Re: [gentoo-dev] Removal of CVS headers

[Kristian Fiskerstrand]

On 02/26/2017 09:16 PM, Lars Wendler wrote:
> Now QA again wants to do a questionable action _without_ any approval
> from neither infra nor council.

The council has reached a majority for the following statement in [bug
Bug 611234 - Council vote: CVS headers and git expansion]

"""
The council confirms its earlier decision (2014-10-14 meeting) to drop
CVS headers after migration to Git.

1) Any $Id$ and $Header$ lines are to be removed from ebuilds and
eclasses in the gentoo repository, as well as from other files, e.g.,
metadata, profiles, and files (except patches) in FILESDIR.

2) Removal should be done at once, and a repoman check should be
implemented to prevent such lines from accidentally being inserted again.

3) Infra is asked not to expand any $Id$ or other keywords, neither at
rsync generation time, nor via git attributes in the development
repository."
"""


References:
[bug Bug 611234 - Council vote: CVS headers and git expansion]
https://bugs.gentoo.org/611234

-- 
Kristian Fiskerstrand
OpenPGP keyblock reachable at hkp://pool.sks-keyservers.net
fpr:94CB AFDD 3034 5109 5618 35AA 0B7F 8B60 E3ED FAE3



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] Removal of CVS headers

[Mike Gilbert]

On Mon, Feb 27, 2017 at 7:56 AM, Lars Wendler  wrote:
>>> This is a very useful feature and should not be removed only because
>>> council was told that it's a mere CVS migration cruft. It is not!
>>
>>If this is about keeping ebuilds in your overlay in sync, you could
>>alternatively use the output of git ls-tree to keep track of changes
>>using the same blob ids.
>
> This is not the same. git's ident mechanism doesn't use the commit blob
> hash but creates a hash from the file's content. And as I am not using
> VCS for my overlay, git ls-tree is no useful replacement solution here.

I think you completely misread what I wrote. git ls-tree displays the
very same blob ids that appear in the $Id$ field. It's exactly the
same information you are already using.

You would run git ls-tree in gentoo.git to obtain the necessary blob
ids, and store them somewhere (maybe a comment in the ebuild in your
overlay). You could then cross-reference the stored ids after a git
fetch from gentoo.git to tell if anything changed.

Re: [gentoo-dev] Removal of CVS headers

[Lars Wendler]

On Sun, 26 Feb 2017 23:30:25 +0100 Michał Górny wrote:

>W dniu 26.02.2017, nie o godzinie 21∶16 +0100, użytkownik Lars Wendler
>napisał:
>> On Sun, 26 Feb 2017 19:59:19 + Robin H. Johnson wrote:
>>   
>> > On Sat, Feb 25, 2017 at 03:05:09PM +0100, Ulrich Mueller wrote:  
>> > > As the council has decided in its 2014-10-14 meeting (and
>> > > confirmed again in the 2016-11-13 meeting), CVS headers should
>> > > be removed after the migration to Git.
>> > 
>> > The 2014-10-14 meeting did NOT specify what CVS headers were in
>> > question, and it was later decided that this was $Header$, not
>> > $Id$. 
>> > > Until recently, this was blocked by repoman still checking for
>> > > the $Id$ line. The latter is now fixed in the stable repoman
>> > > version.
>> > > 
>> > > Therefore, I am going to remove the remaining CVS headers
>> > > throughout the tree (except for patches, of course) in two days
>> > > from now.
>> > 
>> > This was also discussed in August 2015:
>> > Subject: 'Infra plans regarding $Id$ - official answer...'
>> > https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e
>> > 
>> > $Id$ is used by Git as well, and I was a strong advocate that
>> > expansion of $Id$ should be ENABLED in the rsync exports, because
>> > it allowed tracing what version of a file was actually in use.
>> > 
>> > In the case of Git, $Id$ expands to the blob hash, which can be
>> > traced to a commit trivially, and several of the council members
>> > in the 2015 thread did agree it was useful in that format (but I
>> > see no formal vote was ever taken).
>> >   
>> 
>> And that's exactly for what I use the $Id$ header.
>> I am completely against removal of this header line. It does _not_ do
>> any harm and I don't understand why people want it to be removed so
>> badly.
>> Now QA again wants to do a questionable action _without_ any approval
>> from neither infra nor council. Sorry guys but this is not how things
>> work. The official answer from infra regarding $Id$ gives enough good
>> examples why this header line should be kept.
>> This $Id$ header line is the only way how I can safely keep official
>> ebuilds and ebuilds from my overlay in sync. I don't like getting my
>> workflows sabotaged and I consider this a pure act of sabotage...
>> 
>> How about QA finally starts acting on useful issues or at least do
>> actions that make sense?  
>
>How about you give some respect to your fellow developers who simply
>try to do stuff to improve Gentoo, instead of attributing malice and
>taking it as personal attack on you?

