* GitLab CI setup file in scratch/tzz/gitlab
@ 2017-04-26 18:45 Ted Zlatanov
2017-04-26 18:51 ` Eli Zaretskii
0 siblings, 1 reply; 51+ messages in thread
From: Ted Zlatanov @ 2017-04-26 18:45 UTC (permalink / raw)
To: emacs-devel
I've pushed a .gitlab-ci.yml file to the branch scratch/tzz/gitlab
Merging that branch and letting me and Toon maintain that file would
make it much easier to support https://gitlab.com/emacs-ci/emacs and
further evaluate its viability as a CI service.
John, Eli, any objections? I can add a disclaimer, if you wish, that the
presence of the file doesn't imply support or endorsement of GitLab. I
don't think it's necessary personally. Right now it just names me as the
maintainer.
Thanks
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-26 18:45 GitLab CI setup file in scratch/tzz/gitlab Ted Zlatanov
@ 2017-04-26 18:51 ` Eli Zaretskii
2017-04-26 20:55 ` dptd dptd
` (2 more replies)
0 siblings, 3 replies; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-26 18:51 UTC (permalink / raw)
To: emacs-devel
> From: Ted Zlatanov <tzz@lifelogs.com>
> Date: Wed, 26 Apr 2017 14:45:24 -0400
>
> I've pushed a .gitlab-ci.yml file to the branch scratch/tzz/gitlab
>
> Merging that branch and letting me and Toon maintain that file would
> make it much easier to support https://gitlab.com/emacs-ci/emacs and
> further evaluate its viability as a CI service.
>
> John, Eli, any objections?
Frankly, I don't understand why this file should be part of the Emacs
repository. For starters, it seems to be Debian-specific. And if it
should be in our repository, its place is definitely wrong, it should
be somewhere under admin/.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-26 18:51 ` Eli Zaretskii
@ 2017-04-26 20:55 ` dptd dptd
2017-04-27 5:01 ` Lars Brinkhoff
2017-04-27 16:16 ` Ted Zlatanov
2 siblings, 0 replies; 51+ messages in thread
From: dptd dptd @ 2017-04-26 20:55 UTC (permalink / raw)
To: tzz; +Cc: Eli Zaretskii, Emacs developers
> On 26 Apr 2017, at 20:51, Eli Zaretskii <eliz@gnu.org> wrote:
>
>> From: Ted Zlatanov <tzz@lifelogs.com>
>> Date: Wed, 26 Apr 2017 14:45:24 -0400
>>
>> I've pushed a .gitlab-ci.yml file to the branch scratch/tzz/gitlab
>>
>> Merging that branch and letting me and Toon maintain that file would
>> make it much easier to support https://gitlab.com/emacs-ci/emacs and
>> further evaluate its viability as a CI service.
>>
>> John, Eli, any objections?
>
> Frankly, I don't understand why this file should be part of the Emacs
> repository. For starters, it seems to be Debian-specific. And if it
> should be in our repository, its place is definitely wrong, it should
> be somewhere under admin/.
>
Sorry for asking stupid questions but… is this some kind of “official” emacs mirror on gitlab?
What is the purpose of it?
Thanks!
Best regards,
bd
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-26 18:51 ` Eli Zaretskii
2017-04-26 20:55 ` dptd dptd
@ 2017-04-27 5:01 ` Lars Brinkhoff
2017-04-27 14:23 ` Eli Zaretskii
2017-04-27 16:16 ` Ted Zlatanov
2 siblings, 1 reply; 51+ messages in thread
From: Lars Brinkhoff @ 2017-04-27 5:01 UTC (permalink / raw)
To: emacs-devel
Eli Zaretskii wrote:
>> Ted Zlatanov wrote:
>> I've pushed a .gitlab-ci.yml file to the branch scratch/tzz/gitlab
>> John, Eli, any objections?
>
> Frankly, I don't understand why this file should be part of the Emacs
> repository. For starters, it seems to be Debian-specific. And if it
> should be in our repository, its place is definitely wrong, it should
> be somewhere under admin/.
It's not Debian-specific. It does mention "debian", but that's only to
pick one out of many possible images to run the build.
It's probably required to be in the top-level directory to work with
GitLab-CI. I haven't checked GitLab specifically, but most continuous
integration services does have this requirement that the configuration
file must have a specific name, and must be in the root of the project.
With this file in place, GitLab can now run a full build and test
cycle for every commit:
https://gitlab.com/emacs-ci/emacs/builds/14381164
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 5:01 ` Lars Brinkhoff
@ 2017-04-27 14:23 ` Eli Zaretskii
2017-04-27 14:37 ` Dmitry Gutov
` (2 more replies)
0 siblings, 3 replies; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-27 14:23 UTC (permalink / raw)
To: Lars Brinkhoff; +Cc: emacs-devel
> From: Lars Brinkhoff <lars@nocrew.org>
> Date: Thu, 27 Apr 2017 07:01:40 +0200
>
> Eli Zaretskii wrote:
> >> Ted Zlatanov wrote:
> >> I've pushed a .gitlab-ci.yml file to the branch scratch/tzz/gitlab
> >> John, Eli, any objections?
> >
> > Frankly, I don't understand why this file should be part of the Emacs
> > repository. For starters, it seems to be Debian-specific. And if it
> > should be in our repository, its place is definitely wrong, it should
> > be somewhere under admin/.
>
> It's not Debian-specific. It does mention "debian", but that's only to
> pick one out of many possible images to run the build.
I actually meant the 'apt' commands. It's true that some other
distributions besides Debian support that, but AFAIK they have their
own commands to the same effects.
> It's probably required to be in the top-level directory to work with
> GitLab-CI. I haven't checked GitLab specifically, but most continuous
> integration services does have this requirement that the configuration
> file must have a specific name, and must be in the root of the project.
>
> With this file in place, GitLab can now run a full build and test
> cycle for every commit:
>
> https://gitlab.com/emacs-ci/emacs/builds/14381164
OK, but why should the Emacs repository keep this file and maintain
it? It looks simple enough for the interested users to have it on
their systems, and GitLab looks like a full repository anyway, so
maybe the file should be maintained there? Or we could have a small
note file in admin/notes telling the interested users to create such a
file if they want.
IOW, this file looks unrelated to Emacs, so I don't think I understand
why Ted wanted us to maintain it. I'm probably missing something.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 14:23 ` Eli Zaretskii
@ 2017-04-27 14:37 ` Dmitry Gutov
2017-04-27 14:59 ` Eli Zaretskii
2017-04-27 17:43 ` Lars Brinkhoff
2017-04-28 21:34 ` Richard Stallman
2 siblings, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-27 14:37 UTC (permalink / raw)
To: Eli Zaretskii, Lars Brinkhoff; +Cc: emacs-devel
On 27.04.2017 17:23, Eli Zaretskii wrote:
> I actually meant the 'apt' commands. It's true that some other
> distributions besides Debian support that, but AFAIK they have their
> own commands to the same effects.
Ubuntu also uses 'apt', and it's a popular choice for CI servers.
> OK, but why should the Emacs repository keep this file and maintain
> it?
It's a common practice for the projects using Travis or GitLab CI.
> It looks simple enough for the interested users to have it on
> their systems,
Would you ask all interested users to maintain their own configurations,
or even installations, of Hydra as well?
It sounds silly. We should have an accessible, friendly CI instance for
all our users.
> and GitLab looks like a full repository anyway, so
> maybe the file should be maintained there?
And then what? Create endless merge commits in its repo? Or endlessly
rebase the patch? Neither sounds very smart or necessary, and I've no
idea how easy that would be to automate.
In both cases, HEAD will become different from the tip of master in the
Emacs repository, and build failure emails will become less useful.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 14:37 ` Dmitry Gutov
@ 2017-04-27 14:59 ` Eli Zaretskii
2017-04-27 15:16 ` Dmitry Gutov
0 siblings, 1 reply; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-27 14:59 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: lars, emacs-devel
> Cc: emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Thu, 27 Apr 2017 17:37:20 +0300
>
> On 27.04.2017 17:23, Eli Zaretskii wrote:
>
> > I actually meant the 'apt' commands. It's true that some other
> > distributions besides Debian support that, but AFAIK they have their
> > own commands to the same effects.
>
> Ubuntu also uses 'apt', and it's a popular choice for CI servers.
>
> > OK, but why should the Emacs repository keep this file and maintain
> > it?
>
> It's a common practice for the projects using Travis or GitLab CI.
>
> > It looks simple enough for the interested users to have it on
> > their systems,
>
> Would you ask all interested users to maintain their own configurations,
> or even installations, of Hydra as well?
>
> It sounds silly. We should have an accessible, friendly CI instance for
> all our users.
I don't understand this attack. We didn't yet decide that this site
will be our CI site. Until we do, why should we function as someone
else's repository??
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 14:59 ` Eli Zaretskii
@ 2017-04-27 15:16 ` Dmitry Gutov
2017-04-28 21:33 ` Richard Stallman
0 siblings, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-27 15:16 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: lars, emacs-devel
On 27.04.2017 17:59, Eli Zaretskii wrote:
>> It sounds silly. We should have an accessible, friendly CI instance for
>> all our users.
>
> I don't understand this attack. We didn't yet decide that this site
> will be our CI site.
If we don't like it, we can easily delete the file. It's low-conflict,
low-maintenance. There's almost zero cost to adding it.
Further, the contents of the file are not tied to the "site", just to
GitLab as a project. It can be used by other GitLab installations
without changes.
> Until we do, why should we function as someone
> else's repository??
From that perspective, Ted has already done what you're asking
(committed the change into a different repository).
The first message of this thread is a request to take a further step, to
make a fair evaluation easier.
As a result of having this file in master, the GitLab instance will
build all later branches that are pushed to the repository (with some
possible exceptions).
I am restating here the second paragraph of that email, BTW.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-26 18:51 ` Eli Zaretskii
2017-04-26 20:55 ` dptd dptd
2017-04-27 5:01 ` Lars Brinkhoff
@ 2017-04-27 16:16 ` Ted Zlatanov
2018-07-23 13:14 ` Tino Calancha
2 siblings, 1 reply; 51+ messages in thread
From: Ted Zlatanov @ 2017-04-27 16:16 UTC (permalink / raw)
To: emacs-devel
On Wed, 26 Apr 2017 21:51:09 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
>> From: Ted Zlatanov <tzz@lifelogs.com>
>> Date: Wed, 26 Apr 2017 14:45:24 -0400
>>
>> I've pushed a .gitlab-ci.yml file to the branch scratch/tzz/gitlab
>>
>> Merging that branch and letting me and Toon maintain that file would
>> make it much easier to support https://gitlab.com/emacs-ci/emacs and
>> further evaluate its viability as a CI service.
EZ> Frankly, I don't understand why this file should be part of the Emacs
EZ> repository. For starters, it seems to be Debian-specific. And if it
EZ> should be in our repository, its place is definitely wrong, it should
EZ> be somewhere under admin/.
The contents and location are annoyingly necessary in order to try out
Gitlab as a CI service.
On Thu, 27 Apr 2017 17:23:54 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
EZ> OK, but why should the Emacs repository keep this file and maintain
EZ> it? It looks simple enough for the interested users to have it on
EZ> their systems, and GitLab looks like a full repository anyway, so
EZ> maybe the file should be maintained there? Or we could have a small
EZ> note file in admin/notes telling the interested users to create such a
EZ> file if they want.
Because it's a pain otherwise. If we use Gitlab CI more fully, we have
to keep it there. They are working on allowing the location to be
flexible: https://gitlab.com/gitlab-org/gitlab-ce/issues/15041
https://gitlab.com/gitlab-org/gitlab-ce/merge_requests/5682 so this is
probably not a long-term problem.
On Thu, 27 Apr 2017 17:59:30 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
EZ> I don't understand this attack. We didn't yet decide that this site
EZ> will be our CI site. Until we do, why should we function as someone
EZ> else's repository??
I think it's useful, in order to make that evaluation, to allow this
hiden file. I don't think it's a big burden for me, and don't expect
anyone else to maintain it.
On Wed, 26 Apr 2017 22:55:48 +0200 dptd dptd <dptdescribe@gmail.com> wrote:
dd> Sorry for asking stupid questions but… is this some kind of
“official” emacs mirror on gitlab? dd> What is the purpose of it?
Not an official anything. Just a way to evaluate Gitlab CI against
Hydra, Buildbot, and other CI solutions.
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 14:23 ` Eli Zaretskii
2017-04-27 14:37 ` Dmitry Gutov
@ 2017-04-27 17:43 ` Lars Brinkhoff
2017-04-27 20:16 ` Dmitry Gutov
2017-04-27 20:27 ` Eli Zaretskii
2017-04-28 21:34 ` Richard Stallman
2 siblings, 2 replies; 51+ messages in thread
From: Lars Brinkhoff @ 2017-04-27 17:43 UTC (permalink / raw)
To: emacs-devel
Eli Zaretskii wrote:
> I actually meant the 'apt' commands. It's true that some other
> distributions besides Debian support that, but AFAIK they have their
> own commands to the same effects.
Right, the build runs in a Debian image, so it must use Debian package
management commands. Whatever operating system you pick, you must
provide some instructions to install build dependencies.
> We didn't yet decide that this site will be our CI site. Until we do,
> why should we function as someone else's repository??
I think it would be reasonable to take that decision first, and add the
file second.
It has been demonstrated to work. Now it has to be cleared by Emacs
maintainers, FSF policies, etc.
> OK, but why should the Emacs repository keep this file and maintain
> it? It looks simple enough for the interested users to have it on
> their systems
It would be possible, but no that simple. There would to be a commit
(or patch) with the .gitlab-ci.yml file which would have to be rebased
on top of master (or any oher branch to be tested) every time the GitLab
repository is synchronized with savannah.
Yes, this could be done. But it seems to me that this would be brittle
and prone to failures. Someone would have to look after this.
If the file could just sit in the savannah repository, it would be more
likely to be useful.
> IOW, this file looks unrelated to Emacs, so I don't think I understand
> why Ted wanted us to maintain it. I'm probably missing something.
I have experimented with about a dozen different CI services, and to me
it seems quite reasonable that Emacs should have this file, IF the
maintainers decide that GitLab-CI is to be used for continuous
integration.
It's not unrelated to Emacs. It provides build and testing instructions
to an automated service. Those instuctions are specific to Emacs.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 17:43 ` Lars Brinkhoff
@ 2017-04-27 20:16 ` Dmitry Gutov
2017-04-27 20:42 ` Eli Zaretskii
2017-04-27 20:27 ` Eli Zaretskii
1 sibling, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-27 20:16 UTC (permalink / raw)
To: Lars Brinkhoff, emacs-devel
On 27.04.2017 20:43, Lars Brinkhoff wrote:
>> We didn't yet decide that this site will be our CI site. Until we do,
>> why should we function as someone else's repository??
>
> I think it would be reasonable to take that decision first, and add the
> file second.
Allow me to disagree: it's not necessary to make this decision first.
In fact, if we have several different CI servers popular among Emacs
developers, it wouldn't cost us much to keep all their build
configurations in the repo (as long as they are configured in the same
fashion).
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 17:43 ` Lars Brinkhoff
2017-04-27 20:16 ` Dmitry Gutov
@ 2017-04-27 20:27 ` Eli Zaretskii
1 sibling, 0 replies; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-27 20:27 UTC (permalink / raw)
To: Lars Brinkhoff; +Cc: emacs-devel
> From: Lars Brinkhoff <lars@nocrew.org>
> Date: Thu, 27 Apr 2017 19:43:36 +0200
>
> > We didn't yet decide that this site will be our CI site. Until we do,
> > why should we function as someone else's repository??
>
> I think it would be reasonable to take that decision first, and add the
> file second.
I obviously agree. When we've made that decision, adding the file
becomes a no-brainer.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 20:16 ` Dmitry Gutov
@ 2017-04-27 20:42 ` Eli Zaretskii
2017-04-27 23:20 ` Dmitry Gutov
0 siblings, 1 reply; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-27 20:42 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: lars, emacs-devel
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Thu, 27 Apr 2017 23:16:22 +0300
>
> In fact, if we have several different CI servers popular among Emacs
> developers, it wouldn't cost us much to keep all their build
> configurations in the repo (as long as they are configured in the same
> fashion).
Nothing prevents us from deciding that we want to support several CI
servers, not just one.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 20:42 ` Eli Zaretskii
@ 2017-04-27 23:20 ` Dmitry Gutov
2017-04-28 6:49 ` Eli Zaretskii
0 siblings, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-27 23:20 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: lars, emacs-devel
On 27.04.2017 23:42, Eli Zaretskii wrote:
> Nothing prevents us from deciding that we want to support several CI
> servers, not just one.
Then my point is that this kind of decision should be like "yeah, ok,
this looks useful", instead of taking days or even weeks of deliberation.
Like mentioned, the cost for the project is very low.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 23:20 ` Dmitry Gutov
@ 2017-04-28 6:49 ` Eli Zaretskii
2017-04-28 8:33 ` Dmitry Gutov
0 siblings, 1 reply; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-28 6:49 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: lars, emacs-devel
> Cc: lars@nocrew.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Fri, 28 Apr 2017 02:20:23 +0300
>
> On 27.04.2017 23:42, Eli Zaretskii wrote:
>
> > Nothing prevents us from deciding that we want to support several CI
> > servers, not just one.
>
> Then my point is that this kind of decision should be like "yeah, ok,
> this looks useful", instead of taking days or even weeks of deliberation.
AFAIU, it's not the decision that takes days or weeks, it's the actual
work on using the server and collecting the experience which would
allow the decision to be made.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 6:49 ` Eli Zaretskii
@ 2017-04-28 8:33 ` Dmitry Gutov
2017-04-28 9:45 ` Eli Zaretskii
0 siblings, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-28 8:33 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: lars, emacs-devel
On 28.04.2017 9:49, Eli Zaretskii wrote:
> AFAIU, it's not the decision that takes days or weeks, it's the actual
> work on using the server and collecting the experience which would
> allow the decision to be made.
In that case, again, to use the CI to its fullest, the file should be in
the repo.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 8:33 ` Dmitry Gutov
@ 2017-04-28 9:45 ` Eli Zaretskii
2017-04-28 9:58 ` Dmitry Gutov
2017-04-28 12:47 ` Ted Zlatanov
0 siblings, 2 replies; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-28 9:45 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: lars, emacs-devel
> Cc: lars@nocrew.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Fri, 28 Apr 2017 11:33:28 +0300
>
> On 28.04.2017 9:49, Eli Zaretskii wrote:
>
> > AFAIU, it's not the decision that takes days or weeks, it's the actual
> > work on using the server and collecting the experience which would
> > allow the decision to be made.
>
> In that case, again, to use the CI to its fullest, the file should be in
> the repo.
It already is -- on a branch. The question was whether to have that
on master, which IMO should be done after we decide to use this
server.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 9:45 ` Eli Zaretskii
@ 2017-04-28 9:58 ` Dmitry Gutov
2017-04-28 13:17 ` Eli Zaretskii
2017-04-28 12:47 ` Ted Zlatanov
1 sibling, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-28 9:58 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: lars, emacs-devel
On 28.04.2017 12:45, Eli Zaretskii wrote:
>> In that case, again, to use the CI to its fullest, the file should be in
>> the repo.
