > > (The config.log for lzo-1 indicates other reasons that it isn't using > > nasm, which should probably get fixed for both x86 and amd64). > > > > > > > sys-apps/attr -> sys-devel/autoconf > > > > autoconf is in the DEPEND already. > > Do you want it not there? > > > > Not reviewing the rest right now, I'm going to bed instead (03h26 here). > > > > > > -- > > Robin Hugh Johnson > > Gentoo Linux Developer & Infra Guy > > E-Mail : [EMAIL PROTECTED] > > GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 > > > > > > > > -- > Fabio Erculiani > Information and Communication Technologies Consultant > Sabayon Linux Chief Architect > http://www.sabayonlinux.org > -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
Hi Robin, first of all. What I need is _basic_ respect on #gentoo-dev You here seem all polite, but there you like playing me. This is not a good start. On 3/13/08, Robin H. Johnson <[EMAIL PROTECTED]> wrote: > On Thu, Mar 13, 2008 at 10:24:23AM +0100, Fabio Erculiani wrote: > > media-libs/x264-svn -> dev-lang/yasm > > dev-libs/lzo -> dev-lang/nasm > > I responded to you on IRC about these two, please see my message there, > as from everything I can see, the DEPs are actually correct. > (The config.log for lzo-1 indicates other reasons that it isn't using > nasm, which should probably get fixed for both x86 and amd64). > > > > sys-apps/attr -> sys-devel/autoconf > > autoconf is in the DEPEND already. > Do you want it not there? > > Not reviewing the rest right now, I'm going to bed instead (03h26 here). > > > -- > Robin Hugh Johnson > Gentoo Linux Developer & Infra Guy > E-Mail : [EMAIL PROTECTED] > GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 > > -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, Mar 13, 2008 at 10:24:23AM +0100, Fabio Erculiani wrote: > media-libs/x264-svn -> dev-lang/yasm > dev-libs/lzo -> dev-lang/nasm I responded to you on IRC about these two, please see my message there, as from everything I can see, the DEPs are actually correct. (The config.log for lzo-1 indicates other reasons that it isn't using nasm, which should probably get fixed for both x86 and amd64). > sys-apps/attr -> sys-devel/autoconf autoconf is in the DEPEND already. Do you want it not there? Not reviewing the rest right now, I'm going to bed instead (03h26 here). -- Robin Hugh Johnson Gentoo Linux Developer & Infra Guy E-Mail : [EMAIL PROTECTED] GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 pgpPnlmirUp7Z.pgp Description: PGP signature
Re: [gentoo-dev] When does gnome-2.22 come in portage?
Shaochun Wang a écrit : Sorry for the impatience! Where is the gnome overlay? The git repository is here : http://git.overlays.gentoo.org/gitweb/?p=proj/gnome.git;a=summary You can emerge and setup app-portage/layman and then use it to manage the Gnome overlay using the following command : layman -a gnome Things to remember : - don't try to install Gnome 2.22 on a *stable* system. If you do and things break, you get to keep all the pieces. - if you have bugs, don't report them here, use bugzilla with the Gnome component. Enjoy -- Rémi Cardona LRI, INRIA [EMAIL PROTECTED] [EMAIL PROTECTED] -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] When does gnome-2.22 come in portage?
On Thu, 2008-03-13 at 18:02 +0800, Shaochun Wang wrote: > Sorry for the impatience! Where is the gnome overlay? > > -- > Shaochun Wang <[EMAIL PROTECTED]> > > Jabber: [EMAIL PROTECTED] It's listed as gnome-experimental in layman. -- Patrick Ohearn Email: [EMAIL PROTECTED] Jabber: [EMAIL PROTECTED] Site: http://ge3k.net signature.asc Description: This is a digitally signed message part
Re: [gentoo-dev] When does gnome-2.22 come in portage?