Wow, getting that from the perhaps most rude dev with the least
amount of social skills we might have currently...
Everytime I think you can't do worse you prove me wrong...

>As far as I'm concerned, we could tell as well that the Council decided
>on header removal, then Infra went rogue and replaced the header with
>another one, and now it claims that the decision was about $Header$
>and not $Id$. Does that sound nice to you? Does it motivate you to work
>more on Gentoo?

Nice conspiracy plot... here, have a tin hat.

>Of course, we can dispute that Infra might one day actually start
>expanding $Id$. And not break random files in the process. And not
>break Manifest thickening and signing in the process.
>

No real need to do that expansion for everybody. Just keep the $Id$
header field for those who want to make use of git's ident attribute
feature.

-- 
Lars Wendler
Gentoo package maintainer
GPG: 21CC CF02 4586 0A07 ED93  9F68 498F E765 960E 9B39

Attention! New gpg key! See
https://www.gentoofan.org/blog/index.php?/archives/9-New-gpg-keys.html


pgp8WPRCrYCRc.pgp
Description: Digitale Signatur von OpenPGP

Re: [gentoo-dev] Removal of CVS headers

[Lars Wendler]

On Sun, 26 Feb 2017 16:42:59 -0500 Mike Gilbert wrote:

>On Sun, Feb 26, 2017 at 4:16 PM, Lars Wendler
> wrote:
>> On Sun, 26 Feb 2017 22:07:48 +0100 Ulrich Mueller wrote:
>>  
 On Sun, 26 Feb 2017, Robin H Johnson wrote:  
>>>  
 The 2014-10-14 meeting did NOT specify what CVS headers were in
 question, and it was later decided that this was $Header$, not
 $Id$.  
>>>
>>>When and by whom was that decided? The unanimous council decision was
>>>to remove "CVS headers" and the obvious understanding is that this
>>>includes all Header, Id, Date, and so on. Quoting the actual agenda
>>>item (which was submitted by another infra member, BTW):
>>>
>>># 3. CVS Headers
>>>#
>>># The hateful thing. We could supposedly somehow fill them in rsync
>>># but that's complex and very dangerous (think of all the broken
>>>patch # files currently in gx86). I think we should kill them.
>>>#
>>># And while at it, I think it'd be good to actually remove most of
>>># them from our files -- changing header templates and so on. While
>>># not strictly useful, it decreases the size of the repo a bit and
>>># avoids any future nightmares :).
>>>
>>>Then in the summary of the 2016-11-13 council meeting we have this:
>>>
>>>- Bug 579460 "please make repoman ignore a missing "# $Id$" header
>>>line":
>>>  Implemented in repoman-2.3.0, but not yet in stable. Once this is
>>>  done, CVS headers can be removed as per 2014-10-14 council
>>> decision.
>>>
>>>Note that it explicitly mentions $Id$, and until today nobody had
>>>raised any objections against its revoval.  
>>
>> Well, here is my objection now. And it seems like nobody at the
>> council seems to have understood the intention of this $Id$ header in
>> git.
>> See https://git-scm.com/docs/gitattributes and search for "ident".
>>  
 This was also discussed in August 2015:
 Subject: 'Infra plans regarding $Id$ - official answer...'
 https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e
   
>>>  
 $Id$ is used by Git as well, and I was a strong advocate that
 expansion of $Id$ should be ENABLED in the rsync exports, because
 it allowed tracing what version of a file was actually in use.  
>>>
>>>I don't see any expansion of $Id$ in rsync as of today, 18 months
>>>after the Git conversion. If it wasn't missed for 18 months, one can
>>>hardly claim that it would be an important feature.
>>>
>>>Also I suspect that it is too late to enable it now, because it will
>>>potentially break patches that have been added after the conversion
>>>to Git.  
>>
>> There is no need to enable it by default. But it is a very nice way
>> to verify ebuild changes if being enabled locally on a git clone of
>> the tree. Ever since portage was migrated to git I had this line in
>> my .git/info/attributes file on my dev box:
>>
>> *.ebuild ident
>>
>> This is a very useful feature and should not be removed only because
>> council was told that it's a mere CVS migration cruft. It is not!  
>
>If this is about keeping ebuilds in your overlay in sync, you could
>alternatively use the output of git ls-tree to keep track of changes
>using the same blob ids.

This is not the same. git's ident mechanism doesn't use the commit blob
hash but creates a hash from the file's content. And as I am not using
VCS for my overlay, git ls-tree is no useful replacement solution here.

>It may not be as immediately convenient for you, but it would
>ultimately be more reliable in the long run; there are already several
>ebuilds in the gentoo repo that are missing the $Id$ header. Even if
>we restore the repoman check, I'm sure they will creep in.

Which then could be a perfect opportunity for QA to finally do
something useful.