>
> It already is -- on a branch.
This way we cannot evaluate how handy it is to use it "live", for our
current work.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 9:45 ` Eli Zaretskii
2017-04-28 9:58 ` Dmitry Gutov
@ 2017-04-28 12:47 ` Ted Zlatanov
2017-04-28 13:24 ` Lars Brinkhoff
1 sibling, 1 reply; 51+ messages in thread
From: Ted Zlatanov @ 2017-04-28 12:47 UTC (permalink / raw)
To: emacs-devel
On Fri, 28 Apr 2017 12:58:21 +0300 Dmitry Gutov <dgutov@yandex.ru> wrote:
DG> On 28.04.2017 12:45, Eli Zaretskii wrote:
>>> In that case, again, to use the CI to its fullest, the file should be in
>>> the repo.
>>
>> It already is -- on a branch.
DG> This way we cannot evaluate how handy it is to use it "live", for our current
DG> work.
Exactly. The presence of this file on a branch lets us test that branch,
so we can't use it on "master" without merging it there. As I mentioned,
Gitlab developers are working on making the location of the file
flexible, but for now it's a reasonable inconvenience to give us a
viable evaluation of Gitlab CI on the "master" branch.
The alternative is for me or someone else to constantly rebase and push
my branch with that commit on top of "master." That's just annoying.
Thanks
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 9:58 ` Dmitry Gutov
@ 2017-04-28 13:17 ` Eli Zaretskii
2017-04-28 13:35 ` Dmitry Gutov
0 siblings, 1 reply; 51+ messages in thread
From: Eli Zaretskii @ 2017-04-28 13:17 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: lars, emacs-devel
> Cc: lars@nocrew.org, emacs-devel@gnu.org
> From: Dmitry Gutov <dgutov@yandex.ru>
> Date: Fri, 28 Apr 2017 12:58:21 +0300
>
> On 28.04.2017 12:45, Eli Zaretskii wrote:
>
> >> In that case, again, to use the CI to its fullest, the file should be in
> >> the repo.
> >
> > It already is -- on a branch.
>
> This way we cannot evaluate how handy it is to use it "live", for our
> current work.
Of course, you can: just switch to that branch, and perhaps also merge
master, and there you go.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 12:47 ` Ted Zlatanov
@ 2017-04-28 13:24 ` Lars Brinkhoff
2017-04-28 14:16 ` Ted Zlatanov
0 siblings, 1 reply; 51+ messages in thread
From: Lars Brinkhoff @ 2017-04-28 13:24 UTC (permalink / raw)
To: emacs-devel
Ted Zlatanov writes:
> The alternative is for me or someone else to constantly rebase and
> push my branch with that commit on top of "master." That's just
> annoying.
Yes, it would be annoying.
That said, couldn't you run this on your favourite online machine?
git checkout master
while :; do
git fetch savannah
git rebase savannah/master
git push gitlab
sleep 300
done
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 13:17 ` Eli Zaretskii
@ 2017-04-28 13:35 ` Dmitry Gutov
0 siblings, 0 replies; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-28 13:35 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: lars, emacs-devel
On 28.04.2017 16:17, Eli Zaretskii wrote:
>> This way we cannot evaluate how handy it is to use it "live", for our
>> current work.
>
> Of course, you can: just switch to that branch, and perhaps also merge
> master, and there you go.
My "current work" is not in that branch. And, for the sake of the
argument, not in master either.
If you do that, you also can't fully use the nice build failure emails
because they will point to wrong commits.
Anyway, you seem adamant about making the decision first, and I'm
repeating myself already.
Please go ahead and make that decision, and we can resume the argument
if the decision turns out to be negative.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 13:24 ` Lars Brinkhoff
@ 2017-04-28 14:16 ` Ted Zlatanov
0 siblings, 0 replies; 51+ messages in thread
From: Ted Zlatanov @ 2017-04-28 14:16 UTC (permalink / raw)
To: emacs-devel
On Fri, 28 Apr 2017 15:24:24 +0200 Lars Brinkhoff <lars@nocrew.org> wrote:
LB> Ted Zlatanov writes:
>> The alternative is for me or someone else to constantly rebase and
>> push my branch with that commit on top of "master." That's just
>> annoying.
LB> Yes, it would be annoying.
LB> That said, couldn't you run this on your favourite online machine?
LB> while :; do [... rebase and push ...]
No.
I think this direction is a distraction from the real goal: evaluating
Gitlab CI against Hydra and Buildbot. If we really want to evaluate the
tool, let's evaluate it like it's supposed to be used.
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 15:16 ` Dmitry Gutov
@ 2017-04-28 21:33 ` Richard Stallman
2017-04-29 0:41 ` Ted Zlatanov
2017-04-29 1:29 ` Mike Gerwitz
0 siblings, 2 replies; 51+ messages in thread
From: Richard Stallman @ 2017-04-28 21:33 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: lars, eliz, emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
If we install files to enable people to use a variety of CI sites,
including GitLab among them might not matter.
However, to install just one file for one site is tantamount to
selecting that as "the" CI site for Emacs. As has been pointed out,
using GitLab for this purpose would be SaaSS.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 14:23 ` Eli Zaretskii
2017-04-27 14:37 ` Dmitry Gutov
2017-04-27 17:43 ` Lars Brinkhoff
@ 2017-04-28 21:34 ` Richard Stallman
2 siblings, 0 replies; 51+ messages in thread
From: Richard Stallman @ 2017-04-28 21:34 UTC (permalink / raw)
To: Eli Zaretskii; +Cc: lars, emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> I actually meant the 'apt' commands. It's true that some other
> distributions besides Debian support that, but AFAIK they have their
> own commands to the same effects.
Trisquel uses the apt command.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 21:33 ` Richard Stallman
@ 2017-04-29 0:41 ` Ted Zlatanov
2017-04-29 15:58 ` Richard Stallman
2017-04-29 18:22 ` Lars Brinkhoff
2017-04-29 1:29 ` Mike Gerwitz
1 sibling, 2 replies; 51+ messages in thread
From: Ted Zlatanov @ 2017-04-29 0:41 UTC (permalink / raw)
To: emacs-devel
On Fri, 28 Apr 2017 17:33:54 -0400 Richard Stallman <rms@gnu.org> wrote:
RS> If we install files to enable people to use a variety of CI sites,
RS> including GitLab among them might not matter.
RS> However, to install just one file for one site is tantamount to
RS> selecting that as "the" CI site for Emacs. As has been pointed out,
RS> using GitLab for this purpose would be SaaSS.
Great. Only Gitlab requires a file in the master branch, out of the
three we're evaluating (the other two are Hydra and Buildbot). I could
add a .travis.yml for laughs, I guess.
Or as I suggested, I can just add a comment that makes it clear that
Gitlab is not implicitly endorsed by the presence of this file. Would
that be enough?
Thanks
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-28 21:33 ` Richard Stallman
2017-04-29 0:41 ` Ted Zlatanov
@ 2017-04-29 1:29 ` Mike Gerwitz
1 sibling, 0 replies; 51+ messages in thread
From: Mike Gerwitz @ 2017-04-29 1:29 UTC (permalink / raw)
To: Richard Stallman; +Cc: lars, eliz, emacs-devel, Dmitry Gutov
[-- Attachment #1: Type: text/plain, Size: 1348 bytes --]
On Fri, Apr 28, 2017 at 17:33:54 -0400, Richard Stallman wrote:
> If we install files to enable people to use a variety of CI sites,
> including GitLab among them might not matter.
>
> However, to install just one file for one site is tantamount to
> selecting that as "the" CI site for Emacs. As has been pointed out,
> using GitLab for this purpose would be SaaSS.
That file would be recognized by _any_ Gitlab instance, including
self-hosted. Of course, that also means it would be recognized by
Gitlab.com's own instance. I've used Gitlab's pipelines self-hosted,
but haven't on Gitlab.com, so I don't know if it needs to be explicitly
enabled or if it automatically takes effect.
My point being: the presence of that file doesn't itself imply SaaSS in
the case of Gitlab. But it might be worth documenting that fact
explicitly somewhere.
In the case of a service like Travis CI, though, I think that's only
available as SaaSS via GitHub---I haven't heard of anyone self-hosting
it, though it seems like it might technically be possible to do so.[0]
Therefore I think that a Travis configuration file would effectively be
endorsing its SaaSS service.
--
Mike Gerwitz
Free Software Hacker+Activist | GNU Maintainer & Volunteer
GPG: D6E9 B930 028A 6C38 F43B 2388 FEF6 3574 5E6F 6D05
https://mikegerwitz.com
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 818 bytes --]
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-29 0:41 ` Ted Zlatanov
@ 2017-04-29 15:58 ` Richard Stallman
2017-04-29 17:44 ` Dmitry Gutov
2017-04-29 18:22 ` Lars Brinkhoff
1 sibling, 1 reply; 51+ messages in thread
From: Richard Stallman @ 2017-04-29 15:58 UTC (permalink / raw)
To: emacs-devel; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> Great. Only Gitlab requires a file in the master branch, out of the
> three we're evaluating (the other two are Hydra and Buildbot).
Are those SaaSS too?
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-29 15:58 ` Richard Stallman
@ 2017-04-29 17:44 ` Dmitry Gutov
2017-04-30 22:19 ` Richard Stallman
0 siblings, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-29 17:44 UTC (permalink / raw)
To: rms, emacs-devel
On 29.04.2017 18:58, Richard Stallman wrote:
> [[[ To any NSA and FBI agents reading my email: please consider ]]]
> [[[ whether defending the US Constitution against all enemies, ]]]
> [[[ foreign or domestic, requires you to follow Snowden's example. ]]]
>
> > Great. Only Gitlab requires a file in the master branch, out of the
> > three we're evaluating (the other two are Hydra and Buildbot).
>
> Are those SaaSS too?
That's a trick question. Gitlab is not SaaSS. It has a community version
(MIT licensed) which we are free to use.
Buildbot is GPLv2. Hydra is GPLv3.
None of them, of course, normally run on an ordinary user's computer.
But they could.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-29 0:41 ` Ted Zlatanov
2017-04-29 15:58 ` Richard Stallman
@ 2017-04-29 18:22 ` Lars Brinkhoff
2017-04-29 21:11 ` Ted Zlatanov
2017-04-30 22:20 ` Richard Stallman
1 sibling, 2 replies; 51+ messages in thread
From: Lars Brinkhoff @ 2017-04-29 18:22 UTC (permalink / raw)
To: emacs-devel
Ted Zlatanov <tzz@lifelogs.com> writes:
> I could add a .travis.yml for laughs, I guess.
Travis CI isn't free software, is it?
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-29 18:22 ` Lars Brinkhoff
@ 2017-04-29 21:11 ` Ted Zlatanov
2017-04-30 22:20 ` Richard Stallman
1 sibling, 0 replies; 51+ messages in thread
From: Ted Zlatanov @ 2017-04-29 21:11 UTC (permalink / raw)
To: emacs-devel
On Sat, 29 Apr 2017 20:22:23 +0200 Lars Brinkhoff <lars@nocrew.org> wrote:
LB> Ted Zlatanov <tzz@lifelogs.com> writes:
>> I could add a .travis.yml for laughs, I guess.
LB> Travis CI isn't free software, is it?
It was a joke, not a serious suggestion. Sorry that wasn't clear.
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-29 17:44 ` Dmitry Gutov
@ 2017-04-30 22:19 ` Richard Stallman
2017-04-30 23:55 ` Dmitry Gutov
0 siblings, 1 reply; 51+ messages in thread
From: Richard Stallman @ 2017-04-30 22:19 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> > > Great. Only Gitlab requires a file in the master branch, out of the
> > > three we're evaluating (the other two are Hydra and Buildbot).
> >
> > Are those SaaSS too?
> That's a trick question. Gitlab is not SaaSS.
It is an honest and serious question. However, I should
have distinguishd more clearly between the software called GitLab
and the service called GitLab.
The GitLab software is not SaaSS, because it's not a service.
It is a program. One version of that program is free.
It is fine for us to run that program if it serves our purpose.
To use a commercial service for this would be SaaSS, even
if it is running _the exact same program_
SaaSS is not a question of which programs are running on the server.
We can't verify which programs are running on anyone else's server.
We can't change them either, even if they are free software.
Please see https://gnu.org/philosophy/who-does-that-server-really-serve.html
for an explanation of the SaaSS issue.
Perhaps the FSF should find volunteers to set up a server to do CI for
GNU packages using the GitLab software. Would that be useful?
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-29 18:22 ` Lars Brinkhoff
2017-04-29 21:11 ` Ted Zlatanov
@ 2017-04-30 22:20 ` Richard Stallman
1 sibling, 0 replies; 51+ messages in thread
From: Richard Stallman @ 2017-04-30 22:20 UTC (permalink / raw)
To: Lars Brinkhoff; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> Travis CI isn't free software, is it?
My understanding is that CI involves use of a service.
Isn't that right?
A service is not software, so it is not meaningful to ask whether a
service is free or not.
See https://gnu.org/philosophy/network-services-arent-free-or-nonfree.html
for explanation of this point.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-30 22:19 ` Richard Stallman
@ 2017-04-30 23:55 ` Dmitry Gutov
2017-05-01 1:53 ` Ted Zlatanov
2017-05-01 15:54 ` Richard Stallman
0 siblings, 2 replies; 51+ messages in thread
From: Dmitry Gutov @ 2017-04-30 23:55 UTC (permalink / raw)
To: rms; +Cc: emacs-devel
On 01.05.2017 1:19, Richard Stallman wrote:
> > > > Great. Only Gitlab requires a file in the master branch, out of the
> > > > three we're evaluating (the other two are Hydra and Buildbot).
> > >
> > > Are those SaaSS too?
>
> > That's a trick question. Gitlab is not SaaSS.
>
> It is an honest and serious question. However, I should
> have distinguishd more clearly between the software called GitLab
> and the service called GitLab.
It's an important distinction, yes. And I don't think the question of
whether Buildbot and Hydra are "SaaSS too" is important for our
purposes. The important question is whether we can run them for ourselves.
Somewhere, I'm sure some company (or several ones) provide them as paid
services.
> The GitLab software is not SaaSS, because it's not a service.
> It is a program. One version of that program is free.
> It is fine for us to run that program if it serves our purpose.
Naturally.
> To use a commercial service for this would be SaaSS, even
> if it is running _the exact same program_
We can call it SaaS, which is the industry-standard term. To use the
disparaging variation, against a well-behaving company that distributes
a fully-functional version of said software with a Free license, I
wouldn't consider fair.
> SaaSS is not a question of which programs are running on the server.
> We can't verify which programs are running on anyone else's server.
> We can't change them either, even if they are free software.
That's also true, for many users, of the GNU hosted installation of
Hydra. And it will remain true, for most Emacs developers and users, of
the GitLab installation on the FSF premises, when it's set up.
And yet, I wouldn't be comfortable calling them SaaSS.
> Please see https://gnu.org/philosophy/who-does-that-server-really-serve.html
> for an explanation of the SaaSS issue.
Already read it, thanks.
> Perhaps the FSF should find volunteers to set up a server to do CI for
> GNU packages using the GitLab software. Would that be useful?
That has always been the idea, yes. We are only using the GitLab service
to evaluate the features of GitLab the program.
Like already mentioned, by the way, the contents of .gitlab-ci.yml,
which is proposed for adding into the Emacs repository, are not going to
change after that transition.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-30 23:55 ` Dmitry Gutov
@ 2017-05-01 1:53 ` Ted Zlatanov
2017-05-01 15:54 ` Richard Stallman
1 sibling, 0 replies; 51+ messages in thread
From: Ted Zlatanov @ 2017-05-01 1:53 UTC (permalink / raw)
To: emacs-devel
On Mon, 1 May 2017 02:55:16 +0300 Dmitry Gutov <dgutov@yandex.ru> wrote:
DG> On 01.05.2017 1:19, Richard Stallman wrote:
>> Perhaps the FSF should find volunteers to set up a server to do CI for
>> GNU packages using the GitLab software. Would that be useful?
DG> That has always been the idea, yes. We are only using the GitLab service to
DG> evaluate the features of GitLab the program.
Thank you for suggesting that, Richard. I've been asking for a machine
for months, see the emacs-devel and savannah-admins threads.
There should be no problem with switching to a FSF/GNU host for the CI
service, when and if such a host becomes available.
I don't think the choice of CI system matters, this is a necessary step
regardless.
Thanks
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-30 23:55 ` Dmitry Gutov
2017-05-01 1:53 ` Ted Zlatanov
@ 2017-05-01 15:54 ` Richard Stallman
2017-05-02 12:42 ` Richard Stallman
2017-05-03 0:06 ` Dmitry Gutov
1 sibling, 2 replies; 51+ messages in thread
From: Richard Stallman @ 2017-05-01 15:54 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> > To use a commercial service for this would be SaaSS, even
> > if it is running _the exact same program_
> We can call it SaaS, which is the industry-standard term.
It's a standard term for a different meaning, a meaning that right for
this point. That is why I stopped using it.
SaaSS, which stands for Service as a Software Substitute, refers to a
specific kind of situation which raises a specific issue, described in
https://gnu.org/philosophy/who-does-that-server-really-serve.html.
SaaS, or Software as a Service, includes SaaSS and other things too.
I don't have a clear picture of how far that territory goes, so I
don't want to state a judgment of all kinds of SaaS.
> > SaaSS is not a question of which programs are running on the server.
> > We can't verify which programs are running on anyone else's server.
> > We can't change them either, even if they are free software.
> That's also true, for many users, of the GNU hosted installation of
> Hydra. And it will remain true, for most Emacs developers and users, of
> the GitLab installation on the FSF premises, when it's set up.
This server will not be SaaSS because it will be used by the GNU
Project (specifically, GNU Emacs) and run by the GNU Project.
If individual users used it for CI for their own non-GNU projects,
that would be SaaSS.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-01 15:54 ` Richard Stallman
@ 2017-05-02 12:42 ` Richard Stallman
2017-05-03 0:06 ` Dmitry Gutov
1 sibling, 0 replies; 51+ messages in thread
From: Richard Stallman @ 2017-05-02 12:42 UTC (permalink / raw)
To: rms; +Cc: emacs-devel, dgutov
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> It's a standard term for a different meaning, a meaning that right for
> this point. That is why I stopped using it.
should have been
> It's a standard term for a different meaning, a meaning that isn't
> right for this point. That is why I stopped using it.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-01 15:54 ` Richard Stallman
2017-05-02 12:42 ` Richard Stallman
@ 2017-05-03 0:06 ` Dmitry Gutov
2017-05-03 19:15 ` Richard Stallman
1 sibling, 1 reply; 51+ messages in thread
From: Dmitry Gutov @ 2017-05-03 0:06 UTC (permalink / raw)
To: rms; +Cc: emacs-devel
On 01.05.2017 18:54, Richard Stallman wrote:
> This server will not be SaaSS because it will be used by the GNU
> Project (specifically, GNU Emacs) and run by the GNU Project.