Sorry for the impatience! Where is the gnome overlay? -- Shaochun Wang <[EMAIL PROTECTED]> Jabber: [EMAIL PROTECTED] -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
media-libs/x264-svn -> dev-lang/yasm dev-libs/lzo -> dev-lang/nasm sys-apps/attr -> sys-devel/autoconf x11-libs/qt:3 (I reported it a while ago and it got fixed, it was a real mess) net-dialup/capisuite -> sys-devel/autoconf dev-libs/xmlsec -> sys-devel/autoconf x11-misc/fluxbg -> sys-devel/autoconf media-video/effectv -> dev-lang/nasm net-voip/linphone -> dev-lang/nasm media-sound/gogo -> dev-lang/nasm sys-boot/lilo -> sys-devel/bin86 app-text/iso-codes -> sys-devel/automake These depend on sys-devel/bison, are they correct? app-office/mdbtools-0.6_pre1-r1 www-servers/boa-0.94.14_rc21 media-video/sswf-1.8.0-r1 net-firewall/itval-1.0 app-office/openoffice-2.3.1-r1 sci-geosciences/grass-6.0.1 sci-geosciences/grass-6.2.1 media-gfx/gliv-1.9.6 These depend on sys-devel/make sci-geosciences/grass-6.0.1 sci-geosciences/grass-6.2.1 These depend on sys-devel/gcc (remember, only RDEPENDs here) app-text/pdftk-1.41 net-irc/inspircd-1.1.14 app-benchmarks/piozone-1.0-r2 sci-chemistry/xdrawchem-1.9.9 sci-geosciences/grass-6.0.1 dev-lang/mono-1.2.6-r1 sci-geosciences/grass-6.2.1 www-apache/anyterm-1.1.16 dev-lang/ghc-6.8.2 sci-libs/hdf5-1.6.6 x11-proto/xineramaproto: gnome-extra/gnome-screensaver-2.18.2-r1 media-video/ogle-0.9.2-r1 sabayon server # python reagent database query depends --quiet x11-proto/printproto x11-libs/libXp-1.0.0 app-editors/nvu-1.0-r4 x11-libs/openmotif-2.3.0 x11-libs/openmotif-2.2.3-r9 sabayon server # python reagent database query depends --quiet x11-proto/xproto x11-libs/libXevie-1.0.2 x11-libs/libXdmcp-1.0.2 x11-plugins/asclock-2.0.12-r1 dev-libs/libstroke-0.5.1 sys-devel/gcc-3.4.6-r2 x11-libs/libXv-1.0.3 sys-devel/gcc-4.2.2 x11-libs/libXcomposite-0.4.0 x11-plugins/wmmixer-2.0_beta4-r1 x11-plugins/fsviewer-0.2.5 net-www/gnash-0.8.1-r1 x11-libs/libSM-1.0.3 dev-lang/ocaml-3.10.1 x11-libs/libXt-1.0.5 x11-libs/libXaw-1.0.4 x11-libs/libXcursor-1.1.9 gnome-base/nautilus-2.20.0-r1 media-gfx/gifsicle-1.44 x11-libs/xforms-1.0.90-r1 x11-libs/dnd-1.1-r1 x11-libs/libICE-1.0.4 x11-libs/libXft-2.1.12-r90 x11-terms/eterm-0.9.4 media-gfx/tgif-4.1.45 x11-libs/libFS-1.0.0 x11-libs/libXdamage-1.1.1 x11-libs/libXres-1.0.3 x11-libs/libXrandr-1.2.2 x11-libs/libXfont-1.3.1-r1 x11-libs/libXrender-0.9.4 x11-libs/libXau-1.0.3 app-editors/xvile-9.4d-r1 x11-libs/libast-0.7 media-plugins/vdr-xineliboutput-1.0.0_rc2_p20080120 x11-libs/libXvMC-1.0.4 x11-libs/libxsettings-client-0.10 net-dialup/isdn4k-utils-3.11_pre20071003 x11-libs/libX11-1.1.3 x11-libs/libXmu-1.0.3 x11-misc/slim-1.3.0-r1 net-mail/gnubiff-2.2.5 x11-libs/libXfixes-4.0.3 sci-mathematics/snns-4.2-r7 ^^^ do they need x11-proto/xproto as RDEPEND? My time on it for today is over. I'm busy preparing a release, sorry. Probably some of them are ok, but I don't think all. Using http://packages.sabayonlinux.org interface you can query all our bins. -- Fabio Erculiani Information and Communication Technologies Consultant Sabayon Linux Chief Architect http://www.sabayonlinux.org -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] When does gnome-2.22 come in portage?