-- 
Lars Wendler
Gentoo package maintainer
GPG: 21CC CF02 4586 0A07 ED93  9F68 498F E765 960E 9B39

Attention! New gpg key! See
https://www.gentoofan.org/blog/index.php?/archives/9-New-gpg-keys.html


pgpIk_Z6QkF3c.pgp
Description: Digitale Signatur von OpenPGP

Re: [gentoo-dev] Removal of CVS headers

[Ulrich Mueller]

> On Mon, 27 Feb 2017, Robin H Johnson wrote:

> Why did Repoman not complain about either of these?

Repoman has been fixed in bug 579460, following the council decision.

Also app-emacs/ebuild-mode was changed for removal of the $Id$ header,
and I think a similar change was committed for Vim.

Ulrich


pgpmhFqxUuhm_.pgp
Description: PGP signature

Re: [gentoo-dev] Removal of CVS headers

[Robin H. Johnson]

On Sun, Feb 26, 2017 at 10:38:42PM +0100, Ulrich Mueller wrote:
> So what is your suggestion? Keep the status quo and leave all Id,
> Header, etc. in place, including profiles and init scripts? Or are
> these Ids only useful for ebuilds?
I did not ever say it was going to be in init scripts.

Here's the list I had previously, based on where old CVS $Header$ were
most useful, and paths that would have very low chance of breakage.
 */*/*.ebuild ident
 */metadata.xml ident
 */*/metadata.xml ident
 eclass/*.eclass ident
 eclass/tests/* ident
 profiles/* ident
 scripts/* ident

> Also note that there are some 200 ebuilds that don't have any $Id$ in
> their header, and others that still have $Header$.
Why did Repoman not complain about either of these? I will go and fix
them.

-- 
Robin Hugh Johnson
Gentoo Linux: Dev, Infra Lead, Foundation Trustee & Treasurer
E-Mail   : robb...@gentoo.org
GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85
GnuPG FP : 7D0B3CEB E9B85B1F 825BCECF EE05E6F6 A48F6136

Re: [gentoo-dev] Removal of CVS headers

[Michał Górny]

W dniu 26.02.2017, nie o godzinie 21∶16 +0100, użytkownik Lars Wendler
napisał:
> On Sun, 26 Feb 2017 19:59:19 + Robin H. Johnson wrote:
> 
> > On Sat, Feb 25, 2017 at 03:05:09PM +0100, Ulrich Mueller wrote:
> > > As the council has decided in its 2014-10-14 meeting (and confirmed
> > > again in the 2016-11-13 meeting), CVS headers should be removed after
> > > the migration to Git.  
> > 
> > The 2014-10-14 meeting did NOT specify what CVS headers were in
> > question, and it was later decided that this was $Header$, not $Id$.
> > 
> > > Until recently, this was blocked by repoman still checking for the
> > > $Id$ line. The latter is now fixed in the stable repoman version.
> > > 
> > > Therefore, I am going to remove the remaining CVS headers throughout
> > > the tree (except for patches, of course) in two days from now.  
> > 
> > This was also discussed in August 2015:
> > Subject: 'Infra plans regarding $Id$ - official answer...'
> > https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e
> > 
> > $Id$ is used by Git as well, and I was a strong advocate that expansion
> > of $Id$ should be ENABLED in the rsync exports, because it allowed
> > tracing what version of a file was actually in use.
> > 
> > In the case of Git, $Id$ expands to the blob hash, which can be traced
> > to a commit trivially, and several of the council members in the 2015
> > thread did agree it was useful in that format (but I see no formal vote
> > was ever taken).
> > 
> 
> And that's exactly for what I use the $Id$ header.
> I am completely against removal of this header line. It does _not_ do
> any harm and I don't understand why people want it to be removed so
> badly.
> Now QA again wants to do a questionable action _without_ any approval
> from neither infra nor council. Sorry guys but this is not how things
> work. The official answer from infra regarding $Id$ gives enough good
> examples why this header line should be kept.
> This $Id$ header line is the only way how I can safely keep official
> ebuilds and ebuilds from my overlay in sync. I don't like getting my
> workflows sabotaged and I consider this a pure act of sabotage...
> 
> How about QA finally starts acting on useful issues or at least do
> actions that make sense?

How about you give some respect to your fellow developers who simply try
to do stuff to improve Gentoo, instead of attributing malice and taking
it as personal attack on you?

As far as I'm concerned, we could tell as well that the Council decided
on header removal, then Infra went rogue and replaced the header with
another one, and now it claims that the decision was about $Header$
and not $Id$. Does that sound nice to you? Does it motivate you to work
more on Gentoo?

Of course, we can dispute that Infra might one day actually start
expanding $Id$. And not break random files in the process. And not break
Manifest thickening and signing in the process.