Thanks for clarifying. Either way, it seems to be a well-considered
position.
Any news on setting up a GitLab server for GNU packages?
And with that goal in mind, do you sanctify adding .gitlab-ci.yml into
Emacs's repo?
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-03 0:06 ` Dmitry Gutov
@ 2017-05-03 19:15 ` Richard Stallman
2017-05-14 2:25 ` Ted Zlatanov
0 siblings, 1 reply; 51+ messages in thread
From: Richard Stallman @ 2017-05-03 19:15 UTC (permalink / raw)
To: Dmitry Gutov; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> Any news on setting up a GitLab server for GNU packages?
FSF people should contact the Emacs developers to find people
who want to adnminister it as a VM.
> And with that goal in mind, do you sanctify adding .gitlab-ci.yml into
> Emacs's repo?
It is ok, but please add comments to explain that this is anticioating
the GitLab server that the FSF will run for this purpose.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-03 19:15 ` Richard Stallman
@ 2017-05-14 2:25 ` Ted Zlatanov
2017-05-15 1:44 ` Richard Stallman
0 siblings, 1 reply; 51+ messages in thread
From: Ted Zlatanov @ 2017-05-14 2:25 UTC (permalink / raw)
To: emacs-devel
On Wed, 03 May 2017 15:15:42 -0400 Richard Stallman <rms@gnu.org> wrote:
>> And with that goal in mind, do you sanctify adding .gitlab-ci.yml into
>> Emacs's repo?
RS> It is ok, but please add comments to explain that this is anticioating
RS> the GitLab server that the FSF will run for this purpose.
Sorry for the delay! I've now pushed this to master.
I added a copyright notice (assuming it's needed) and a disclaimer that
there is no endorsement implied:
# GitLab CI support for GNU Emacs
# The presence of this file does not imply any FSF/GNU endorsement of
# GitLab CI or any related entities and is intended only for
# evaluation purposes.
I didn't think it needed anything else because the forthcoming GitLab
server is too specific a task. But if you disagree, please go ahead and
add or edit the message.
Thanks
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-14 2:25 ` Ted Zlatanov
@ 2017-05-15 1:44 ` Richard Stallman
2017-05-15 20:08 ` Ted Zlatanov
` (2 more replies)
0 siblings, 3 replies; 51+ messages in thread
From: Richard Stallman @ 2017-05-15 1:44 UTC (permalink / raw)
To: emacs-devel; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> # GitLab CI support for GNU Emacs
> # The presence of this file does not imply any FSF/GNU endorsement of
> # GitLab CI or any related entities and is intended only for
> # evaluation purposes.
As I understand it, "GitLab CI" is the name of a suite of programs
that anyone can run, and also the name of a particular service.
That ambiguity came up in the discussion. So this message should not
be ambiguous.
Would someone please change this to
> # GNU Emacs support for the GitLab protocol for CI
> # The presence of this file does not imply any FSF/GNU endorsement of
> # any particular servce that uses that protocol. Also, it is intended for
> # evaluation purposes, thus possibly temporary.
(I would do it myself if using git were easy for me.)
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-15 1:44 ` Richard Stallman
@ 2017-05-15 20:08 ` Ted Zlatanov
2017-05-18 1:12 ` Perry E. Metzger
2017-05-21 15:19 ` Tino Calancha
2 siblings, 0 replies; 51+ messages in thread
From: Ted Zlatanov @ 2017-05-15 20:08 UTC (permalink / raw)
To: emacs-devel
On Sun, 14 May 2017 21:44:25 -0400 Richard Stallman <rms@gnu.org> wrote:
RS> As I understand it, "GitLab CI" is the name of a suite of programs
RS> that anyone can run, and also the name of a particular service.
RS> That ambiguity came up in the discussion. So this message should not
RS> be ambiguous.
Calling it a "GitLab protocol" is best, you're right.
RS> Would someone please change this to [...]
Sure--done.
Ted
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-15 1:44 ` Richard Stallman
2017-05-15 20:08 ` Ted Zlatanov
@ 2017-05-18 1:12 ` Perry E. Metzger
2017-05-19 1:19 ` Richard Stallman
2017-05-21 15:19 ` Tino Calancha
2 siblings, 1 reply; 51+ messages in thread
From: Perry E. Metzger @ 2017-05-18 1:12 UTC (permalink / raw)
To: Richard Stallman; +Cc: emacs-devel
On Sun, 14 May 2017 21:44:25 -0400 Richard Stallman <rms@gnu.org>
wrote:
>
> (I would do it myself if using git were easy for me.)
>
Just as an aside, what are you finding difficult? It would be best
for us to help you get past the learning curve, I think. (There are
also fantastic Emacs tools for using Git, like magit mode.)
Perry
--
Perry E. Metzger perry@piermont.com
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-18 1:12 ` Perry E. Metzger
@ 2017-05-19 1:19 ` Richard Stallman
2017-05-19 1:53 ` Perry E. Metzger
0 siblings, 1 reply; 51+ messages in thread
From: Richard Stallman @ 2017-05-19 1:19 UTC (permalink / raw)
To: Perry E. Metzger; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
Thanks for the offer, but it would not be feasible for anyone to help
me deal with git while communicating by email. We would need to be in
the same room.
As for magit, we don't have legal papers to include it in Emacs,
so please don't encourage poeple to use it.
To keep Emacs honest, we must promote the tools we can include in
Emacs. If those seem insufficient, then we should improve them.
Perhaps that includes replacing magit (if we can't get papers for
magit).
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-19 1:19 ` Richard Stallman
@ 2017-05-19 1:53 ` Perry E. Metzger
2017-05-21 3:25 ` Richard Stallman
0 siblings, 1 reply; 51+ messages in thread
From: Perry E. Metzger @ 2017-05-19 1:53 UTC (permalink / raw)
To: Richard Stallman; +Cc: emacs-devel
On Thu, 18 May 2017 21:19:36 -0400 Richard Stallman <rms@gnu.org>
wrote:
> Thanks for the offer, but it would not be feasible for anyone to
> help me deal with git while communicating by email. We would need
> to be in the same room.
Although I can't be in the same room as you, I'm sure getting someone
else to you in Cambridge to tutor you through it would be
straightforward.
> As for magit, we don't have legal papers to include it in Emacs,
> so please don't encourage poeple to use it.
I'm curious as to why. It is GPL, even GPLv3. Given that it is free
software, is there a good reason why one should discourage use of
it?
> To keep Emacs honest, we must promote the tools we can include in
> Emacs. If those seem insufficient, then we should improve them.
> Perhaps that includes replacing magit (if we can't get papers for
> magit).
Has it been tried?
Perry
--
Perry E. Metzger perry@piermont.com
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-19 1:53 ` Perry E. Metzger
@ 2017-05-21 3:25 ` Richard Stallman
2017-05-21 3:49 ` Jean-Christophe Helary
0 siblings, 1 reply; 51+ messages in thread
From: Richard Stallman @ 2017-05-21 3:25 UTC (permalink / raw)
To: Perry E. Metzger; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> > As for magit, we don't have legal papers to include it in Emacs,
> > so please don't encourage poeple to use it.
> I'm curious as to why. It is GPL, even GPLv3. Given that it is free
> software, is there a good reason why one should discourage use of
> it?
We want the Emacs distribution to be complete -- to contain everything
needed to use Emacs in the _recommended_ way, aside from low level
general-purpose facilities that Emacs uses (and other programs use).
We can't include magit in Emacs. If we treated magit as part of the
way we recommend for people to use Emacs, that creates a problem
since Emacs could not be complete.
> > To keep Emacs honest, we must promote the tools we can include in
> > Emacs. If those seem insufficient, then we should improve them.
> > Perhaps that includes replacing magit (if we can't get papers for
> > magit).
> Has it been tried?
I think that a year or two ago people found it would be hard to get
papers from the developers of magit.
I don't think people have tried replacing the parts of magit that we
can't get papers for. Maybe that is a good approach to use.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-21 3:25 ` Richard Stallman
@ 2017-05-21 3:49 ` Jean-Christophe Helary
2017-05-22 12:15 ` Richard Stallman
0 siblings, 1 reply; 51+ messages in thread
From: Jean-Christophe Helary @ 2017-05-21 3:49 UTC (permalink / raw)
To: emacs-devel
> On May 21, 2017, at 12:25, Richard Stallman <rms@gnu.org> wrote:
>
> We want the Emacs distribution to be complete -- to contain everything
> needed to use Emacs in the _recommended_ way, aside from low level
> general-purpose facilities that Emacs uses (and other programs use).
Are there references that document the _recommended_ way ?
Jean-Christophe Helary
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-15 1:44 ` Richard Stallman
2017-05-15 20:08 ` Ted Zlatanov
2017-05-18 1:12 ` Perry E. Metzger
@ 2017-05-21 15:19 ` Tino Calancha
2 siblings, 0 replies; 51+ messages in thread
From: Tino Calancha @ 2017-05-21 15:19 UTC (permalink / raw)
To: Richard Stallman; +Cc: Emacs developers
On Sun, 14 May 2017, Richard Stallman wrote:
> > # GitLab CI support for GNU Emacs
>
> > # The presence of this file does not imply any FSF/GNU endorsement of
> > # GitLab CI or any related entities and is intended only for
> > # evaluation purposes.
>
> As I understand it, "GitLab CI" is the name of a suite of programs
> that anyone can run, and also the name of a particular service.
> That ambiguity came up in the discussion. So this message should not
> be ambiguous.
>
> Would someone please change this to
>
> > # GNU Emacs support for the GitLab protocol for CI
>
> > # The presence of this file does not imply any FSF/GNU endorsement of
> > # any particular service that uses that protocol. Also, it is intended for
> > # evaluation purposes, thus possibly temporary.
>
> (I would do it myself if using git were easy for me.)
> it would not be feasible for anyone to help
> me deal with git while communicating by email. We would need to be in
> the same room.
Hi Richard,
You can do this kind of things with easy using gited.el without
knowing Git in detail. This is a simple lib without big pretensions,
just written to help me with my typical Git usage.
Due to its simplicity, it's really easy to start using it: specially
for people familiar with Dired.
You can think on gited.el as a Git interfaz 'a la Dired'. In Gited,
your branches are displayed in a similar way as the files are shown
in Dired. With a few exceptions, Gited commands with an
equivalent Dired command share same keybindings.
=== Following is how to push your .gitlab-ci.xml changes above with Gited ===
I) Assuming you have cloned the Emacs repository in your
local machine, then call from that directory:
M-x gited-list-branches RET
;; This creates a buffer *gited* with your local branches: 'master' branch,
;; and possible more if you have created others.
II) From the *gited* buffer:
c master RET ; Checkout the master branch.
* < ; Pull the latest changes in the remote repository.
III) Update .gitlab-ci.xml with your changes and save this file.
IV) From the *gited* buffer:
A ; Stage your changes.
C-c c ; Commit your changes (you are prompted for the commit message).
* > ; Push your local changes to the remote repository.
I am OK to add this lib to Elpa in case people find it useful.
Cheers,
Tino
Here is the latest gited.el version:
--8<-----------------------------cut here---------------start------------->8---
;;; gited.el --- Operate on Git branches like dired -*- lexical-binding:t -*-
;;
;; Filename: gited.el
;; Description: Operate on Git branches like dired
;;
;; Author: Tino Calancha <tino.calancha@gmail.com>
;; Maintainer: Tino Calancha <tino.calancha@gmail.com>
;; Copyright (C) 2016-2017, Tino Calancha, all rights reserved.
;; Created: Wed Oct 26 01:28:54 JST 2016
;; Compatibility: GNU Emacs: 24.4
;; Version: 0.1
;; Package-Requires: ((emacs "24.4") (cl-lib "0.5"))
;; Last-Updated: Sun May 21 22:53:34 JST 2017
;; By: calancha
;; Update #: 605
;;
;; Features that might be required by this library:
;;
;; `vc-git', `cl-lib', `dired', `tabulated-list',
;; `find-func'.
;;
;; Keywords: vc, convenience
;;
;;
;;; Commentary:
;;
;; This library lists the branches in a Git repository. Then you can
;; operate on them with a dired-like interface.
;;
;; The command `gited-list-branches' prompts for the kind of branch
;; (local branches, remote branches or tags) and lists them.
;; This command is used quite often, thus it might be convenient
;; to give it a key binding. For instance, if `gited.el' is in
;; your `load-path', then you can bind it to `C-x C-g' in Dired buffers
;; by adding the following lines into your .emacs file:
;;
;; (require 'gited)
;; (define-key dired-mode-map "\C-x\C-g" 'gited-list-branches)
;;
;; If you are familiar with Dired, then you already know how to use
;; Gited; that's because most of the Gited commands with a Dired equivalent
;; share same keybindings.
;; For instance `gited-rename-branch' is bound to `R' as `dired-do-rename'.
;; Similarly, `gited-mark' is bound to `m' as `dired-mark'.
;;
;;
;; Internal variables defined here:
;;
;; `gited--hide-details-set', `gited--op',
;; `gited--revert-commit', `gited--running-async-op',
;; `gited-actual-switches', `gited-after-change-hook',
;; `gited-author-face', `gited-author-idx',
;; `gited-bisect-buf-name', `gited-bisect-buffer',
;; `gited-bisect-buffer', `gited-bisect-output-name',
;; `gited-branch-after-op', `gited-branch-alist',
;; `gited-branch-idx', `gited-branch-name-face',
;; `gited-buffer', `gited-buffer-name',
;; `gited-commit-idx', `gited-commit-msg-face',
;; `gited-current-branch', `gited-date-idx',
;; `gited-date-regexp', `gited-date-time-face',
;; `gited-del-char', `gited-deletion-branch-face',
;; `gited-deletion-face', `gited-edit-commit-mode-map',
;; `gited-flag-mark-face', `gited-flag-mark-line-face',
;; `gited-header', `gited-list-format',
;; `gited-list-refs-format-command', `gited-log-buffer',
;; `gited-mark-face', `gited-mark-idx',
;; `gited-marker-char', `gited-mode',
;; `gited-mode-map', `gited-modified-branch',
;; `gited-new-or-deleted-files-re', `gited-op-string',
;; `gited-original-buffer', `gited-output-buffer',
;; `gited-output-buffer-name', `gited-re-mark',
;; `gited-ref-kind', `gited-section-highlight-face',
;; `gited-toplevel-dir'.
;;
;; Coustom variables defined here:
;;
;; `gited-current-branch-face', `gited-date-format',
;; `gited-expert', `gited-patch-options',
;; `gited-patch-program', `gited-protected-branches',
;; `gited-reset-mode', `gited-short-log-cmd',
;; `gited-switches', `gited-use-header-line',
;; `gited-verbose'.
;;
;; Macros defined here:
;;
;; `gited-map-over-marks', `gited-mark-if',
;; `gited-with-current-branch'.
;;
;; Commands defined here:
;;
;; `gited--mark-merged-branches-spec', `gited--mark-unmerged-branches-spec',
;; `gited-add-patched-files', `gited-apply-add-and-commit-patch',
;; `gited-apply-patch', `gited-async-operation',
;; `gited-bisect', `gited-branch-clear',
;; `gited-checkout-branch', `gited-commit',
;; `gited-copy-branch', `gited-copy-branchname-as-kill',
;; `gited-delete-branch', `gited-diff-with-branch',
;; `gited-do-delete', `gited-do-flagged-delete',
;; `gited-do-kill-lines', `gited-edit-commit-mode',
;; `gited-extract-patches', `gited-finish-commit-edit',
;; `gited-flag-branch-deletion', `gited-goto-branch',
;; `gited-goto-first-branch', `gited-goto-last-branch',
;; `gited-kill-line', `gited-list-branches',
;; `gited-log', `gited-log-last-n-commits',
;; `gited-mark', `gited-mark-branches-containing-commit',
;; `gited-mark-branches-containing-regexp', `gited-mark-branches-regexp',
;; `gited-mark-merged-branches', `gited-mark-unmerged-branches',
;; `gited-merge-branch', `gited-move-to-author',
;; `gited-move-to-branchname', `gited-move-to-date',
;; `gited-move-to-end-of-author', `gited-move-to-end-of-branchname',
;; `gited-move-to-end-of-date', `gited-next-line',
;; `gited-next-marked-branch', `gited-number-marked',
;; `gited-origin', `gited-prev-line',
;; `gited-prev-marked-branch', `gited-pull',
;; `gited-push', `gited-rename-branch',
;; `gited-reset-branch', `gited-revert-commit',
;; `gited-set-branch-upstream', `gited-show-commit',
;; `gited-stash', `gited-stash-apply',
;; `gited-stash-branch', `gited-stash-drop',
;; `gited-stash-pop', `gited-status',
;; `gited-summary', `gited-sync-with-trunk',
;; `gited-toggle-marks', `gited-unmark',
;; `gited-unmark-all-branches', `gited-unmark-all-marks',
;; `gited-unmark-backward', `gited-update',
;; `gited-visit-branch-sources', `gited-why'.
;;
;; Non-interactive functions defined here:
;;
;; `gited--bisect-after-run', `gited--bisect-executable-p',
;; `gited--case-ref-kind', `gited--check-unmerged-marked-branches',
;; `gited--clean-previous-patches', `gited--fill-branch-alist',
;; `gited--fontify-current-row', `gited--get-branches-from-command',
;; `gited--get-column', `gited--get-merged-branches',
;; `gited--get-patch-or-commit-buffers', `gited--get-unmerged-branches',
;; `gited--goto-column', `gited--goto-first-branch',
;; `gited--handle-new-or-delete-files', `gited--list-format-init',
;; `gited--mark-branches-in-region',
;; `gited--mark-merged-or-unmerged-branches',
;; `gited--mark-merged-or-unmerged-branches-spec', `gited--merged-branch-p',
;; `gited--move-to-end-of-column', `gited--output-buffer',
;; `gited--patch-or-commit-buffer', `gited--set-output-buffer-mode',
;; `gited--stash-branch', `gited--update-padding',
;; `gited--valid-ref-p', `gited-all-branches',
;; `gited-async-operation-sentinel', `gited-at-header-line-p',
;; `gited-bisecting-p', `gited-branch-exists-p',
;; `gited-buffer-p', `gited-commit-title',
;; `gited-current-branch', `gited-current-branches-with-marks',
;; `gited-current-state-list', `gited-date-string-to-time',
;; `gited-dir-under-Git-control-p', `gited-edit-commit',
;; `gited-fontify-current-branch', `gited-fontify-marked-branch-name',
;; `gited-format-columns-of-files', `gited-get-branchname',
;; `gited-get-commit', `gited-get-date',
;; `gited-get-element-in-row', `gited-get-last-commit-time',
;; `gited-get-mark', `gited-get-marked-branches',
;; `gited-git-command', `gited-git-command-on-region',
;; `gited-hide-details-update-invisibility-spec',
;; `gited-insert-marker-char', `gited-internal-do-deletions',
;; `gited-last-commit-title', `gited-listed-branches',
;; `gited-log-msg', `gited-log-summary',
;; `gited-map-lines', `gited-mark-pop-up',
;; `gited-mark-remembered', `gited-modified-files',
;; `gited-modified-files-p', `gited-next-branch',
;; `gited-number-of-commits', `gited-prev-branch',
;; `gited-print-entry', `gited-remember-marks',
;; `gited-remote-repository-p', `gited-repeat-over-lines',
;; `gited-stashes', `gited-tabulated-list-entries',
;; `gited-trunk-branches', `gited-untracked-files'.