Shaochun Wang a écrit : Gnome 2.22 was out recently. No, it was released less than 24 hours ago. "Recently" would have been something like a week or two. I have to say that this is by far one of the "best" 0-day bump reports I've seen in a long time... When does it come in portage? When we feel it's ready and when most outstanding bugs are solved. The Gnome 2.21/22 series of packages have been available for months in the gnome overlay. If you want to try it out on an ~ system, use layman and knock yourself out. -- Rémi Cardona LRI, INRIA [EMAIL PROTECTED] [EMAIL PROTECTED] -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
> end up copying the ebuild from the tree into our overlay and fix. great! where is it? does it have a webvc or trac interface? thanks Thilo signature.asc Description: This is a digitally signed message part.
Re: [gentoo-dev] When does gnome-2.22 come in portage?
2008/3/13, Shaochun Wang <[EMAIL PROTECTED]>: > Gnome 2.22 was out recently. When does it come in portage? When the gentoo-devs add it! They are aware of the newly released gnome, so be patient! Regards, Daniel -- gentoo-dev@lists.gentoo.org mailing list
[gentoo-dev] When does gnome-2.22 come in portage?
Gnome 2.22 was out recently. When does it come in portage? -- Shaochun Wang <[EMAIL PROTECTED]> Jabber: [EMAIL PROTECTED] -- gentoo-dev@lists.gentoo.org mailing list
Re: [gentoo-dev] Help offered - Portage tree
On Thu, Mar 13, 2008 at 12:35 AM, Fabio Erculiani <[EMAIL PROTECTED]> wrote: > After having discussed with one of your dev about it, he suggested me > to ask here looking for a mentor. If there's anything I can do, I'm > ready. On Thu, Mar 13, 2008 at 12:57 AM, Fabio Erculiani <[EMAIL PROTECTED]> wrote: > I know what you mean, but take into account I don't have much time > left for the reporting. What I ask is [...] getting me able to fix stuff So you don't have time to file bugs but you would have time to fix them ? Interesting... In any case, we require users to have a consistent history of helping the project before they are considered for recruitment. What you are doing for Sabayon is great but it can't be taken into account. Please find below some information that may be useful to get you started. There are many ways you can help. Two good ways to start helping out are proposing solutions for bugs [1] and contributing to an overlay [2] like Sunrise for example [3]. There is more information on how to get involved with overlay development at [4]. When your contributions become significant enough, developers may contact you (or you can contact them). You may also want to have a look at the staffing needs page [5]. You will need to read the Gentoo Documentation Resources [6], and more specifically the Gentoo Developer Handbook [7] and the Gentoo Development Guide [8]. Another way to help, especially for non-technical projects, is to contact people directly [9]. Be aware that they can be away though, so be patient, try others on the same project, and finally get back to us in case you fail to reach anybody. Do not hesitate to contact recruiters in the future in case you need more information. Best regards, Denis. [1] http://bugs.gentoo.org/ [2] http://overlays.gentoo.org/ [3] http://overlays.gentoo.org/proj/sunrise/ [4] http://www.gentoo.org/proj/en/overlays/userguide.xml#doc_chap3 [5] http://www.gentoo.org/proj/en/devrel/staffing-needs/ [6] http://www.gentoo.org/doc/en/index.xml?catid=gentoodev [7] http://www.gentoo.org/proj/en/devrel/handbook/handbook.xml [8] http://devmanual.gentoo.org/ [9] http://www.gentoo.org/proj/en/index.xml?showlevel=2 -- gentoo-dev@lists.gentoo.org mailing list