-- 
Best regards,
Michał Górny


signature.asc
Description: This is a digitally signed message part

Re: [gentoo-dev] Removal of CVS headers

[Mike Gilbert]

On Sun, Feb 26, 2017 at 4:16 PM, Lars Wendler  wrote:
> On Sun, 26 Feb 2017 22:07:48 +0100 Ulrich Mueller wrote:
>
>>> On Sun, 26 Feb 2017, Robin H Johnson wrote:
>>
>>> The 2014-10-14 meeting did NOT specify what CVS headers were in
>>> question, and it was later decided that this was $Header$, not
>>> $Id$.
>>
>>When and by whom was that decided? The unanimous council decision was
>>to remove "CVS headers" and the obvious understanding is that this
>>includes all Header, Id, Date, and so on. Quoting the actual agenda
>>item (which was submitted by another infra member, BTW):
>>
>># 3. CVS Headers
>>#
>># The hateful thing. We could supposedly somehow fill them in rsync
>># but that's complex and very dangerous (think of all the broken patch
>># files currently in gx86). I think we should kill them.
>>#
>># And while at it, I think it'd be good to actually remove most of
>># them from our files -- changing header templates and so on. While
>># not strictly useful, it decreases the size of the repo a bit and
>># avoids any future nightmares :).
>>
>>Then in the summary of the 2016-11-13 council meeting we have this:
>>
>>- Bug 579460 "please make repoman ignore a missing "# $Id$" header
>>line":
>>  Implemented in repoman-2.3.0, but not yet in stable. Once this is
>>  done, CVS headers can be removed as per 2014-10-14 council decision.
>>
>>Note that it explicitly mentions $Id$, and until today nobody had
>>raised any objections against its revoval.
>
> Well, here is my objection now. And it seems like nobody at the
> council seems to have understood the intention of this $Id$ header in
> git.
> See https://git-scm.com/docs/gitattributes and search for "ident".
>
>>> This was also discussed in August 2015:
>>> Subject: 'Infra plans regarding $Id$ - official answer...'
>>> https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e
>>
>>> $Id$ is used by Git as well, and I was a strong advocate that
>>> expansion of $Id$ should be ENABLED in the rsync exports, because it
>>> allowed tracing what version of a file was actually in use.
>>
>>I don't see any expansion of $Id$ in rsync as of today, 18 months
>>after the Git conversion. If it wasn't missed for 18 months, one can
>>hardly claim that it would be an important feature.
>>
>>Also I suspect that it is too late to enable it now, because it will
>>potentially break patches that have been added after the conversion to
>>Git.
>
> There is no need to enable it by default. But it is a very nice way to
> verify ebuild changes if being enabled locally on a git clone of the
> tree. Ever since portage was migrated to git I had this line in
> my .git/info/attributes file on my dev box:
>
> *.ebuild ident
>
> This is a very useful feature and should not be removed only because
> council was told that it's a mere CVS migration cruft. It is not!

If this is about keeping ebuilds in your overlay in sync, you could
alternatively use the output of git ls-tree to keep track of changes
using the same blob ids.

It may not be as immediately convenient for you, but it would
ultimately be more reliable in the long run; there are already several
ebuilds in the gentoo repo that are missing the $Id$ header. Even if
we restore the repoman check, I'm sure they will creep in.

Re: [gentoo-dev] Removal of CVS headers

[Ulrich Mueller]

> On Sun, 26 Feb 2017, Lars Wendler wrote:

> There is no need to enable it by default. But it is a very nice way
> to verify ebuild changes if being enabled locally on a git clone of
> the tree. Ever since portage was migrated to git I had this line in
> my .git/info/attributes file on my dev box:

> *.ebuild ident

> This is a very useful feature and should not be removed only because
> council was told that it's a mere CVS migration cruft. It is not!

So what is your suggestion? Keep the status quo and leave all Id,
Header, etc. in place, including profiles and init scripts? Or are
these Ids only useful for ebuilds?

Also note that there are some 200 ebuilds that don't have any $Id$ in
their header, and others that still have $Header$.

Ulrich


pgpxdbLw8q66e.pgp
Description: PGP signature

Re: [gentoo-dev] Removal of CVS headers

[Lars Wendler]

On Sun, 26 Feb 2017 22:07:48 +0100 Ulrich Mueller wrote:

>> On Sun, 26 Feb 2017, Robin H Johnson wrote:  
>
>> The 2014-10-14 meeting did NOT specify what CVS headers were in
>> question, and it was later decided that this was $Header$, not
>> $Id$.  
>
>When and by whom was that decided? The unanimous council decision was
>to remove "CVS headers" and the obvious understanding is that this
>includes all Header, Id, Date, and so on. Quoting the actual agenda
>item (which was submitted by another infra member, BTW):
>
># 3. CVS Headers
>#
># The hateful thing. We could supposedly somehow fill them in rsync
># but that's complex and very dangerous (think of all the broken patch
># files currently in gx86). I think we should kill them.
>#
># And while at it, I think it'd be good to actually remove most of
># them from our files -- changing header templates and so on. While
># not strictly useful, it decreases the size of the repo a bit and
># avoids any future nightmares :).
>
>Then in the summary of the 2016-11-13 council meeting we have this:
>
>- Bug 579460 "please make repoman ignore a missing "# $Id$" header
>line":
>  Implemented in repoman-2.3.0, but not yet in stable. Once this is
>  done, CVS headers can be removed as per 2014-10-14 council decision.
>
>Note that it explicitly mentions $Id$, and until today nobody had
>raised any objections against its revoval.