;;
;; Faces defined here:
;;
;; `gited-author', `gited-branch-name',
;; `gited-commit-msg', `gited-date-time',
;; `gited-deletion', `gited-deletion-branch',
;; `gited-flag-mark', `gited-flag-mark-line',
;; `gited-header', `gited-mark',
;; `gited-modified-branch', `gited-section-highlight',
;; `gited-status-branch-local', `gited-status-tag'.
;;
;;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;
;; This file is NOT part of GNU Emacs.
;;
;; GNU Emacs is free software: you can redistribute it and/or modify
;; it under the terms of the GNU General Public License as published by
;; the Free Software Foundation, either version 3 of the License, or
;; (at your option) any later version.
;;
;; GNU Emacs is distributed in the hope that it will be useful,
;; but WITHOUT ANY WARRANTY; without even the implied warranty of
;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
;; GNU General Public License for more details.
;;
;; You should have received a copy of the GNU General Public License
;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
;;; Code:
(require 'cl-lib)
(require 'tabulated-list)
(require 'dired)
(require 'vc-git)
(require 'find-func) ; `find-library-name'
(defgroup gited nil
"Git branch editing."
:version "26.1"
:group 'vc)
(defvar gited-mode nil
"Variable saving the status of `gited-mode'.")
(make-variable-buffer-local 'gited-mode)
(put 'gited-mode 'permanent-local t)
(defvar gited-current-branch nil
"The branch currently checked out.")
(make-variable-buffer-local 'gited-current-branch)
(put 'gited-current-branch 'permanent-local t)
(defvar gited-toplevel-dir nil
"Absolute path of the top-level directory for the current repository.")
(make-variable-buffer-local 'gited-toplevel-dir)
(put 'gited-toplevel-dir 'permanent-local t)
;; Stolen from ediff-ptch.el
(defcustom gited-patch-program "patch"
"Name of the program that applies patches.
It is recommended to use GNU-compatible versions."
:type 'string
:group 'gited)
;; Stolen from ediff-ptch.el
(defcustom gited-patch-options "-f"
"Options to pass to `gited-patch-program'.
It is recommended to pass the `-f' option to the patch program, so it won't ask
questions. However, some implementations don't accept this option, in which
case the default value for this variable should be changed."
:type 'string
:group 'gited)
(defcustom gited-use-header-line t
"If non-nil, use the header line to display Gited column titles."
:type 'boolean
:group 'gited)
(defcustom gited-verbose nil
"If non-nil, show the author name and commit date."
:type 'boolean
:group 'gited)
(defvar gited--hide-details-set nil)
(make-variable-buffer-local 'gited--hide-details-set)
(defvar gited-ref-kind nil
"The kind of Git reference displayed in `gited-buffer'.
It must be `local', `remote' or `tags'.")
(make-variable-buffer-local 'gited-ref-kind)
(put 'gited-ref-kind 'permanent-local t)
(defvar gited-mark-idx 0
"Position of marker char in array `tabulated-list-entries'.")
(defvar gited-author-idx 1
"Position of author in array `tabulated-list-entries'.")
(defvar gited-date-idx 2
"Position of date in array `tabulated-list-entries'.")
(defvar gited-branch-idx 3
"Position of branch name in array `tabulated-list-entries'.")
(defvar gited-commit-idx 4
"Position of newest commit tittle in array `tabulated-list-entries'.")
(defvar gited-after-change-hook nil
"Hook run after make a change in the list of branches.")
(defvar gited-branch-after-op nil
"Branch where to set point after an asynchronous operation.")
(defvar gited-op-string nil
"String to describe the actual operation.")
(defvar gited-marker-char ?*
"In gited, the current mark character.
This is what the do-commands look for, and what the mark-commands store.")
(defvar gited-del-char ?D
"Character used to flag branches for deletion.")
;; "Regexp matching a marked line.
(defvar gited-re-mark "^[^ \n]")
(defvar gited-branch-alist nil
"Alist with elements (INDEX MARK TIME BRANCH-NAME AUTHOR-NAME TITLE).
INDEX identify each row in the table.
MARK is the mark character shown in the table for that row.
TIME is the time of the last commit in that branch.
BRANCH-NAME is the name of the branch.
AUTHOR-NAME is the author of the last commit in that branch.
TITLE is the title of the last commit.")
(make-variable-buffer-local 'gited-branch-alist)
(defvar gited-buffer-name "*gited*"
"Name of the buffer where to list the repository branches.")
(defvar gited-buffer nil
"Buffer where to list the repository branches.")
(make-variable-buffer-local 'gited-buffer)
(put 'gited-buffer 'permanent-local t)
(defvar gited-output-buffer-name "*gited-output*"
"Name of the buffer where to show output from Git commands.")
(defvar gited-output-buffer nil
"Output buffer for Git commands.")
(make-variable-buffer-local 'gited-output-buffer)
(put 'gited-output-buffer 'permanent-local t)
(defvar gited-bisect-buf-name "*gited-bisect*"
"Name of the buffer where to show bisect output.")
(defvar gited-bisect-buffer nil
"Output buffer for Git bisects.")
(make-variable-buffer-local 'gited-bisect-buffer)
(put 'gited-bisect-buffer 'permanent-local t)
(defvar gited-bisect-output-name "*gited-bisect*"
"Name of the output buffer for Git bisects.")
(defvar gited-bisect-buffer nil
"Output buffer for Git bisects.")
(make-variable-buffer-local 'gited-bisect-buffer)
(put 'gited-bisect-buffer 'permanent-local t)
(defvar gited-list-refs-format-command
'("for-each-ref" "--format='(%(authordate:raw) \
\"%(refname:short)\" \"%(authorname)\")'" "refs/%s")
"Format strings to build a Git command to list references.")
(defvar gited-date-regexp (concat "\\("
(substring
directory-listing-before-filename-regexp
(length "\\([0-9][BkKMGTPEZY]? ")))
"Regular expression to match a date in `gited-buffer' buffer.")
(defcustom gited-switches "-g"
"Control how to sort `gited-branch-alist'.
Option -r reverse order while sorting.
Option -g do not show the author name."
:type '(choice
(const :tag "Unset" nil) string)
:group 'gited)
(defvar gited-actual-switches gited-switches
"Switches used on this buffer.")
(make-variable-buffer-local 'gited-actual-switches)
(put 'gited-actual-switches 'permanent-local t)
(defvar gited-list-format nil
"Format of the columns in the branch list.")
(make-variable-buffer-local 'gited-list-format)
(defcustom gited-reset-mode "mixed"
"Default mode of a Git reset."
:type '(choice
(const :tag "soft" "soft")
(const :tag "mixed" "mixed")
(const :tag "hard" "hard")
(const :tag "merge" "merge")
(const :tag "keep" "keep"))
:group 'gited)
(defun gited--list-format-init (&optional col-names col-sizes)
"Initialize `gited-list-format'.
Optional arguments COL-NAMES and COL-SIZES are the column names
and sizes."
(setq gited-actual-switches gited-switches)
(setq gited-list-format
(vector `(,(if col-names (nth 0 col-names) "M")
,(if col-sizes (nth 0 col-sizes) 2) t)
`(,(if col-names (nth 1 col-names) "Authors")
,(if col-sizes (nth 1 col-sizes) 16) t)
`(,(if col-names (nth 2 col-names) "Date")
,(if col-sizes (nth 2 col-sizes) 17)
(lambda (row1 row2)
(let* ((reverse-order
(member "-r" (split-string gited-actual-switches)))
(t1 (aref (cadr row1) gited-date-idx))
(t2 (aref (cadr row2) gited-date-idx))
(earlierp
(time-less-p
(apply #'encode-time (parse-time-string t1))
(apply #'encode-time (parse-time-string t2)))))
(if reverse-order
earlierp
(not earlierp)))))
`(,(if col-names (nth 3 col-names) "Branches")
,(if col-sizes (nth 3 col-sizes) 50) t)
`(,(if col-names (nth 4 col-names) "Last Commit")
,(if col-sizes (nth 4 col-sizes) 65) t))))
(setq gited-list-format (gited--list-format-init))
(defcustom gited-expert nil
"If non-nil, don't ask for confirmation for some operations on branches."
:type 'boolean
:group 'gited)
(defcustom gited-date-format "%F %R"
"Format to display the date in `gited-buffer'."
:type 'string
:group 'gited)
(defcustom gited-current-branch-face 'font-lock-keyword-face
"Face used for displaying current checkout branch."
:type 'face
:group 'gited)
(defface gited-section-highlight
'((((class color) (background light)) :background "grey95")
(((class color) (background dark)) :background "grey20"))
"Face for highlighting the current branch."
:group 'gited)
(defvar gited-section-highlight-face 'gited-section-highlight)
(defface gited-flag-mark-line
'((((background dark)) (:background "#787831311414")) ; ~ dark red brown
(t (:background "Skyblue")))
"Face used for flagged and marked lines in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-flag-mark-line-face 'gited-flag-mark-line)
(defface gited-flag-mark
'((((background dark))
(:foreground "Blue" :background "#7575D4D41D1D")) ; ~ olive green
(t (:foreground "Yellow" :background "Blueviolet")))
"Face used for flags and marks (except D) in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-flag-mark-face 'gited-flag-mark)
(defface gited-deletion-branch
'((t (:foreground "Red")))
"Face used for branches flagged for deletion in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-deletion-branch-face 'gited-deletion-branch)
(defface gited-deletion
'((t (:foreground "Yellow" :background "Red")))
"Face used for deletion flags (D) in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-deletion-face 'gited-deletion)
(defface gited-header
'((t (:inherit font-lock-type-face)))
"Face used used for header when listing Git branches."
:group 'gited)
(defface gited-status-branch-local ; Same as magit-branch-local.
'((((class color) (background light)) :foreground "SkyBlue4")
(((class color) (background dark)) :foreground "LightSkyBlue1"))
"Face for local branches in status."
:group 'gited)
(defface gited-status-tag ; Same as gited-tag.
'((((class color) (background light)) :foreground "Goldenrod4")
(((class color) (background dark)) :foreground "LightGoldenrod2"))
"Face for tag labels shown in log buffer."
:group 'gited)
(defcustom gited-protected-branches nil
"Name of protected branches.
These branches cannot be deleted or renamed."
:type '(repeat (string :tag "Branch name"))
:group 'gited)
(defcustom gited-short-log-cmd
'("log" "--pretty=format:'%h %ad | %s%d [%an]'" "--graph" "--date=short")
"Default short format for Git log."
:type 'string
:group 'gited)
(defvar gited-header 'gited-header
"Face name used for header when listing Git branches.")
(defface gited-modified-branch
'((((background dark)) (:foreground "green1"))
(t (:foreground "red")))
"*Face used for branches with unstaged/uncommitted changes."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-modified-branch 'gited-modified-branch)
(defface gited-mark
'((t (:inherit font-lock-constant-face)))
"Face used for Gited marks."
:group 'gited-faces)
(defvar gited-mark-face 'gited-mark
"Face name used for Gited marks.")
(defface gited-author
'((((background dark)) (:foreground "orange"))
(t (:foreground "black")))
"*Face used for AUTHOR in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-author-face 'gited-author)
(defface gited-date-time
'((((background dark)) (:foreground "#74749A9AF7F7")) ; ~ med blue
(t (:foreground "DarkGoldenrod4")))
"Face used for date and time in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-date-time-face 'gited-date-time)
(defface gited-branch-name
'((((background dark)) (:foreground "Yellow"))
(t (:foreground "Blue")))
"Face used for branch names in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-branch-name-face 'gited-branch-name)
(defface gited-commit-msg ; Same as magit-branch-remote.
'((((class color) (background light)) :foreground "DarkOliveGreen4")
(((class color) (background dark)) :foreground "DarkSeaGreen2"))
"Face used for commit-msg in Gited buffers."
:group 'gited :group 'font-lock-highlighting-faces)
(defvar gited-commit-msg-face 'gited-commit-msg)
;;; Macros.
(defmacro gited-with-current-branch (branch &rest body)
"Set BRANCH temporarily current and execute forms in BODY.
BRANCH must be the name of an existing branch.
The value returned is the value of the last form in BODY."
(declare (indent 1) (debug t))
(let ((cur-branch (make-symbol "cur-branch")))
`(let ((,cur-branch gited-current-branch))
(unwind-protect
(progn
(vc-git-checkout nil ,branch)
(setq gited-current-branch ,branch)
,@body)
;; Restore original current branch.
(vc-git-checkout nil ,cur-branch)
(setq gited-current-branch ,cur-branch)))))
;;; Map over marks.
(defmacro gited-map-over-marks (body arg)
"Eval BODY with point on each marked line. Return a list of BODY's results.
If no marked branch could be found, execute BODY on the current
line. ARG, if non-nil, specifies the branches to use instead of the
marked branches.
If ARG is an integer, use the next ARG (or previous -ARG, if
ARG<0) files. In that case, point is dragged along. This is so
that commands on the next ARG (instead of the marked) files can
be chained easily.
For any other non-nil value of ARG, use the current file.
Search starts at the beginning of the buffer, thus the car of the
list corresponds to the line nearest to the buffer's bottom.
This is also true for (positive and negative) integer values of
ARG.
BODY should not be too long as it is expanded four times.
If DISTINGUISH-ONE-MARKED is non-nil, then if we find just one
marked file, return (t FILENAME) instead of (FILENAME)."
`(prog1
(let ((inhibit-read-only t) case-fold-search found results)
(if ,arg
(if (integerp ,arg)
(progn ;; no save-excursion, want to move point.
(gited-repeat-over-lines
,arg
(lambda ()
(setq results (cons ,body results))))
(if (< ,arg 0)
(nreverse results)
results))
;; non-nil, non-integer ARG means use current file:
(list ,body))
(let ((regexp
(concat "^"
(regexp-quote
(char-to-string gited-marker-char))))
next-position)
(save-excursion
(goto-char (point-min))
;; remember position of next marked file before BODY
;; can insert lines before the just found file,
;; confusing us by finding the same marked file again
;; and again and...
(setq next-position (and (re-search-forward regexp nil t)
(point-marker))
found (not (null next-position)))
(while next-position
(goto-char next-position)
(setq results (cons ,body results))
;; move after last match
(goto-char next-position)
(forward-line)
(set-marker next-position nil)
(setq next-position (and (re-search-forward regexp nil t)
(point-marker)))))
(if found
results
(list ,body)))))
;; save-excursion loses, again
(gited-move-to-branchname)))
(defmacro gited-mark-if (predicate msg)
"Mark all branches for which PREDICATE eval to non-nil.
PREDICATE is evaluated on each line, with point at beginning of line.
MSG is a noun phrase for the type of branches being marked.
It should end with a noun that can be pluralized by adding `s'.
Return value is the number of files marked, or nil if none were marked."
`(let ((inhibit-read-only t) count)
(save-excursion
(setq count 0)
(when ,msg
(message "%s %ss%s..."
(cond ((eq gited-marker-char ?\s) "Unmarking")
((eq gited-marker-char gited-del-char)
"Flagging")
(t "Marking"))
,msg
(if (eq gited-marker-char gited-del-char)
" for deletion"
"")))
(gited--goto-first-branch)
(while (not (eobp))
(if ,predicate
(progn
(gited-insert-marker-char)
(gited-fontify-marked-branch-name)
(cl-incf count)))
(forward-line))
(if ,msg (message "%s %s%s %s%s."
count
,msg
(gited-plural-s count)
(if (eq gited-marker-char ?\s) "un" "")
(if (eq gited-marker-char gited-del-char)
"flagged" "marked"))))
(and (> count 0) count)))
;;; Convenience functions.
;; Following functions are stolen from Dired.
;; To avoid code duplication we just bind `dired-log-buffer' to
;; `gited-log-buffer' and use the original definitions in dired.el.
(defvar gited-log-buffer "*Gited log*")
(defalias 'gited-plural-s 'dired-plural-s)
(defun gited-why ()
"Pop up a buffer with error log output from Gited.
A group of errors from a single command ends with a formfeed.
Thus, use \\[backward-page] to find the beginning of a group of errors."
(interactive)
(let ((dired-log-buffer gited-log-buffer))
(dired-why)))
(defun gited-summary ()
"Summarize basic Gited commands and show recent Gited errors."
(interactive)
(gited-why)
(message
"d-elete, u-ndelete, x-punge, f-ind, o-rigin, R-ename, C-opy, h-elp"))
(defun gited-log-msg (log &rest args)
(let ((dired-log-buffer gited-log-buffer))
(dired-log log args)))
(defun gited-log-summary (string failures)
(let ((dired-log-buffer gited-log-buffer))
(dired-log-summary string failures)))
(defun gited-format-columns-of-files (branches)
(let ((beg (point)))
(completion--insert-strings branches)
(put-text-property beg (point) 'mouse-face nil)))
(defun gited-next-branch ()
"Return name of next branch."
(save-excursion
(forward-line)
(ignore-errors
(gited-get-branchname))))
(defun gited-prev-branch ()
"Return name of previous branch."
(save-excursion
(forward-line -1)
(ignore-errors
(gited-get-branchname))))
(defun gited-current-branch ()
"Return name of current branch."
(car (vc-git-branches)))
(defun gited-get-last-commit-time (branch)
"Return last commit time of BRANCH."
(apply #'encode-time
(parse-time-string
(cadr
(cdr
(nth
(cl-position-if
(lambda (x) (equal branch (nth 3 x))) gited-branch-alist)
gited-branch-alist))))))
(defun gited--get-column (col)
(mapcar (lambda (x)
(elt (cadr x) col))
tabulated-list-entries))
(defun gited-listed-branches ()
"Return a list with all listed branches/tags names."
(gited--get-column gited-branch-idx))
(defalias 'gited-get-branches 'gited-listed-branches)
(defun gited-git-command (args &optional buffer display unquote)
"Execute a Git command with arguments ARGS.
Optional arg BUFFER is the output buffer.
Optional arg DISPLAY means redisplay buffer as output is inserted.
Optional arg UNQUOTE removes single quotes from the output."
(prog1
(apply #'call-process vc-git-program nil buffer display args)
(when (and unquote buffer (buffer-live-p buffer))
(with-current-buffer buffer
;; Drop single quotes.
(save-excursion
(goto-char (point-min))
(while (re-search-forward "^'\\(.*\\)'$" nil t)
(replace-match "\\1")))))))
(defun gited-git-command-on-region (args &optional buffer display)
"Execute a Git command with arguments ARGS and region as input.