Well, here is my objection now. And it seems like nobody at the
council seems to have understood the intention of this $Id$ header in
git.
See https://git-scm.com/docs/gitattributes and search for "ident".

>> This was also discussed in August 2015:
>> Subject: 'Infra plans regarding $Id$ - official answer...'
>> https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e
>>   
>
>> $Id$ is used by Git as well, and I was a strong advocate that
>> expansion of $Id$ should be ENABLED in the rsync exports, because it
>> allowed tracing what version of a file was actually in use.  
>
>I don't see any expansion of $Id$ in rsync as of today, 18 months
>after the Git conversion. If it wasn't missed for 18 months, one can
>hardly claim that it would be an important feature.
>
>Also I suspect that it is too late to enable it now, because it will
>potentially break patches that have been added after the conversion to
>Git.

There is no need to enable it by default. But it is a very nice way to
verify ebuild changes if being enabled locally on a git clone of the
tree. Ever since portage was migrated to git I had this line in
my .git/info/attributes file on my dev box:

*.ebuild ident

This is a very useful feature and should not be removed only because
council was told that it's a mere CVS migration cruft. It is not!

-- 
Lars Wendler
Gentoo package maintainer
GPG: 21CC CF02 4586 0A07 ED93  9F68 498F E765 960E 9B39

Attention! New gpg key! See
https://www.gentoofan.org/blog/index.php?/archives/9-New-gpg-keys.html


pgpBNWn6sTwOy.pgp
Description: Digitale Signatur von OpenPGP

Re: [gentoo-dev] Removal of CVS headers

[Ulrich Mueller]

> On Sun, 26 Feb 2017, Robin H Johnson wrote:

> The 2014-10-14 meeting did NOT specify what CVS headers were in
> question, and it was later decided that this was $Header$, not $Id$.

When and by whom was that decided? The unanimous council decision was
to remove "CVS headers" and the obvious understanding is that this
includes all Header, Id, Date, and so on. Quoting the actual agenda
item (which was submitted by another infra member, BTW):

# 3. CVS Headers
#
# The hateful thing. We could supposedly somehow fill them in rsync
# but that's complex and very dangerous (think of all the broken patch
# files currently in gx86). I think we should kill them.
#
# And while at it, I think it'd be good to actually remove most of
# them from our files -- changing header templates and so on. While
# not strictly useful, it decreases the size of the repo a bit and
# avoids any future nightmares :).

Then in the summary of the 2016-11-13 council meeting we have this:

- Bug 579460 "please make repoman ignore a missing "# $Id$" header line":
  Implemented in repoman-2.3.0, but not yet in stable. Once this is
  done, CVS headers can be removed as per 2014-10-14 council decision.

Note that it explicitly mentions $Id$, and until today nobody had
raised any objections against its revoval.

> This was also discussed in August 2015:
> Subject: 'Infra plans regarding $Id$ - official answer...'
> https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e

> $Id$ is used by Git as well, and I was a strong advocate that
> expansion of $Id$ should be ENABLED in the rsync exports, because it
> allowed tracing what version of a file was actually in use.

I don't see any expansion of $Id$ in rsync as of today, 18 months
after the Git conversion. If it wasn't missed for 18 months, one can
hardly claim that it would be an important feature.

Also I suspect that it is too late to enable it now, because it will
potentially break patches that have been added after the conversion to
Git.

Ulrich


pgpnZG3A0ANcp.pgp
Description: PGP signature

Re: [gentoo-dev] Removal of CVS headers

[Lars Wendler]

On Sun, 26 Feb 2017 15:32:56 -0500 Rich Freeman wrote:

>On Sun, Feb 26, 2017 at 3:27 PM, Lars Wendler
> wrote:
>> On Sun, 26 Feb 2017 21:24:38 +0100 Andreas K. Huettel wrote:
>>  
>>>Am Sonntag, 26. Februar 2017, 21:16:28 CET schrieb Lars Wendler:  
 I am completely against removal of this header line. It does _not_
 do any harm and I don't understand why people want it to be
 removed so badly.
 Now QA again wants to do a questionable action _without_ any
 approval from neither infra nor council.  
>>>[snip]
>>>
>>>October 2014 council meeting:
>>>
>>>Can we drop CVS headers post-migration?
>>>Aye: blueness, creffett (proxy for ulm), dberkholz, dilfridge,
>>>radhermit, rich0, williamh
>>>  
>>
>> $Id$ is _NOT_ the CVS header.
>>  
>
>Do we really need to put it on the Council agenda just so that the
>same group of people can approve it again?
>

Yes please. I'd like to raise my concerns if this is really the only
way to keep this header.

-- 
Lars Wendler
Gentoo package maintainer
GPG: 21CC CF02 4586 0A07 ED93  9F68 498F E765 960E 9B39

Attention! New gpg key! See
https://www.gentoofan.org/blog/index.php?/archives/9-New-gpg-keys.html


pgp0Mxy1KfT5m.pgp
Description: Digitale Signatur von OpenPGP

Re: [gentoo-dev] Removal of CVS headers

[Rich Freeman]

On Sun, Feb 26, 2017 at 3:27 PM, Lars Wendler  wrote:
> On Sun, 26 Feb 2017 21:24:38 +0100 Andreas K. Huettel wrote:
>
>>Am Sonntag, 26. Februar 2017, 21:16:28 CET schrieb Lars Wendler:
>>> I am completely against removal of this header line. It does _not_ do
>>> any harm and I don't understand why people want it to be removed so
>>> badly.
>>> Now QA again wants to do a questionable action _without_ any approval
>>> from neither infra nor council.
>>[snip]
>>
>>October 2014 council meeting:
>>
>>Can we drop CVS headers post-migration?
>>Aye: blueness, creffett (proxy for ulm), dberkholz, dilfridge,
>>radhermit, rich0, williamh
>>
>
> $Id$ is _NOT_ the CVS header.
>

Do we really need to put it on the Council agenda just so that the
same group of people can approve it again?

-- 
Rich

Re: [gentoo-dev] Removal of CVS headers

[Lars Wendler]

On Sun, 26 Feb 2017 21:24:38 +0100 Andreas K. Huettel wrote:

>Am Sonntag, 26. Februar 2017, 21:16:28 CET schrieb Lars Wendler:
>> I am completely against removal of this header line. It does _not_ do
>> any harm and I don't understand why people want it to be removed so
>> badly.
>> Now QA again wants to do a questionable action _without_ any approval
>> from neither infra nor council.  
>[snip]
>
>October 2014 council meeting:
>
>Can we drop CVS headers post-migration?
>Aye: blueness, creffett (proxy for ulm), dberkholz, dilfridge,
>radhermit, rich0, williamh 
>
>
>
>

$Id$ is _NOT_ the CVS header.

-- 
Lars Wendler
Gentoo package maintainer
GPG: 21CC CF02 4586 0A07 ED93  9F68 498F E765 960E 9B39

Attention! New gpg key! See
https://www.gentoofan.org/blog/index.php?/archives/9-New-gpg-keys.html


pgpL6ISAlDEuc.pgp
Description: Digitale Signatur von OpenPGP

Re: [gentoo-dev] Removal of CVS headers

[Andreas K. Huettel]

Am Sonntag, 26. Februar 2017, 21:16:28 CET schrieb Lars Wendler:
> I am completely against removal of this header line. It does _not_ do
> any harm and I don't understand why people want it to be removed so
> badly.
> Now QA again wants to do a questionable action _without_ any approval
> from neither infra nor council.
[snip]

October 2014 council meeting:

Can we drop CVS headers post-migration?
Aye: blueness, creffett (proxy for ulm), dberkholz, dilfridge, radhermit, 
rich0, williamh 




-- 
Andreas K. Hüttel
dilfri...@gentoo.org
Gentoo Linux developer (council, perl, libreoffice)

signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-dev] Removal of CVS headers

[Ciaran McCreesh]

On Sun, 26 Feb 2017 21:16:28 +0100
Lars Wendler  wrote:
> How about QA finally starts acting on useful issues or at least do
> actions that make sense?

Part of the job of QA is to improve the overall quality of the tree.
This includes going through and fixing historical mistakes. You may
think it's not important, but if Gentoo finally gets into the habit of
ongoing improvements, the tree will slowly get better rather than
worse over time.