Optional arg BUFFER is the output buffer.
Optional arg DISPLAY means redisplay buffer as output is inserted."
(apply #'call-process-region nil nil vc-git-program nil buffer display args))
(defun gited-all-branches ()
"Return a list with all (local and remote) branches and tags."
(let ((args '("for-each-ref" "--format='%(refname:short)'" "refs/heads"
"refs/remotes" "refs/tags")))
(with-temp-buffer
(gited-git-command args (current-buffer) nil 'unquote)
(split-string (buffer-string) "\n" 'omit-nulls))))
(defun gited-copy-branchname-as-kill ()
"Copy names of marked branches into the kill ring.
The names are separated by a space.
If on a subdir headerline, use absolute subdirname instead;
prefix arg and marked files are ignored in this case.
You can then feed the file name(s) to other commands with \\[yank]."
(interactive)
(let ((string
(mapconcat (function identity)
(or (gited-get-marked-branches)
(list (gited-get-branchname)))
" ")))
(unless (string= string "")
(if (eq last-command 'kill-region)
(kill-append string nil)
(kill-new string))
(message "%s" string))))
(defun gited--output-buffer (&optional buf-name)
(unless buf-name
(setq buf-name gited-output-buffer-name))
(let* ((buf
(cond ((equal buf-name gited-bisect-buf-name)
(and (buffer-live-p gited-bisect-buffer) gited-bisect-buffer))
(t
(and (buffer-live-p gited-output-buffer) gited-output-buffer))))
(res (or (and (buffer-live-p buf)
(equal default-directory
(buffer-local-value 'default-directory buf))
buf)
(generate-new-buffer buf-name))))
(if (equal buf-name gited-bisect-buf-name)
(setq gited-bisect-buffer res)
(setq gited-output-buffer res))))
;;; Predicates.
(defun gited-modified-files ()
"Return a list with all unstaged files."
(let ((regexp "^[[:blank:]]*[MAU]+[[:blank:]]*\\(.+\\)")
(case-fold-search)
res)
(with-temp-buffer
(gited-git-command '("status" "--porcelain") (current-buffer))
(goto-char (point-min))
(while (re-search-forward regexp nil t)
(push (match-string 1) res)))
(nreverse res)))
(defun gited-modified-files-p ()
"Return non-nil if there are unstaged changes."
(and (gited-modified-files) t))
(defun gited-dir-under-Git-control-p ()
"Return non-nil if current directory is under Git version control."
(zerop (gited-git-command '("status"))))
(defun gited-branch-exists-p (branch)
"Return non-nil if BRANCH exists."
(member branch (gited-listed-branches)))
(defun gited-buffer-p ()
"Return non-nil if current buffer is a gited buffer."
(string-prefix-p gited-buffer-name
(buffer-name (current-buffer))))
(defun gited-at-header-line-p ()
"Return non-nil if point is at header line."
(and (not gited-use-header-line)
(= 1 (line-number-at-pos (point)))))
(defun gited-remote-repository-p ()
"Return non-nil if current repository is remote."
(let ((regexp "^remote.origin"))
(with-temp-buffer
(gited-git-command '("config" "--local" "--list") (current-buffer))
(goto-char 1)
(and (re-search-forward regexp nil t) t))))
;;; Operations on branches (copy, merge, ...).
(defun gited--get-branches-from-command (cmd)
(with-temp-buffer
(gited-git-command cmd (current-buffer) nil 'unquote)
(goto-char (point-min))
(while (re-search-forward "^\\( \\|\\* \\)" nil t)
(replace-match ""))
(split-string (buffer-string) "\n" 'omit-nulls)))
(defun gited-trunk-branches ()
"Return a list with branch names tracked from a remote repository."
(let ((regexp "^branch\.\\([^.]+\\)\.merge=")
res)
(with-temp-buffer
(gited-git-command '("config" "--local" "--list") (current-buffer))
(goto-char 1)
(while (re-search-forward regexp nil t)
(push (match-string 1) res))
(nreverse res))))
(defun gited--get-unmerged-branches ()
(let ((args `("branch" "--no-merged" ,(car (gited-trunk-branches)))))
(gited--get-branches-from-command args)))
(defun gited--get-merged-branches ()
(let ((args `("branch" "--merged" ,(car (gited-trunk-branches)))))
(gited--get-branches-from-command args)))
(defun gited--check-unmerged-marked-branches (&optional marker)
(let ((marked-branches (or (gited-get-marked-branches marker)
(list (gited-get-branchname)))))
(dolist (b marked-branches)
(let ((unmerged (ignore-errors (gited--get-unmerged-branches))))
(dolist (x unmerged)
(when (string= b x)
(error "Cannot delete unmerged branches. Try C-u %s"
(substitute-command-keys (this-command-keys)))))))))
(defun gited--merged-branch-p (branch)
(and (member branch (gited--get-merged-branches))
t))
(defun gited-untracked-files ()
"Return a list with all untracked files."
(let ((args '("status" "--porcelain"))
(regexp "^[?]\\{2\\}[[:blank:]]+\\(.+\\)")
res)
(with-temp-buffer
(gited-git-command args (current-buffer))
(goto-char (point-min))
(while (re-search-forward regexp nil t)
(push (match-string 1) res)))
(nreverse res)))
(defun gited-stashes ()
"Return a list with all the stashes."
(let ((args '("stash" "list"))
res)
(with-temp-buffer
(gited-git-command args (current-buffer))
(goto-char (point-min))
(while (not (eobp))
(push (buffer-substring (point-at-bol) (point-at-eol)) res)
(forward-line)))
(nreverse res)))
(defun gited-commit-title (commit)
"Return title of COMMIT, a string."
(let ((args `("log" "--pretty=format:'%s'" "-n1" ,commit)))
(with-temp-buffer
(gited-git-command args (current-buffer) nil 'unquote)
(buffer-string))))
(defun gited-last-commit-title ()
"Return title of the last commit."
(gited-commit-title "HEAD"))
;; Non-nil while running an asynchronous Gited subprocess.
(defvar gited--running-async-op nil)
(make-variable-buffer-local 'gited--running-async-op)
(defun gited-async-operation (command &optional remote-op-p buffer)
"Run COMMAND asynchronously.
COMMAND perform a branch operation, i.e., rename or delete a branch.
Optional arg REMOTE-OP-P, means the operation modify the remote
repository. Otherwise, the operation just change local branches.
Optional arg BUFFER is the output buffer for the operation. Otherwise,
use `gited-output-buffer'."
(interactive)
(if gited--running-async-op
(error "Cannot run 2 Gited async process in parallel")
(let* ((gited-buf (current-buffer))
(out-buf (or buffer (gited--output-buffer)))
(directory default-directory)
proc)
(with-current-buffer out-buf
;; Always display out-buf for remote operations. This is to prompt
;; for Git user/password.
(when remote-op-p
(display-buffer out-buf '(nil (allow-no-window . t))))
(setq default-directory directory
proc (start-process-shell-command
"*gited-async-operation*" out-buf command)
mode-line-process '(":%s"))
(with-current-buffer gited-buf
(setq gited--running-async-op t))
(with-no-warnings
(require 'shell) (shell-mode))
(set-process-sentinel proc 'gited-async-operation-sentinel)
(set-process-filter proc 'comint-output-filter)
;; Associate out-buf with gited-buf; this is used in the sentinel.
(setq gited-buffer gited-buf)))))
(defun gited-async-operation-sentinel (proc state)
"Sentinel for asynchronous operations on branches.
PROC is the process.
STATE is the state of process PROC."
(let* ((buf (process-buffer proc))
(gited-buf (and (buffer-live-p buf)
(buffer-local-value 'gited-buffer buf)))
(op-string gited-op-string)
(inhibit-read-only t))
(ignore state)
(when (memq (process-status proc) '(exit signal))
(with-current-buffer gited-buf
(setq gited--running-async-op nil)
(when (gited-bisecting-p)
(gited--bisect-after-run (process-buffer proc)))))
(when (buffer-live-p gited-buf)
(set-buffer gited-buf)
(run-hooks 'gited-after-change-hook)
(when gited-branch-after-op
(gited-goto-branch gited-branch-after-op)
(setq gited-branch-after-op nil
gited-op-string nil))
(message "%s done!" op-string))))
(define-minor-mode gited-hide-details-mode
"Toggle visibility of detailed information in current Gited buffer.
When this minor mode is enabled, details such as last commit author and
date are hidden from view."
:group 'gited
(unless (derived-mode-p 'gited-mode)
(error "Not a Gited buffer"))
(gited-hide-details-update-invisibility-spec))
(put 'gited-hide-details-mode 'permanent-local t)
(defun gited-update ()
"Update `gited-branch-alist' and redisplay the list of branches."
(interactive)
(unless (derived-mode-p major-mode 'gited-mode)
(error "Cannot enable Gited mode in this buffer"))
(let ((target-br (ignore-errors (gited-get-branchname)))
(at-headr-p (gited-at-header-line-p))
(hide-details gited-hide-details-mode))
(gited-list-branches gited-ref-kind nil 'update)
(gited-hide-details-mode (if hide-details 1 0))
(if (not at-headr-p)
(gited-goto-branch target-br)
(gited-goto-branch gited-current-branch))))
(add-hook 'gited-after-change-hook 'gited-update)
(defun gited-rename-branch (old-name new-name)
"Rename branch OLD-NAME to NEW-NAME."
(interactive
(let* ((old (gited-get-branchname))
(new (read-string
(format "Rename %s to: " old)
nil nil old)))
(list old new)))
(when (member old-name gited-protected-branches)
(error "Cannot rename a protected branch"))
(let ((buf (gited--output-buffer))
(inhibit-read-only t) remote-op-p)
(setq gited-output-buffer buf)
(with-current-buffer buf (erase-buffer))
(if (and (not gited-expert)
(not (y-or-n-p
(format "Rename branch '%s' to '%s'? "
old-name new-name))))
(message "OK, rename canceled")
(if (gited-async-operation
(pcase gited-ref-kind
("remote"
(let ((old (substring old-name (length "origin/")))
(new (if (string-prefix-p "origin/" new-name)
(substring new-name (length "origin/"))
new-name)))
(setq remote-op-p t)
(format "%s push origin origin/%s:refs/heads/%s :%s"
vc-git-program old new old)))
("local" (format "%s branch --move %s %s"
vc-git-program old-name new-name))
("tags" (error "Rename tags not implemented!"))
(_ (error "Unsupported gited-ref-kind: must be \
local, remote or tags")))
remote-op-p)
(progn
(setq gited-branch-after-op new-name
gited-op-string
(format "Rename branch '%s' to '%s'" old-name new-name)))
(error "Cannot rename branch '%s' to '%s'" old-name new-name)))))
(defun gited-merge-branch (branch)
"Merge BRANCH with another one.
That means, changes from another branch are added into BRANCH."
(interactive
(list (gited-get-branchname)))
(let ((branch-new
(completing-read
(format "Merge %s with: " branch)
(gited-listed-branches)
nil t)))
(gited-with-current-branch branch
(if (zerop (gited-git-command `("merge" ,branch-new)))
(message "Merged %s into %s!" branch-new branch)
(error "Cannot merge '%s' into '%s'" branch-new branch)))))
(defun gited-reset-branch (commit &optional mode)
"Reset current branch to an earlier state.
COMMIT is a SHA1 string or HEAD~N, to reset BRANCH to that commit.
Interactively prompt for the limit commit: 0 means HEAD,
1 means HEAD~, and so on.
Interactively with a prefix argument prompts for the reset mode.
Defaults to `gited-reset-mode'."
(interactive
(let* ((alist
'(("s" . "soft") ("m" . "mixed") ("h" . "hard")
("g" . "merged") ("k" . "keep")))
(mode
(if current-prefix-arg
(cdr (assoc
(completing-read
"Reset mode (s = soft, m = mixed, h = hard, \
g = merged, k = keep): "
'("s" "m" "h" "g" "k")
nil 'mustmatch nil nil "h")
alist))
gited-reset-mode))
(input
(read-string
(format "Reset --%s to commit (0 = HEAD, \
1 = HEAD~1, ... or SHA1): " mode)
nil nil "0")))
(list (if (gited--valid-ref-p input)
input
(concat "HEAD~" input))
mode)))
(unless mode (setq mode gited-reset-mode))
(let ((branch (gited-current-branch))
(args `("reset" ,(concat "--" mode) ,commit)))
(if (not (y-or-n-p
(format "Reset --%s '%s' to '%s'? " mode branch commit)))
(message "OK, reset canceled")
(if (zerop (gited-git-command args))
(message "Reseted --%s '%s' to '%s'!" mode branch commit)
(error "Cannot reset --%s '%s' to '%s'" mode branch commit)))))
(defun gited-delete-branch (branch &optional force)
"Delete branch BRANCH.
BRANCH default to the branch at current line.
Optional arg FORCE, if non-nil then delete non-fully merged branches
as well."
(interactive
(list (gited-get-branchname) current-prefix-arg))
(let ((br-after (or (gited-next-branch)
(gited-prev-branch)
(gited-current-branch)))
(buf (gited--output-buffer))
(inhibit-read-only t))
(setq gited-output-buffer buf)
(with-current-buffer buf (erase-buffer))
(when (string= branch gited-current-branch)
(error "Cannot delete the current branch"))
(when (member branch gited-protected-branches)
(error "Cannot delete a protected branch"))
(if (and (not gited-expert)
(not (y-or-n-p (format "Delete branch '%s'? " branch))))
(message "OK, deletion canceled")
(pcase gited-ref-kind
("tags" (error "Delete tags not implemented!"))
("local"
(if (zerop (gited-git-command
;; --delete --force as shortcut of -D doesn't exist
;; in old Git versions.
`("branch"
,(if force "-D" "--delete")
,branch)
buf))
(progn
(gited-goto-branch br-after)
(message "Delete branch '%s'!" branch))
(error "Cannot delete unmerged branch '%s'. Try C-u %s"
branch
(substitute-command-keys "\\[gited-do-flagged-delete\]"))))
("remote"
(gited-async-operation
(format "%s push origin :%s"
vc-git-program
(substring branch (length "origin/"))) 'remote-op-p)
(setq gited-branch-after-op br-after
gited-op-string (format "Delete branch '%s'" branch)))
(_ (error "Unsupported gited-ref-kind: must be \
local, remote or tags"))))))
(defun gited-do-delete (&optional arg force)
"Delete all marked (or next ARG) branches.
Optional arg FORCE, if non-nil then delete non-fully merged branches
as well."
(interactive
(let ((prefix current-prefix-arg))
(list prefix (equal prefix '(4)))))
(unless force
(gited--check-unmerged-marked-branches gited-del-char))
(gited-internal-do-deletions
;; this may move point if ARG is an integer
(gited-map-over-marks (cons (or (gited-get-branchname)
(error "No branch on this line"))
(point))
arg)
arg force)
(run-hooks 'gited-after-change-hook))
(defun gited-mark-pop-up (buffer-or-name op-symbol branches function &rest args)
"Return FUNCTION's result on ARGS after showing which branches are marked.
Displays the branch names in a window showing a buffer named
BUFFER-OR-NAME; the default name being \" *Marked Branches*\". The
window is not shown if there is just one branch, `gited-no-confirm'
is t, or OP-SYMBOL is a member of the list in `gited-no-confirm'.
By default, Gited shrinks the display buffer to fit the marked branches.
To disable this, use the Customization interface to add a new rule
to `display-buffer-alist' where condition regexp is
\"^ \\*Marked Branches\\*$\",
action argument symbol is `window-height' and its value is nil.
BRANCHES is the list of marked branches. It can also be (t BRANCHNAME)
in the case of one marked branch, to distinguish that from using
just the current branch.
FUNCTION should not manipulate branches, just read input (an
argument or confirmation)."
(if (= (length branches) 1)
(apply function args)
(let ((buffer (get-buffer-create (or buffer-or-name " *Marked Branches*")))
;; Mark *Marked Branches* window as softly-dedicated, to prevent
;; other buffers e.g. *Completions* from reusing it (bug#17554).
(display-buffer-mark-dedicated 'soft))
(ignore op-symbol) ; ignore unused symbol.
(with-current-buffer buffer
(with-current-buffer-window
buffer
(cons 'display-buffer-below-selected
'((window-height . fit-window-to-buffer)))
#'(lambda (window _value)
(with-selected-window window
(unwind-protect
(apply function args)
(when (window-live-p window)
(quit-restore-window window 'kill)))))
(gited-format-columns-of-files
branches)
(remove-text-properties (point-min) (point-max)
'(mouse-face nil help-echo nil)))))))
(defun gited-internal-do-deletions (l arg &optional force)
;; L is an alist of branches to delete, with their buffer positions.
;; ARG is the prefix arg.
;; FORCE, if non-nil then delete non-fully merged branches as well.
;; (car L) *must* be the *last* (bottommost) branch in the gited buffer.
;; That way as changes are made in the buffer they do not shift the
;; lines still to be changed, so the (point) values in L stay valid.
;; Also, for subdirs in natural order, a subdir's branches are deleted
;; before the subdir itself - the other way around would not work.
(let* ((branches (mapcar (function car) l))
(count (length l))
(succ 0))
;; canonicalize branch list for pop up
(if (gited-mark-pop-up
" *Deletions*" 'delete branches 'y-or-n-p
(format "%s %s "
"Delete"
(replace-regexp-in-string "files"
"branches"
(dired-mark-prompt arg branches))))
(save-excursion
(let ((progress-reporter
(make-progress-reporter
"Deleting..."
succ count))
failures) ;; branches better be in reverse order for this loop!
(while l
(goto-char (cdar l))
(let ((inhibit-read-only t))
(condition-case err
(let ((fn (caar l))
(gited-expert t)) ;; Don't ask confirmation again.
(gited-delete-branch fn force)
;; if we get here, removing worked
(cl-incf succ)
(progress-reporter-update progress-reporter succ))
(error ;; catch errors from failed deletions
(gited-log-msg "%s\n" err)
(setq failures (cons (caar l) failures)))))
(setq l (cdr l)))
(if (not failures)
(progress-reporter-done progress-reporter)
(gited-log-summary
(format "%d of %d deletion%s failed"
(length failures) count
(gited-plural-s count))
failures))))
(message "(No deletions performed)")))
(gited-goto-branch gited-current-branch))
(defun gited-do-flagged-delete (&optional force)
"In Gited, delete the branches flagged for deletion.
Optional arg FORCE, if non-nil then delete non-fully merged branches
as well."
(interactive "P")
(let* ((gited-marker-char gited-del-char)
(regexp (concat "^" (regexp-quote (char-to-string gited-marker-char))))
case-fold-search)
(unless force
(gited--check-unmerged-marked-branches gited-del-char))
(if (save-excursion (goto-char (point-min))
(re-search-forward regexp nil t))
(progn
(gited-internal-do-deletions
;; this can't move point since ARG is nil
(gited-map-over-marks (cons (gited-get-branchname) (point))
nil)
nil force)
(run-hooks 'gited-after-change-hook))
(message "(No deletions requested)"))))
(defun gited-copy-branch (old new &optional commit)
"Copy branch OLD to branch NEW.