-- 
Ciaran McCreesh

Re: [gentoo-dev] Removal of CVS headers

[Lars Wendler]

On Sun, 26 Feb 2017 19:59:19 + Robin H. Johnson wrote:

>On Sat, Feb 25, 2017 at 03:05:09PM +0100, Ulrich Mueller wrote:
>> As the council has decided in its 2014-10-14 meeting (and confirmed
>> again in the 2016-11-13 meeting), CVS headers should be removed after
>> the migration to Git.  
>The 2014-10-14 meeting did NOT specify what CVS headers were in
>question, and it was later decided that this was $Header$, not $Id$.
>
>> Until recently, this was blocked by repoman still checking for the
>> $Id$ line. The latter is now fixed in the stable repoman version.
>> 
>> Therefore, I am going to remove the remaining CVS headers throughout
>> the tree (except for patches, of course) in two days from now.  
>This was also discussed in August 2015:
>Subject: 'Infra plans regarding $Id$ - official answer...'
>https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e
>
>$Id$ is used by Git as well, and I was a strong advocate that expansion
>of $Id$ should be ENABLED in the rsync exports, because it allowed
>tracing what version of a file was actually in use.
>
>In the case of Git, $Id$ expands to the blob hash, which can be traced
>to a commit trivially, and several of the council members in the 2015
>thread did agree it was useful in that format (but I see no formal vote
>was ever taken).
>

And that's exactly for what I use the $Id$ header.
I am completely against removal of this header line. It does _not_ do
any harm and I don't understand why people want it to be removed so
badly.
Now QA again wants to do a questionable action _without_ any approval
from neither infra nor council. Sorry guys but this is not how things
work. The official answer from infra regarding $Id$ gives enough good
examples why this header line should be kept.
This $Id$ header line is the only way how I can safely keep official
ebuilds and ebuilds from my overlay in sync. I don't like getting my
workflows sabotaged and I consider this a pure act of sabotage...

How about QA finally starts acting on useful issues or at least do
actions that make sense?

-- 
Lars Wendler
Gentoo package maintainer
GPG: 21CC CF02 4586 0A07 ED93  9F68 498F E765 960E 9B39

Attention! New gpg key! See
https://www.gentoofan.org/blog/index.php?/archives/9-New-gpg-keys.html


pgpXlAc1_vLo2.pgp
Description: Digitale Signatur von OpenPGP

Re: [gentoo-dev] Removal of CVS headers

[Gordon Pettey]

On Sun, Feb 26, 2017 at 1:59 PM, Robin H. Johnson 
wrote:

> On Sat, Feb 25, 2017 at 03:05:09PM +0100, Ulrich Mueller wrote:
> > As the council has decided in its 2014-10-14 meeting (and confirmed
> > again in the 2016-11-13 meeting), CVS headers should be removed after
> > the migration to Git.
> The 2014-10-14 meeting did NOT specify what CVS headers were in
> question, and it was later decided that this was $Header$, not $Id$.
>
> > Until recently, this was blocked by repoman still checking for the
> > $Id$ line. The latter is now fixed in the stable repoman version.
> >
> > Therefore, I am going to remove the remaining CVS headers throughout
> > the tree (except for patches, of course) in two days from now.
> This was also discussed in August 2015:
> Subject: 'Infra plans regarding $Id$ - official answer...'
> https://archives.gentoo.org/gentoo-dev/message/
> d01ce943a9f9404c454c26bdb7efdf0e
>
> $Id$ is used by Git as well, and I was a strong advocate that expansion
> of $Id$ should be ENABLED in the rsync exports, because it allowed
> tracing what version of a file was actually in use.
>
> In the case of Git, $Id$ expands to the blob hash, which can be traced
> to a commit trivially, and several of the council members in the 2015
> thread did agree it was useful in that format (but I see no formal vote
> was ever taken).
>

Which can also be generated trivially (
https://stackoverflow.com/questions/223678/which-commit-has-this-blob),
which kind of obviates the need for $Id$.

Re: [gentoo-dev] Removal of CVS headers

[Robin H. Johnson]

On Sat, Feb 25, 2017 at 03:05:09PM +0100, Ulrich Mueller wrote:
> As the council has decided in its 2014-10-14 meeting (and confirmed
> again in the 2016-11-13 meeting), CVS headers should be removed after
> the migration to Git.
The 2014-10-14 meeting did NOT specify what CVS headers were in
question, and it was later decided that this was $Header$, not $Id$.

> Until recently, this was blocked by repoman still checking for the
> $Id$ line. The latter is now fixed in the stable repoman version.
> 
> Therefore, I am going to remove the remaining CVS headers throughout
> the tree (except for patches, of course) in two days from now.
This was also discussed in August 2015:
Subject: 'Infra plans regarding $Id$ - official answer...'
https://archives.gentoo.org/gentoo-dev/message/d01ce943a9f9404c454c26bdb7efdf0e

$Id$ is used by Git as well, and I was a strong advocate that expansion
of $Id$ should be ENABLED in the rsync exports, because it allowed
tracing what version of a file was actually in use.

In the case of Git, $Id$ expands to the blob hash, which can be traced
to a commit trivially, and several of the council members in the 2015
thread did agree it was useful in that format (but I see no formal vote
was ever taken).