OLD default to branch at current line.
Optional arg COMMIT, if non-nil then is a SHA1 string or
HEAD~N, to copy OLD until that commit, inclusive.
When called with a prefix, prompt for the limit commit: 0 means HEAD,
1 means HEAD~, and so on."
(interactive
(let* ((old-br (gited-get-branchname))
(new-br (read-string
(format "Copy %s to: " old-br)
nil nil old-br))
(ask current-prefix-arg)
(num-or-sha1
(if ask
(read-string "Show commit (0 = HEAD, 1 = HEAD~1, ... or SHA1): "
nil nil "0")
"HEAD")))
(list old-br new-br num-or-sha1)))
(unless commit (setq commit "HEAD"))
(unless (gited--valid-ref-p commit)
(setq commit (concat old "~" commit)))
(let ((cmd1 `("checkout" ,old))
(cmd2 `("checkout" "-b" ,new ,commit))
(cmd3 `("checkout" ,gited-current-branch)))
(if (and (zerop (gited-git-command cmd1))
(zerop (gited-git-command cmd2))
(zerop (gited-git-command cmd3)))
(progn
(message "Copied %s to %s!" old new)
(run-hooks 'gited-after-change-hook)
(gited-goto-branch new))
(error "Cannot copy '%s' to '%s'" old new))))
;;; Checkout/visit sources.
(defun gited-visit-branch-sources (&optional other-window)
"Visit source code for current revision.
If optional arg OTHER-WINDOW is non-nil, then use another window."
(interactive "P")
(when (and (gited-modified-files-p)
(not (equal gited-current-branch (gited-get-branchname))))
(error "Cannot checkout a new branch: there are modified files"))
(let* ((branch (gited-get-branchname))
(visit-sources
(y-or-n-p (format "Visit '%s' branch sources? " branch))))
(if (not visit-sources)
(error "OK, canceled")
(let ((gited-expert visit-sources))
(gited-checkout-branch branch)
(if other-window
(dired-other-window default-directory)
(dired default-directory))
(dired-revert)))))
(defun gited--fontify-current-row ()
(remove-text-properties
(point-at-bol) (point-at-eol) '(face))
(let ((inhibit-read-only t) pos)
(save-excursion
(gited-move-to-author)
(setq pos (point))
(gited-move-to-end-of-author)
(put-text-property
pos (point) 'face gited-author-face)
(gited-move-to-date)
(setq pos (point))
(gited-move-to-end-of-date)
(put-text-property
pos (point) 'face gited-date-time-face)
(gited-move-to-branchname)
(setq pos (point))
(gited-move-to-end-of-branchname)
(put-text-property
pos (point) 'face gited-branch-name-face)
(setq pos (point))
(put-text-property
pos (point-at-eol) 'face gited-commit-msg-face))))
(defun gited-checkout-branch (branch)
"Checkout BRANCH.
If the gited buffer lists local branches and BRANCH is not
local, then prompt for a branch name where to check out BRANCH."
(interactive
(list (completing-read "Checkout branch: "
(gited-all-branches)
nil 'mustmatch (gited-get-branchname) nil)))
(when (and (gited-modified-files-p)
(not (equal gited-current-branch (gited-get-branchname))))
(error "Cannot checkout a new branch: there are modified files"))
(let* ((cur-br gited-current-branch)
(new-branch-p (and (equal gited-ref-kind "local")
(not (member branch (gited-get-branches)))))
(inhibit-read-only t)
(local-branch (if new-branch-p
(read-string
"Checkout in local branch: "
nil nil (file-name-nondirectory branch))
branch)))
(save-excursion
(gited-goto-branch cur-br)
(gited--fontify-current-row)
(if (not new-branch-p)
(vc-git-checkout nil branch)
(gited-git-command `("checkout" "-b" ,local-branch ,branch))
(run-hooks 'gited-after-change-hook))
(setq gited-current-branch local-branch))
(gited-fontify-current-branch)
(when new-branch-p
(gited-goto-branch gited-current-branch)))
(message "Current branch is %s" gited-current-branch))
;;; Apply patches.
(defun gited--patch-or-commit-buffer (&optional commit)
(let ((regexp
(if commit
"\\`\\*gited-commit-\\([0-9]+\\)\\*\\'"
"\\`\\*gited-patch-\\([0-9]+\\)\\*\\'")))
(get-buffer
(completing-read
(if commit
"Use commit message in buffer: "
"Apply patch from buffer: ")
(cl-delete-if-not
(lambda (x) (string-match regexp x))
(mapcar #'buffer-name (buffer-list)))))))
(defconst gited-new-or-deleted-files-re
(format "^\\(%s\\|%s\\|%s\\|%s\\|%s\\|%s\\)"
"deleted file mode"
"new file mode"
"copy from"
"copy to"
"rename from"
"rename to")
"Regexp to match new or deleted files in a Git diff.")
;; Currently just handle new files.
(defun gited--handle-new-or-delete-files (patch-buf)
(let ((new-files))
(goto-char 1)
(with-current-buffer patch-buf
(while (re-search-forward gited-new-or-deleted-files-re nil t)
(unless (string= "new file mode" (match-string-no-properties 0))
(error "Only creation of files is implementing: %s"
(match-string-no-properties 0)))
(let* ((str (buffer-substring-no-properties
(point-at-bol 0) (point-at-eol 0)))
(file
(progn
(string-match "diff --git a/\\(.*\\) b/.*" str)
(match-string-no-properties 1 str))))
(push file new-files))))
(if (zerop (gited-git-command (nconc '("add") new-files)))
(message "Sucessfully staged new files: %s"
(mapconcat #'shell-quote-argument new-files " "))
(error "Cannot stage some new files. Please check"))))
(defun gited-apply-patch (buf-patch &optional update)
"Apply patch at BUF-PATCH into current branch.
If optional arg UPDATE is non-nil, then call `gited-update'
after checkout."
(interactive
(list (gited--patch-or-commit-buffer)
current-prefix-arg))
(let ((toplevel gited-toplevel-dir)
create-or-del-files-p)
(with-temp-buffer
;; Apply patches from top-level dir.
(setq default-directory (file-name-as-directory toplevel))
(insert-buffer-substring-no-properties buf-patch)
(goto-char 1)
(setq create-or-del-files-p
(re-search-forward gited-new-or-deleted-files-re nil t))
(if (not (zerop (gited-git-command-on-region '("apply" "--check"))))
(error "Cannot apply patch at '%s'. Please check"
(buffer-name buf-patch))
(gited-git-command-on-region '("apply"))
(when create-or-del-files-p
(gited--handle-new-or-delete-files (current-buffer)))
(and update (gited-update))
(message "Patch applied successfully!")))))
(defun gited-add-patched-files (files &optional ask)
"Stage FILES for next commit.
If ASK is non-nil, then prompt the user before to add every hunk
and display the output buffer in other window."
(interactive
(list (gited-modified-files) current-prefix-arg))
(unless files (error "No modified files"))
(let ((buf (gited--output-buffer)))
(cond (ask
;; Output buffer must be editable.
(with-current-buffer buf
(setq buffer-read-only nil)
(erase-buffer))
(gited-async-operation (format "%s add --patch" vc-git-program)
nil buf)
(setq gited-op-string "add --patch")
(display-buffer buf))
(t
(let ((toplevel gited-toplevel-dir))
(with-temp-buffer
;; Add files from top-level dir.
(setq default-directory (file-name-as-directory toplevel))
(if (not (zerop (gited-git-command (nconc '("add") files))))
(error "Cannot add files. Please check")
(message "Successfully added files: %s"
(mapconcat #'shell-quote-argument files " ")))))))))
(defun gited-commit (comment &optional author)
"Commit latest changes using COMMENT as the message.
Optional argument AUTHOR is the author of the commit.
A prefix argument prompts for AUTHOR."
(interactive
(let ((_files (or (gited-modified-files) (error "No changes to commit")))
(name (and current-prefix-arg (read-string "Author: ")))
(msg (read-string "Message: ")))
(list msg name)))
(or (gited-modified-files) (error "No changes to commit"))
(let* ((buf (generate-new-buffer "*git-commit*"))
(args
(delete ""
`("commit" ,(if author (format "--author=\"%s\"" author) "")
"-m"
,comment))))
(unless (zerop (gited-git-command args buf))
(display-buffer buf)
(error "Commit failt: please check %s" (buffer-name buf)))
(with-current-buffer buf
(insert (format "\nCommit successfully with message:\n\n\"%s\"" comment)))
(gited--set-output-buffer-mode buf)
(display-buffer buf)))
(defun gited-apply-add-and-commit-patch (buf-patch buf-commit)
"Apply patch at BUF-PATCH, stage it and commit it with message BUF-COMMIT."
(interactive
(list
(gited--patch-or-commit-buffer)
(gited--patch-or-commit-buffer 'commit)))
(if (null (ignore-errors (gited-apply-patch buf-patch)))
(error "Cannot apply patch at %s" (buffer-name buf-patch))
(let ((cur-buf (current-buffer)))
;; Stage changes.
(gited-add-patched-files (gited-modified-files)) ; Switch to another buf.
(switch-to-buffer cur-buf)
(let* ((commit-msg
(with-current-buffer buf-commit
(buffer-string)))
(args `("commit" "-m" ,commit-msg)))
(if (zerop (gited-git-command args))
(message "Patch applied and committed successfully!")
(error "Cannot commit patch at %s" (buffer-name buf-patch)))))))
;;; Revert a commit.
;; Following is inspired `edit-kbd-macro'.
(defvar gited-original-buffer)
(defvar gited--revert-commit)
(defvar gited-edit-commit-mode-map
(let ((map (make-sparse-keymap)))
(define-key map "\C-c\C-c" 'gited-finish-commit-edit)
map))
(defun gited-finish-commit-edit ()
(interactive)
(unless (eq major-mode 'gited-edit-commit-mode)
(error
"This command is valid only in buffers created by `gited-edit-commit-mode'"))
(goto-char 1)
(while (re-search-forward "^#.*
?" nil t)
(replace-match ""))
(unless (zerop (gited-git-command `("revert" "--no-commit" ,gited--revert-commit)))
(error "Cannot revert commit %s" gited--revert-commit))
(unless (zerop (gited-git-command `("commit" "-m" ,(buffer-string))))
(error "Reverted commit %s but cannot commit the revert" gited--revert-commit))
(switch-to-buffer gited-original-buffer)
(run-hooks 'gited-after-change-hook))
(defun gited-edit-commit-mode ()
(interactive)
(error "This mode can be enabled only by `gited-edit-commit'"))
(put 'gited-edit-commit-mode 'mode-class 'special)
(defun gited-edit-commit (commit)
"Edit message to revert a commit."
(let ((string (format "Revert '%s'\n\nThis reverts commit %s\n\n%s\n%s\n%s\n"
(gited-commit-title commit)
commit
"# Please enter the commit message for your changes. Lines starting"
"# with '#' will be ignored, and an empty message aborts the commit."
"# Press 'C-c C-c' once done to commit. Press C-x k RET to cancel.")))
(let ((oldbuf (current-buffer))
(buf (get-buffer-create "*Edit Commit*")))
(switch-to-buffer buf)
(kill-all-local-variables)
(use-local-map gited-edit-commit-mode-map)
(setq buffer-read-only nil)
(setq major-mode 'gited-edit-commit-mode)
(setq mode-name "Edit Commit")
(set (make-local-variable 'gited-original-buffer) oldbuf)
(set (make-local-variable 'gited--revert-commit) commit)
(erase-buffer)
(insert string)
(recenter '(4)))))
(defun gited-revert-commit (commit)
(interactive
(let ((last-commit
(with-temp-buffer
(gited-git-command '("rev-parse" "HEAD") (current-buffer))
(buffer-substring 1 (1- (point-max))))))
(list (read-string "Revert commit: " last-commit))))
(gited-edit-commit commit))
;;; Common operations on Git repositiores: pull, diff, log, etc.
(defun gited-number-of-commits ()
"Return number of Git commits in current buffer."
(let ((regexp "^commit[:]? \\([[:xdigit:]]+\\)"))
(save-excursion
(goto-char (point-min))
(let ((count 0))
(while (re-search-forward regexp nil t)
(cl-incf count))
(if (called-interactively-p 'interactive)
(message "%d commits in current buffer" count)
count)))))
(defun gited--case-ref-kind ()
(pcase gited-ref-kind
("remote" "remotes/")
("local" "heads/")
("tags" "tags/")
(_ (error "Unsupported gited-ref-kind: must be local, remote or tags"))))
(defun gited--set-output-buffer-mode (buffer &optional mode editable)
(let ((win (get-buffer-window buffer)))
(when win (set-window-point win (point-min))))
(with-current-buffer buffer
(goto-char (point-min))
(set-buffer-modified-p nil)
(or editable (setq buffer-read-only t))
(pcase mode
(`diff (diff-mode))
(`outline (outline-mode))
(_ (fundamental-mode)))))
(defun gited-diff-with-branch (branch)
"Show diff of BRANCH with a second branch.
BRANCH default to the branch at current line.
The actual diff run is:
diff BRANCH SECOND-BRANCH."
(interactive
(list (gited-get-branchname)))
(let ((branch-new
(completing-read (format "Diff %s with: " branch)
(gited-listed-branches)))
(buf (gited--output-buffer)))
(setq gited-output-buffer buf)
(with-current-buffer buf
(let ((inhibit-read-only t))
(erase-buffer)
(gited-git-command `("diff" ,branch ,branch-new)
(current-buffer)))
(display-buffer buf))
(gited--set-output-buffer-mode buf 'diff)))
(defun gited--valid-ref-p (str)
(let ((args `("rev-parse" ,str)))
(zerop (gited-git-command args))))
(defun gited-show-commit (branch &optional commit)
"Show a commit of BRANCH.
BRANCH default to the branch at current line.
Optional arg COMMIT, if non-nil then is a SHA1 string or
HEAD~N, to indicate which commit to display.
Interactively prompt for the limit commit: 0 means HEAD,
1 means HEAD~, and so on."
(interactive
(let* ((branch (gited-get-branchname))
(num-or-sha1
(read-string "Show commit (0 = HEAD, 1 = HEAD~1, ... or SHA1): "
nil nil "0")))
(list branch num-or-sha1)))
(let ((buf (gited--output-buffer))
(args (if (gited--valid-ref-p commit)
`("show" ,commit)
(list "show" (concat (gited--case-ref-kind)
branch "~" commit)))))
(setq gited-output-buffer buf)
(with-current-buffer buf
(let ((inhibit-read-only t))
(erase-buffer)
(gited-git-command args (current-buffer)))
(display-buffer buf))
(gited--set-output-buffer-mode buf 'diff)))
(defun gited-status ()
"Show repository status.
If `magit' is installed, then this calls `magit-status'. Otherwise,
show similar info as that command."
(interactive)
(let ((buf (gited--output-buffer))
(branch gited-current-branch)
(gited-buf gited-buffer))
(setq gited-output-buffer buf)
(if (ignore-errors (find-library-name "magit"))
(with-no-warnings (magit-status-internal default-directory))
(with-current-buffer buf
(let ((inhibit-read-only t)
(untracked (gited-untracked-files))
(unstaged (gited-modified-files))
(bisectingp (gited-bisecting-p))
(stashes (gited-stashes)))
(erase-buffer)
;; (outline-mode)
(insert (format "Head: %s %s\n"
(propertize branch
'font-lock-face 'gited-status-branch-local)
(gited-last-commit-title)))
;; (insert (format "Tag: %s (%s)\n" tag tag-id))
(when bisectingp
(insert (format "\n%s\nCall C-u C-u %s to reset\n"
(propertize
"You are bisecting:"
'font-lock-face 'gited-status-tag)
(with-current-buffer gited-buf
(substitute-command-keys "\\[gited-bisect]"))))
(let ((nentries -1)
bad-commit res)
(with-temp-buffer
(gited-git-command '("bisect" "log") (current-buffer))
(goto-char (point-min))
(save-excursion
(when (re-search-forward
"# first bad commit: \\[\\([[:xdigit:]]+\\)\\]"
nil t)
(setq bad-commit (match-string 1))))
(while (re-search-forward "^[^#]+$" nil t)
(push (buffer-substring-no-properties (point-at-bol) (point))
res))
(setq res (nreverse res) nentries (length res)))
(insert (format "\n%s (%d)\n"
(propertize
"Bisect Log:"
'font-lock-face 'gited-status-tag)
(1+ nentries)))
(while res
(insert (concat (pop res) "\n")))
(when bad-commit
(insert (concat bad-commit " is the first bad commit\n")))))
(when untracked
(insert (format "\n%s (%d)\n"
(propertize "Untracked files"
'font-lock-face 'gited-status-tag)
(length untracked)))
(insert (mapconcat 'identity untracked "\n")))
(when unstaged
(insert (format "\n\n%s (%d)\n"
(propertize "Unstaged changes"
'font-lock-face 'gited-status-tag)
(length unstaged)))
(insert (mapconcat 'identity unstaged "\n")))
(when stashes
(insert (format "\n\n%s (%d)\n"
(propertize "Stashes"
'font-lock-face 'gited-status-tag)
(length stashes)))
(insert (mapconcat 'identity stashes "\n")))
(display-buffer buf)
(gited--set-output-buffer-mode buf 'outline)))
)))
(defun gited-pull ()
"Run git pull in current branch."
(interactive)
(let ((branch gited-current-branch))
(if (not (or gited-expert
(y-or-n-p (format "Pull on '%s' branch? " branch))))
(message "OK, pull canceled")
(let ((buf (gited--output-buffer))
(cmd (format "%s pull" vc-git-program))
(inhibit-read-only t))
(setq gited-output-buffer buf
gited-op-string cmd)
(with-current-buffer buf (erase-buffer))
(gited-async-operation cmd 'remote-op-p)))))
(defun gited-push ()
"Run git push in current branch."
(interactive)
(if (not (or gited-expert
(y-or-n-p (format "Push '%s' branch? "
gited-current-branch))))
(message "OK, push canceled")
(let ((buf (gited--output-buffer))
(cmd (format "%s push" vc-git-program)))
(setq gited-output-buffer buf
gited-op-string cmd)
(with-current-buffer buf
(setq buffer-read-only nil) ; Editable, they can ask username.
(erase-buffer))
(gited-async-operation cmd 'remote-op-p))))
(defun gited-set-branch-upstream (branch)
"Push a local BRANCH to origin."
(interactive
(list (gited-get-branchname)))
(unless (string= gited-ref-kind "local")
(error "Gited should be listing local branches"))
(if (not (or gited-expert
(y-or-n-p (format "Push '%s' branch up stream? "
branch))))
(message "OK, push canceled")
(let ((buf (gited--output-buffer))
(cmd (format "%s push --set-upstream origin %s"
vc-git-program branch))
(inhibit-read-only t))
(setq gited-output-buffer buf
gited-op-string (format "Set branch '%s' upsetream" branch))
(with-current-buffer buf (erase-buffer))
(gited-async-operation cmd 'remote-op-p))))
(defun gited-origin (branch &optional no-display)
"Run git log origin..BRANCH.