-- 
Robin Hugh Johnson
Gentoo Linux: Dev, Infra Lead, Foundation Trustee & Treasurer
E-Mail   : robb...@gentoo.org
GnuPG FP : 11ACBA4F 4778E3F6 E4EDF38E B27B944E 34884E85
GnuPG FP : 7D0B3CEB E9B85B1F 825BCECF EE05E6F6 A48F6136

Re: [gentoo-dev] Removal of CVS headers

[Jan Chren (rindeal)]

On 25 February 2017 at 15:05, Ulrich Mueller  wrote:
> As the council has decided in its 2014-10-14 meeting (and confirmed
> again in the 2016-11-13 meeting), CVS headers should be removed after
> the migration to Git. Until recently, this was blocked by repoman
> still checking for the $Id$ line. The latter is now fixed in the
> stable repoman version.
>
> Therefore, I am going to remove the remaining CVS headers throughout
> the tree (except for patches, of course) in two days from now.

Don't forget to clean up wiki and dev docs as well.

Re: [gentoo-dev] Removal of CVS headers

[Rich Freeman]

On Sat, Feb 25, 2017 at 10:12 AM, Ulrich Mueller  wrote:
>> On Sat, 25 Feb 2017, Sergei Trofimovich wrote:
>
>> Typical questions for tree-wide cleanups:
>
>> - Are new ebuilds forbidden to have '$Id$' or just discouraged?
>> - [same as above] Will new version of repoman complain about
>>   leftover '$Id$'?
>
> Not sure. That would have to be controlled via layout.conf, in case
> someone uses CVS or Subversion for their overlay?

It should certainly be forbidden in the main tree though, otherwise
what is the point in cleaning them up?

By all means be nice to people who want to have it in their overlays
and make it controllable in layout.conf, but it isn't like SVN or CVS
rely on keywords to function.

-- 
Rich

Re: [gentoo-dev] Removal of CVS headers

[Ulrich Mueller]

> On Sat, 25 Feb 2017, Sergei Trofimovich wrote:

> Typical questions for tree-wide cleanups:

> - Are new ebuilds forbidden to have '$Id$' or just discouraged?
> - [same as above] Will new version of repoman complain about
>   leftover '$Id$'?

Not sure. That would have to be controlled via layout.conf, in case
someone uses CVS or Subversion for their overlay?

> - Will the tool be available for people to cleanup their overlays?
>   sed 1-liner is good enough. Just make sure it will be in commit
>   message.

Good idea, I'll post my script once it's ready.

Ulrich


pgpRveN4KCTgj.pgp
Description: PGP signature

Re: [gentoo-dev] Removal of CVS headers

[Sergei Trofimovich]

On Sat, 25 Feb 2017 15:05:09 +0100
Ulrich Mueller  wrote:

> As the council has decided in its 2014-10-14 meeting (and confirmed
> again in the 2016-11-13 meeting), CVS headers should be removed after
> the migration to Git. Until recently, this was blocked by repoman
> still checking for the $Id$ line. The latter is now fixed in the
> stable repoman version.

Would be nice to have a link to the meeting notes. I guess it's:
https://projects.gentoo.org/council/meeting-logs/20141014-summary.txt

"""
Can we drop CVS headers post-migration?
Aye
"""

> Therefore, I am going to remove the remaining CVS headers throughout
> the tree (except for patches, of course) in two days from now.

Typical questions for tree-wide cleanups:

- Are new ebuilds forbidden to have '$Id$' or just discouraged?
- [same as above] Will new version of repoman complain about leftover '$Id$'?
- Will the tool be available for people to cleanup their overlays?
  sed 1-liner is good enough. Just make sure it will be in commit message.

Thanks!

-- 

  Sergei


pgpKKJaKir8M2.pgp
Description: Цифровая подпись OpenPGP

Re: [gentoo-dev] Removal of CVS headers

[Rich Freeman]

On Sat, Feb 25, 2017 at 9:05 AM, Ulrich Mueller  wrote:
>
> Therefore, I am going to remove the remaining CVS headers throughout
> the tree (except for patches, of course) in two days from now.
>

Speaking from git migration experience, I'd be really careful about
how you go about it.  Filtering the filename and ensuring only the top
few lines get touched would probably prevent a lot of issues.  Patches
are the biggest problem as you've noted, but offhand I'm not sure if
there were others.  I just remember that CVS keywords were a constant
headache during the migration as they seemed to pop up in places you
wouldn't think of.  Now, you're only concerned with the active tree so
that cuts down the opportunity for issues quite a bit (vs fixing all
the old commits as well).

-- 
Rich

[gentoo-dev] Removal of CVS headers

[Ulrich Mueller]

As the council has decided in its 2014-10-14 meeting (and confirmed
again in the 2016-11-13 meeting), CVS headers should be removed after
the migration to Git. Until recently, this was blocked by repoman
still checking for the $Id$ line. The latter is now fixed in the
stable repoman version.

Therefore, I am going to remove the remaining CVS headers throughout
the tree (except for patches, of course) in two days from now.

Ulrich


pgp74__LGVYF4.pgp
Description: PGP signature