BRANCH defaults to the branch at point.
If optional arg NO-DISPLAY is non-nil, then don't display the
output buffer.
Return output buffer.
Note that this command only has sense if you have a remote branch
called origin in your Git configuration. Otherwise, if you wish to
see the newest N commits then use `\\[gited-log-last-n-commits\]'."
(interactive
(list (gited-get-branchname) current-prefix-arg))
(unless (string= gited-ref-kind "local")
(error "Not listing local branches"))
(unless (gited-remote-repository-p)
(error "Not a remote repository. Try '%s' or '%s'"
(substitute-command-keys "\\[gited-log\]")
(substitute-command-keys "\\[gited-log-last-n-commits\]")))
(let ((buf (gited--output-buffer))
(args (list "log" (concat "origin.." (gited--case-ref-kind) branch)))
(gited-buf gited-buffer) count)
(setq gited-output-buffer buf)
(with-current-buffer buf
(let ((inhibit-read-only t) res)
(erase-buffer)
(setq res (gited-git-command args (current-buffer)))
(unless (zerop res)
(with-current-buffer gited-buf
(error "Command 'git log origin..BRANCH' fails. Try '%s' or '%s'"
(substitute-command-keys "\\[gited-log\]")
(substitute-command-keys "\\[gited-log-last-n-commits\]"))))
(goto-char (point-min)))
(setq count (gited-number-of-commits))
(or no-display (display-buffer buf)))
(gited--set-output-buffer-mode buf 'diff)
(message "%d commit%s in '%s'"
count (if (> count 1) "s" "") (buffer-name buf))
buf))
(defun gited-log (branch start-date end-date &optional short)
"Show Git log for BRANCH between START-DATE and END-DATE.
If optional arg SHORT is non-nil, then use a short format.
Interactively, prompt for START-DATE and END-DATE."
(interactive
(let* ((branch (gited-get-branchname))
(last-commit-time
(gited-get-last-commit-time branch))
(default
(format-time-string
"%F"
(time-subtract
last-commit-time
(days-to-time 7))))
(start (read-string "start-date: " default nil nil nil))
(default (format-time-string "%F"))
(end (read-string "end-date: " default nil nil nil)))
(list branch start end current-prefix-arg)))
(let ((buf (gited--output-buffer))
(args (append (if short gited-short-log-cmd '("log"))
(list (concat (gited--case-ref-kind) branch)))))
(setq gited-output-buffer buf)
(with-current-buffer buf
(let ((inhibit-read-only t))
(erase-buffer)
(setq args (append args
(list (format "--after=\"%s\"" start-date))
(list (format "--before=\"%s\"" end-date))))
(gited-git-command args buf))
(display-buffer buf))
(gited--set-output-buffer-mode buf 'outline)))
(defun gited-log-last-n-commits (branch n &optional short)
"Show newest N log entries for BRANCH.
When N is of the form N1-N2, then skip the N1 newest log entries
and show the remaining newest N2 entries.
The actual command run in this case is as follows:
git-log --skip=N1 --max-count=N2.
If optional arg SHORT is non-nil use a short format."
(interactive
(list (gited-get-branchname)
(read-string "Show newest N commits, or those in (N1, N1 + N2]: "
nil nil "1")
current-prefix-arg))
(let ((buf (gited--output-buffer)))
(setq gited-output-buffer buf)
(with-current-buffer buf
(let* ((inhibit-read-only t)
(skip (and (string-match "\\([0-9]+\\)-\\([0-9]+\\)" n)
(string-to-number (match-string 1 n))))
(max (string-to-number (if skip (match-string 2 n) n)))
(args (append (if short gited-short-log-cmd '("log"))
(and skip (list (format "--skip=%d" skip)))
(list (format "--max-count=%d" max))
(list branch))))
(erase-buffer)
(gited-git-command args buf))
(display-buffer buf))
(gited--set-output-buffer-mode buf 'outline)))
;;; Extract patches
(defun gited--clean-previous-patches ()
(mapc (lambda (x)
(when (buffer-live-p x)
(kill-buffer x)))
(nconc
(gited--get-patch-or-commit-buffers)
(gited--get-patch-or-commit-buffers 'commit))))
(defun gited-extract-patches (n &optional origin write-file)
"Extract the patches from the N newest commits.
Optional arg ORIGIN, means extract the patches from all commits accesible
from the trunk, and not being in the trunk.
Optional arg WRITE-FILE if non-nil, then write the patches to disk."
(interactive
(let* ((prefix current-prefix-arg)
(num (unless prefix
(read-string "Extract N newest patches: "
nil nil "1")))
(from-origin (and prefix (equal prefix '(4))))
(write (and prefix (equal prefix '(16)))))
(list num from-origin write)))
(let* ((branch (gited-get-branchname))
(buffer (if origin
(gited-origin branch 'no-display)
(gited-log-last-n-commits branch n)
gited-output-buffer))
num-commits count)
(with-current-buffer buffer
(if (zerop (buffer-size))
(error "No new patches")
;; Previous patch buffers must be deleted.
(gited--clean-previous-patches)
(save-excursion
(goto-char (point-min))
(setq num-commits (gited-number-of-commits)
count num-commits))))
;; Following form must be evalled with branch temporary current.
(gited-with-current-branch branch
(dotimes (i num-commits)
(let ((buf-patch (get-buffer-create (format "*gited-patch-%d*" count)))
(buf-commit (get-buffer-create
(format "*gited-commit-%d*" count))))
(with-current-buffer buf-patch
(gited-git-command
`("format-patch" "-1" ,(format "HEAD~%d" i) "--stdout")
(current-buffer))
(gited--set-output-buffer-mode (current-buffer) 'diff 'editable))
(with-current-buffer buf-commit
(gited-git-command `("show" "-s" "--format=%B" ,(format "HEAD~%d" i))
(current-buffer))
(while (looking-at "^$") ; Delete empty lines.
(delete-char -1)))
(when write-file
(with-temp-file (expand-file-name
(substring (buffer-name buf-patch) 1 -1)
temporary-file-directory)
(insert
(with-current-buffer buf-patch
(buffer-string)))))
(cl-decf count))))
(if write-file
(message "Extracted %d patches and saved in %s"
num-commits temporary-file-directory)
(message "Extracted %d patches" num-commits))) t)
(defun gited--get-patch-or-commit-buffers (&optional commit)
(let ((regexp
(if commit
"\\`\\*gited-commit-\\([0-9]+\\)\\*\\'"
"\\`\\*gited-patch-\\([0-9]+\\)\\*\\'")))
(sort
(cl-delete-if-not
(lambda (x)
(string-match regexp (buffer-name x)))
(buffer-list))
(lambda (x y)
(string< (buffer-name x) (buffer-name y)))
)))
(defun gited-sync-with-trunk (branch-target)
"Extract latest patches in branch at point and apply then into BRANCH-TARGET.
BRANCH-TARGET is a new branch copied from (car (gited-trunk-branches)).
The effect is similar than merge the branch at point with the trunk;
one difference is that we don't modify the trunk, instead we copy it;
another difference that we don't get a 'Merge branch...' commit in the log.
this command set BRANCH-TARGET current."
(interactive
(let* ((br (gited-get-branchname))
(prompt
(format "Syncronized '%s' into new branch: " br))
(def (if (string-match "-new\\([0-9]*\\)\\'" br)
(format "%s%d" (substring br 0 (match-beginning 1))
(1+ (string-to-number (match-string 1 br))))
(concat br "-new1"))))
(list
(completing-read prompt
(gited-listed-branches)
nil nil def))))
;; Previous patch buffers must be deleted.
(gited--clean-previous-patches)
(unless (gited-remote-repository-p)
(error "This command only works for repositories \
tracking a remote repository"))
(if (null (ignore-errors (gited-extract-patches nil t)))
(error "No new patches to apply")
;; If branch-target doesn't exists create it as copy of master.
(unless (member branch-target (gited-listed-branches))
(cond ((gited-trunk-branches)
(gited-copy-branch (car (gited-trunk-branches)) branch-target))
(t (error "I don't know what is your master branch"))))
(let (num-commits)
(gited-with-current-branch branch-target
(let* ((buf-patches
(gited--get-patch-or-commit-buffers))
(buf-commits
(gited--get-patch-or-commit-buffers 'commit)))
(setq num-commits (length buf-patches))
(while buf-patches
(gited-apply-add-and-commit-patch (car buf-patches)
(car buf-commits))
(setq buf-patches (cdr buf-patches)
buf-commits (cdr buf-commits)))))
(gited-checkout-branch branch-target)
(gited-update)
(message "Successfully applied and committed %d commits!"
num-commits))))
(defun gited-bisecting-p ()
"Return non-nil if a Git bisect is on process."
(zerop (gited-git-command '("bisect" "log"))))
(defun gited--bisect-executable-p (command)
(let ((file (car (split-string command))))
(unless (file-executable-p file)
(error "File '%s' not executable" file))))
(defun gited--bisect-after-run (buffer)
(let ((regexp "^[[:xdigit:]]+ is the first bad commit")
pos window)
(gited--set-output-buffer-mode buffer)
(with-current-buffer buffer
(save-excursion
(goto-char (point-min))
(when (re-search-forward regexp nil t)
(setq pos (match-beginning 0))
(put-text-property (match-beginning 0)
(match-end 0)
'font-lock-face 'gited-status-tag))))
(setq window (display-buffer buffer))
(when pos (set-window-point window pos))))
(defun gited-bisect (&optional script reset)
"Execute a Git bisect.
Optional arg SCRIPT if non-nil, then is a script to pass to
git bisect run.
Optional arg RESET if non-nil, then means abort the current bisect.
Interactively, a prefix 'C-u' prompts for SCRIPT; a prefix 'C-u C-u'
set RESET non-nil."
(interactive
(let ((prefix current-prefix-arg))
(list (equal prefix '(4))
(equal prefix '(16)))))
(let ((bisectingp (gited-bisecting-p))
(obuf (gited--output-buffer gited-bisect-buf-name))
(branch (gited-get-branchname)))
;; Ensure output buffer is editable.
(with-current-buffer obuf (setq buffer-read-only nil))
(cond (reset
(if (not bisectingp)
(message "Not bisecting. Nothing to do")
(setq gited--running-async-op nil)
(gited-git-command '("bisect" "reset") obuf)
(message "Successfully reseted git bisect!")))
((not bisectingp)
(with-current-buffer obuf
(erase-buffer))
(let ((bad (read-string "Start bisect with bad/new revision: "
nil nil branch))
(good (read-string "Good/Old revision: " nil nil branch))
(cmd (and script
(read-shell-command "Bisect shell command: "))))
(and cmd
(gited--bisect-executable-p cmd)) ; File must be executable.
(gited-git-command `("bisect" "start" ,bad ,good) obuf)
(when cmd
;; (when (zerop (gited-git-command `("bisect" "run" ,cmd) obuf))
;; (gited--bisect-after-run obuf))
(gited-async-operation
(format "%s bisect run %s" vc-git-program cmd) nil obuf)
(setq gited-op-string "bisect run"))
(display-buffer obuf)))
((and bisectingp script)
(let ((cmd
(and script (read-shell-command "Bisect shell command: "))))
(and cmd
(gited--bisect-executable-p cmd)) ; File must be executable.
;; (when (zerop (gited-git-command `("bisect" "run" ,cmd) obuf))
;; (gited--bisect-after-run obuf))
(gited-async-operation
(format "%s bisect run %s" vc-git-program cmd) nil obuf)
(setq gited-op-string "bisect run"))
(display-buffer obuf))
(t
(let* ((is-badp (y-or-n-p "Is current revision Bad/New? "))
(is-goodp (and (not is-badp)
(y-or-n-p "Is current revision Good/Old? ")))
(skip (and (not (or is-badp is-goodp))
(y-or-n-p "Do you want to skip this commit? ")))
(args
(cond (is-badp '("bisect" "bad"))
(is-goodp '("bisect" "good"))
(skip '("bisect" "skip"))
(t (error "Commit should be either bad, \
good or skip")))))
(gited-git-command args obuf)
(display-buffer obuf))))))
;;; Git Stash commands.
(defun gited-stash (&optional untracked)
"Stash the changes in a dirty working directory away.
If called with prefix argument, then include untracked files. With two
prefix arguments includes the ignored files as well."
(interactive
(let ((prefix current-prefix-arg))
(list (cond ((and prefix (equal prefix '(4))) "--include-untracked")
((and prefix (equal prefix '(16))) "--all")))))
(let* ((msg (read-string
"Stash message: "
(format "WIP on %s: " (gited-current-branch))))
(args (append '("stash") '("save") `(,msg)
(and untracked `(,untracked)))))
(gited-git-command args)))
(defun gited-stash-apply ()
"Apply a stash to the working tree."
(interactive)
(if (null (gited-stashes))
(message "Empty stash list")
(let* ((stash
(read-string "Apply stash: " nil nil "stash@{0}"))
(args `("stash" "apply" ,stash)))
(gited-git-command args))))
(defun gited-stash-pop ()
"Apply a stash to the working tree and remove it from stash list."
(interactive)
(if (null (gited-stashes))
(message "Empty stash list")
(let* ((stash
(read-string "Apply pop: " nil nil "stash@{0}"))
(args `("stash" "pop" ,stash)))
(gited-git-command args))))
(defun gited--stash-branch ()
(cond ((null (gited-stashes))
(error "Empty stash list"))
((gited-modified-files)
(error "Commit your local changes before you switch branches"))
(t)))
(defun gited-stash-branch (branch stash)
"Create and checkout a new BRANCH from STASH."
(interactive
(let ((stash
(and (gited--stash-branch)
(read-string "Branch stash: " nil nil "stash@{0}")))
(br (read-string "Branch name: ")))
(list br stash)))
(when (gited--stash-branch)
(let ((args `("stash" "branch" ,branch ,stash)))
(gited-git-command args))))
(defun gited-stash-drop ()
"Remove a stash from the stash list."
(interactive)
(if (null (gited-stashes))
(message "Empty stash list")
(let* ((stash
(read-string "Drop stash: " nil nil "stash@{0}"))
(args `("stash" "drop" ,stash)))
(gited-git-command args))))
(defun gited-branch-clear ()
"Remove all stashes from the stash list."
(interactive)
(if (null (gited-stashes))
(message "Empty stash list")
(if (y-or-n-p "Remove all stashes? ")
(gited-git-command '("stash" "clear"))
(error "OK, canceled"))))
(defalias 'gited-delete-all-stashes 'gited-branch-clear)
;;; Moving around.
(defun gited-next-line (&optional arg)
"Go to start of branch name in next ARG lines."
(interactive "p")
(forward-line arg)
(condition-case nil
(gited-move-to-branchname)
(error
(forward-line -1)
(gited-move-to-branchname)
(message "At last branch!")
(ding)
(sit-for 1)
(message nil))))
(defun gited-prev-line (&optional arg)
"Go to start of branch name in previous ARG lines."
(interactive "p")
(let ((oline (line-number-at-pos))
nb-line)
(when (natnump arg) (setq arg (- arg)))
(forward-line arg)
(gited-move-to-branchname)
(setq nb-line (line-number-at-pos))
(when (or (= oline nb-line)
(< (- oline (abs arg)) nb-line))
(message "At first branch!")
(ding)
(sit-for 1)
(message nil))))
(defun gited--goto-first-branch ()
(goto-char (point-min))
(when (overlays-at (point-min))
(forward-line)))
(defun gited-goto-first-branch ()
"Go to first branch in current Gited buffer."
(interactive)
(gited--goto-first-branch)
(gited-move-to-branchname))
(defun gited-goto-last-branch ()
"Go to last branch in current Gited buffer."
(interactive)
(goto-char (point-max))
(forward-line -1)
(gited-move-to-branchname))
(defun gited--goto-column (col)
(forward-line 0)
(dotimes (_ (1- col))
(goto-char
(next-single-property-change
(point)
'tabulated-list-column-name))))
(defun gited-goto-branch (branch)
"Go to line describing BRANCH in this Gited buffer.
Return buffer position on success, else nil."
(interactive
(let ((cur-branch gited-current-branch))
(list (completing-read "Jump to branch: "
(gited-listed-branches)
nil t nil nil cur-branch))))
(let ((row (cl-position branch (gited-listed-branches) :test #'equal)))
(goto-char (point-min))
(forward-line (if (overlays-at (point-min)) (1+ row) row))
(gited--goto-column (1+ gited-branch-idx))))
(defun gited-next-marked-branch (arg &optional wrap opoint)
"Move to the next ARG marked branch.
If WRAP is non-nil, wrap around to the beginning of the buffer if
we reach the end."
(interactive "p\np")
(or opoint (setq opoint (point)));; return to where interactively started
(if (if (> arg 0)
(re-search-forward gited-re-mark nil t arg)
(beginning-of-line)
(re-search-backward gited-re-mark nil t (- arg)))
(gited-move-to-branchname)
(if (null wrap)
(progn
(goto-char opoint)
(error "No next marked branch"))
(message "(Wraparound for next marked branch)")
(goto-char (if (> arg 0) (point-min) (point-max)))
(gited-next-marked-branch arg nil opoint))))
(defun gited-prev-marked-branch (arg &optional wrap)
"Move to the previous ARG marked branch.
If WRAP is non-nil, wrap around to the end of the buffer if we
reach the beginning of the buffer."
(interactive "p\np")
(gited-next-marked-branch (- arg) wrap))
;; Fill and return `gited-branch-alist'.
(defun gited-get-element-in-row (idx)
(let ((row (tabulated-list-get-entry)))
(if row
(aref row idx)
(error "No row at point"))))
(defun gited-get-branchname ()
(gited-get-element-in-row gited-branch-idx))
(defun gited-get-mark ()
(gited-get-element-in-row gited-mark-idx))
(defun gited-get-date ()
(gited-get-element-in-row gited-date-idx))
(defun gited-get-commit ()
(gited-get-element-in-row gited-commit-idx))
(defalias 'gited--move-to-column #'gited--goto-column)
(defun gited--move-to-end-of-column (col)
(gited--move-to-column col)
(goto-char (next-single-property-change (point) 'tabulated-list-column-name)))
(defun gited--fill-branch-alist (&optional pattern)
(let* ((args (append (butlast gited-list-refs-format-command)
(list (format (car (last gited-list-refs-format-command))
(if pattern
(pcase pattern
("local" "heads")
("remote" "remotes")
(_ pattern))
"heads")))))
(alist
(with-temp-buffer
(insert "(\n")
(unless (zerop (gited-git-command args (current-buffer)
nil 'unquote))
(error "No Git repository in current directory"))
(insert ")")
(mapcar (lambda (x)
(when (stringp (car x)) ; No time: set it to beginning
; of epoch.
(push 0 x))
(when (= (length x) 4) ; Drop time zone.
(setf (cdr x) (cddr x)))
(when (and (stringp (car (last x))) ; If no Author, set
; it Unknown.
(string= "" (car (last x))))
(setf (car (last x)) "Unknown"))
x)
(car (read-from-string (buffer-string))))))
(prep
(make-progress-reporter
"Collecting brach info..."
0 (length alist))))
(cl-flet ((format-time-fn (time-secs)
(format-time-string
gited-date-format
(apply #'encode-time
(decode-time
(if (< emacs-major-version 25)
(seconds-to-time time-secs)
time-secs)))))
(get-mark-fn (x)
(let ((table
(save-excursion
(gited--goto-first-branch)
(ignore-errors (gited-get-mark)))))
(if (and table (ignore-errors
(gited-goto-branch (nth 1 x))))
(cons (gited-get-mark) nil)
'(" ")))))
;; Get title of first commit for each listed branch.
(setq gited-branch-alist
(cl-loop for entry the elements of alist using (index idx) collect
(progn
(progress-reporter-update prep idx)
(let* ((args (list "log"
"--pretty=format:%h | %s"
(cadr entry) "-n1" "--"))
(str (with-temp-buffer
(gited-git-command args
(current-buffer) nil)
(buffer-string))))
;; Format time in seconds as `gited-date-format'.
(setf (car entry) (format-time-fn (car entry)))
(append `(,(1+ idx)) (get-mark-fn entry)
entry `(,str)))))))
(progress-reporter-done prep)
gited-branch-alist))
;;; Toggles.
(defun gited-hide-details-update-invisibility-spec ()
(let ((col-names
(if gited-hide-details-mode
'("M" "Branches" "" "" "Last Commit")
'("M" "Authors" "Date" "Branches" "Last Commit")))
(col-sizes
(if gited-hide-details-mode
'(2 50 -1 -1 65)
'(2 16 17 50 65))))
(gited--list-format-init col-names col-sizes)
(setq tabulated-list-format gited-list-format)
(funcall (if gited-hide-details-mode
'add-to-invisibility-spec
'remove-from-invisibility-spec)
'gited-hide-details-author)
(funcall (if gited-hide-details-mode
'add-to-invisibility-spec
'remove-from-invisibility-spec)
'gited-hide-details-date)
(tabulated-list-init-header)
(gited--update-padding (not gited-hide-details-mode))))
(defun gited--update-padding (undo)
"Update columns padding after `gited-hide-details-mode'."
(let ((inhibit-read-only t))
(save-excursion
(gited-goto-first-branch)
(while (not (eobp))
(gited-move-to-end-of-branchname)
(skip-chars-backward " \t")
(forward-char 1)
(when (equal (get-text-property (point) 'display)
`(space :align-to ,(if undo 54 89)))
(put-text-property (point) (1+ (point))
'display `(space :align-to ,(if undo 89 54)))
(delete-region (1+ (point))
(gited-move-to-end-of-branchname)))
(forward-line 1))) nil))
;;; Marked branches.
(defun gited-map-lines (fn)
"Map FN on each Gited line."
(let (br-name mark)
(save-excursion
(gited--goto-first-branch)
(while (ignore-errors
(setq br-name (gited-get-branchname)
mark (gited-get-mark)))
(funcall fn br-name mark)
(forward-line)))))
(defun gited-get-marked-branches (&optional marker)
"Return a list of branches currently marked."
(delq nil
(mapcar (lambda (e)
(when (equal (string-to-char (cdr e))
(or marker gited-marker-char))
(car e)))
(gited-current-state-list))))
(defun gited-current-state-list (&optional pos)
"Return a list like (BRANCH . MARK) of all branches in an Gited buffer.
If POS is non-nil, return a list like (BRANCH MARK POINT), where POINT is
the value of point at the beginning of the line for that buffer."
(let ((gited-current-state-list-tmp '()))
(if pos
(gited-map-lines
(lambda (branch mark)
(push (list branch mark (point))
gited-current-state-list-tmp)))
(gited-map-lines
(lambda (branch mark)
(push (cons branch mark) gited-current-state-list-tmp))))
(nreverse gited-current-state-list-tmp)))
(defun gited-current-branches-with-marks ()
"Return a list like (BRANCH . MARK) of all listed branches."
(let ((branches (gited-current-state-list)))
(mapcar (lambda (x)
(let ((e (assoc x branches)))
(if e
e
(cons x ?\s))))
(gited-listed-branches))))
;;; Mark/unmark.
(defun gited-remember-marks (beg end)
"Return alist of branches and their marks, from BEG to END."
(if selective-display ; must unhide to make this work.
(let ((inhibit-read-only t))
(subst-char-in-region beg end ?\r ?\n)))
(let (branch chr alist)
(save-excursion
(goto-char beg)
(while (re-search-forward gited-re-mark end t)
(when (setq branch (gited-get-branchname))
(setq chr (preceding-char)
alist (cons (cons branch chr) alist)))))
alist))
(defun gited-mark-remembered (alist)
"Mark all branches remembered in ALIST.
Each element of ALIST looks like (BRANCH . MARKERCHAR)."
(let (elt branch chr)
(save-excursion
(while alist
(setq elt (car alist)
alist (cdr alist)
branch (car elt)
chr (cdr elt))
(when (gited-goto-branch branch)
(beginning-of-line)
(delete-char 1)
(insert chr))))))
(defun gited-fontify-current-branch ()
"Set font for current branch."
(let ((inhibit-read-only t))
(save-excursion
(gited-goto-branch gited-current-branch)
(when (string= " " (gited-get-mark))
(remove-text-properties
(point-at-bol) (point-at-eol) '(face))
(put-text-property
(point)
(gited-move-to-end-of-branchname)
'face gited-current-branch-face)
(put-text-property
(point-at-bol)
(point-at-eol)
'face gited-section-highlight-face)))))
(defun gited-fontify-marked-branch-name (&optional mark)
"Set font for a marked branch."
(let ((marker (or mark (string-to-char (gited-get-mark))))
(inhibit-read-only t))
(gited-move-to-branchname)
(remove-text-properties
(point-at-bol) (point-at-eol) '(face))
(cond ((eq marker ?\s)
(if (string= (gited-get-branchname)
gited-current-branch)
(put-text-property
(point-at-bol)
(point-at-eol)
'face gited-section-highlight-face)
(gited--fontify-current-row)))
((eq marker gited-marker-char)
(put-text-property
(point-at-bol) (1+ (point-at-bol)) 'face gited-flag-mark-face)
(put-text-property
(1+ (point-at-bol))
(point-at-eol)
'face gited-flag-mark-line-face))
((eq marker gited-del-char)
(put-text-property
(point-at-bol) (1+ (point-at-bol)) 'face gited-deletion-face)
(put-text-property
(1+ (point-at-bol))
(point-at-eol)
'face gited-deletion-branch-face)))))
(defun gited-insert-marker-char (&optional marker)
(tabulated-list-set-col gited-mark-idx
(char-to-string (or marker gited-marker-char))
'change))
(defun gited-flag-branch-deletion (arg &optional interactive)
"In Gited, flag the branch at current line (or next ARG) for deletion.
If the region is active, flag all branches in the region.
Otherwise, with a prefix arg, flag branches on the next ARG lines.
If the region is active in Transient Mark mode, flag all branches
in the active region."
(interactive (list current-prefix-arg t))
(gited-mark arg gited-del-char interactive))
(defun gited-toggle-marks ()
"Toggle marks: marked branches become unmarked, and vice versa.
Branches marked with other flags (such as `D') are not affected.
As always, hidden subdirs are not affected."
(interactive)
(save-excursion
(gited--goto-first-branch)
(while (not (eobp))
(let* ((mark (string-to-char (gited-get-mark)))
(flag
(cond ((eq ?\s mark) gited-marker-char)
((eq gited-marker-char mark) ?\s)
(t nil))))
(when flag
(gited-insert-marker-char flag)
(gited-fontify-marked-branch-name flag)))
(forward-line))))
(defun gited-kill-line (&optional arg)
"Kill the current line or next ARG lines (not the branches).
With a prefix argument, kill that many lines starting with the current line.
\(A negative argument kills backward.)"
(interactive "P")
(setq arg (prefix-numeric-value arg))
(let (buffer-read-only branch)
(while (/= 0 arg)
(setq branch (gited-get-branchname))
(if (not branch)
(error "Can only kill branch lines")
(setq tabulated-list-entries
(assq-delete-all
(car (tabulated-list-delete-entry))
tabulated-list-entries))
(if (> arg 0)
(cl-decf arg)
(cl-incf arg)
(forward-line -1))))
(gited-move-to-branchname)))
(defun gited-do-kill-lines (&optional arg fmt)
"Kill all marked lines (not the branches).
With a prefix argument, kill that many lines starting with the current line.
\(A negative argument kills backward.)"
;; Returns count of killed lines. FMT="" suppresses message.
(interactive "P")
(if arg
(gited-kill-line arg)
(save-excursion
(gited--goto-first-branch)
(let (buffer-read-only
(count 0)
(regexp
(concat "^" (regexp-quote (char-to-string gited-marker-char)))))
(while (and (not (eobp))
(re-search-forward regexp nil t))
(cl-incf count)
(setq tabulated-list-entries
(assq-delete-all
(car (tabulated-list-delete-entry))
tabulated-list-entries)))
(or (equal "" fmt)
(message (or fmt "Killed %d line%s.") count (gited-plural-s count)))
count))))
(defun gited-mark-branches-regexp (regexp &optional marker-char)
"Mark all files matching REGEXP for use in later commands.
A prefix argument means to unmark them instead.
REGEXP is an Emacs regexp, not a shell wildcard. Thus, use `\\.o$' for
object files--just `.o' will mark more than you might think."
(interactive
(list (read-regexp (concat (if current-prefix-arg "Unmark" "Mark")
" branches (regexp): ")
nil)
(and current-prefix-arg ?\s)))
(let ((gited-marker-char (or marker-char gited-marker-char)))
(gited-mark-if
(and (not (eolp)) ; empty line
(let ((fn (gited-get-branchname)))
(and fn (string-match-p regexp fn))))
"matching branch")))
(defun gited-date-string-to-time (str)
(apply #'encode-time
(parse-time-string str)))
(defun gited-mark-branches-containing-regexp (regexp &optional marker-char days)
"Mark all branches containing REGEXP in some commit message.
A prefix argument means to unmark them instead.
Optional arg DAYS, if non-nil then limit the search to DAYS before the
newest commit. Otherwise, limit the search to commits until 1 month earlier
than the newest commit.
In interactive calls, a prefix C-u C-u prompts for DAYS."
(interactive
(let ((def current-prefix-arg)
regex marker interval)
(pcase def
(`(16) (setq interval
(string-to-number
(read-string "Number of days before newest commit: "
nil nil "30"))
marker (pcase (read-number "Mark (0) or Unmark (1): " 0)
(0 gited-marker-char)
(1 ?\s)
(_ gited-marker-char))
regex (read-string
(format "%s branches containing regexp: "
(if (char-equal marker gited-marker-char)
"Mark" "UNmark")))))
(_ (setq interval 30
marker (if current-prefix-arg
?\s
gited-marker-char)
regex (read-string
(format "%s branches containing regexp: "
(if (char-equal marker gited-marker-char)
"Mark" "UNmark"))))))
(list regex marker interval)))
(let ((gited-marker-char (or marker-char gited-marker-char)))
(gited-mark-if
(and (not (eolp))
(gited-get-branchname)
(let* ((fn (gited-get-branchname))
(time-max
(gited-date-string-to-time
(car
(cddr
(cl-find-if
(lambda (x)
(string= (nth 3 x) fn)) gited-branch-alist)))))
(time-min (time-subtract time-max (days-to-time (or days 30))))
(args (list "log"
(format "--after=%s"
(format-time-string "%F" time-min))
(format "--before=%s"
(format-time-string "%F" time-max))
(format "--grep=%s" regexp)
fn "--")))
(with-temp-buffer
(gited-git-command args (current-buffer))
(not (string= "" (buffer-string))))))
"matching branch")))
(defun gited-mark-branches-containing-commit (commit &optional marker-char)
"Mark all branches containing COMMIT.
A prefix argument means to unmark them instead.
COMMIT is the sha1 of the commit."
(interactive
(list (read-string (format "%s branches containing sha1 commit: "
(if current-prefix-arg "UNmark" "Mark")))
(and current-prefix-arg ?\s)))
(let* ((args `("branch" ,(concat "--contains=" commit)))
(branches (gited--get-branches-from-command args))
(gited-marker-char (or marker-char gited-marker-char)))
(gited-mark-if
(and (not (eolp))
(let ((fn (gited-get-branchname)))
(and fn (member fn branches))))
"matching branch")))
(defvar gited--op nil)
(defun gited--mark-merged-or-unmerged-branches-spec (op-val)
(setq gited--op op-val)
(read-string (format "%s branches %s with branch: "
(if current-prefix-arg "UNmark" "Mark")
(pcase gited--op
(`merged "merged")
(_ "unmerged")))
nil nil gited-current-branch))
(defun gited--mark-merged-branches-spec (branch prefix)
(interactive "i\nP")
(ignore branch prefix)
(list
(gited--mark-merged-or-unmerged-branches-spec 'merged)
(if current-prefix-arg ?\s gited-marker-char)))
(defun gited--mark-unmerged-branches-spec (branch prefix)
(interactive "i\nP")
(ignore branch prefix)
(list
(gited--mark-merged-or-unmerged-branches-spec 'unmerged)
(if current-prefix-arg ?\s gited-marker-char)))
(defun gited--mark-merged-or-unmerged-branches (branch op marker-char)
(let* ((args
(list "branch"
(if (string= op "merged") "--merged" "--no-merged")
branch))
(branches (gited--get-branches-from-command args))
(gited-marker-char (or marker-char gited-marker-char)))
(gited-mark-if
(and (not (eolp))
(let ((fn (gited-get-branchname)))
(and fn (member fn branches))))
"matching branch")))
(defun gited-mark-merged-branches (branch &optional marker-char)
"Mark all merged branches with BRANCH.
A prefix argument means to unmark them instead."
(interactive
(call-interactively #'gited--mark-merged-branches-spec))
(gited--mark-merged-or-unmerged-branches branch "merged" marker-char))
(defun gited-mark-unmerged-branches (branch &optional marker-char)
"Mark all unmerged branches with BRANCH.
A prefix argument means to unmark them instead."
(interactive
(call-interactively #'gited--mark-unmerged-branches-spec))
(gited--mark-merged-or-unmerged-branches branch "unmerged" marker-char))
(defun gited-repeat-over-lines (arg function)
;; This version skips non-file lines.
(let ((pos (make-marker)))
(beginning-of-line)
(while (and (> arg 0) (not (eobp)))
(cl-decf arg)
(beginning-of-line)
(when (gited-at-header-line-p) (forward-line))
(save-excursion
(forward-line)
(move-marker pos (1+ (point))))
(save-excursion (funcall function))
;; Advance to the next line--actually, to the line that *was* next.
;; (If FUNCTION inserted some new lines in between, skip them.)
(goto-char pos))
(while (and (< arg 0) (not (bobp)))
(cl-incf arg)
(forward-line -1)
(while (not (gited-at-header-line-p)) (forward-line -1))
(beginning-of-line)
(save-excursion (funcall function)))
(move-marker pos nil)
(ignore-errors (gited-move-to-branchname))
(when (eobp)
(forward-line -1)
(ignore-errors (gited-move-to-branchname)))))
(defun gited-mark (arg mark &optional interactive)
"Mark the branch at point in the Gited buffer with MARK.
If the region is active, mark all branches in the region.
Otherwise, with a prefix arg, mark branches on the next ARG lines.
Use \\[gited-unmark-all-branches] to remove all marks
and \\[gited-unmark] on a subdir to remove the marks in
this subdir."
(interactive (list current-prefix-arg gited-marker-char t))
(cond
;; Mark branches in the active region.
((and interactive (use-region-p))
(save-excursion
(let ((beg (region-beginning))
(end (region-end)))
(gited--mark-branches-in-region
(progn (goto-char beg) (point-at-bol))
(progn (goto-char end) (point-at-bol))
mark))))
;; Mark the current (or next ARG) branches.
(t
(let ((inhibit-read-only t))
(gited-repeat-over-lines
(prefix-numeric-value arg)
(lambda ()
(when (ignore-errors (gited-get-branchname))
(gited-insert-marker-char mark)
(gited-fontify-marked-branch-name mark))))))))
(defun gited--mark-branches-in-region (start end mark)
(when (> start end)
(error "Wrong input values: start, end, <"))
(goto-char start) ; assumed at beginning of line
(while (< (point) end)
(when (gited-get-branchname)
(gited-insert-marker-char mark)
(gited-fontify-marked-branch-name mark))
(forward-line)))
(defun gited-unmark-backward ()
"Unmark the branches in the region, or ARG branches."
(interactive)
(forward-line -1)
(when (gited-get-branchname)
(gited-insert-marker-char ?\s)
(gited-fontify-marked-branch-name ?\s)
(gited-move-to-branchname)))
(defun gited-unmark-all-branches (mark)
"Remove all branches with flag MARK."
(interactive "cRemove marks (RET means all):")
(save-excursion
(gited--goto-first-branch)
(while (not (eobp))
(let ((str (aref (tabulated-list-get-entry)
gited-mark-idx)))
(when (and (gited-get-branchname)
(or (and (eq mark ?\r) (not (string= str " ")))
(string= (char-to-string mark) str)))
(gited-insert-marker-char ?\s)
(gited-fontify-marked-branch-name ?\s))
(forward-line)))))
(defun gited-unmark-all-marks ()
"Remove all marks from all marked branches in the Gited buffer."
(interactive)
(gited-unmark-all-branches ?\r))
(defun gited-move-to-branchname ()
(interactive)
(gited--move-to-column (1+ gited-branch-idx)))
;; Return point.
(defun gited-move-to-end-of-branchname ()
(interactive)
(gited--move-to-end-of-column (1+ gited-branch-idx)))
(defun gited-move-to-author ()
(interactive)
(gited--move-to-column (1+ gited-author-idx)))
(defun gited-move-to-end-of-author ()
(interactive)
(gited--move-to-end-of-column (1+ gited-author-idx)))
(defun gited-move-to-date ()
(interactive)
(gited--move-to-column (1+ gited-date-idx)))
(defun gited-move-to-end-of-date ()
(interactive)
(gited--move-to-end-of-column (1+ gited-date-idx)))
(defun gited-unmark (arg &optional interactive)
"Unmark the branch at point in the Gited buffer.
If the region is active, unmark all branches in the region.
Otherwise, with a prefix arg, unmark branches on the next ARG lines.
If the region is active in Transient Mark mode, unmark all branches
in the active region."
(interactive (list current-prefix-arg t))
(gited-mark arg ?\s interactive))
(defun gited-number-marked ()
"Return number of marked files."
(interactive)
(save-excursion
(goto-char (point-min))
(let ((count 0))
(while (not (eobp))
(unless (looking-at "^[[:blank:]]")
(cl-incf count))
(forward-line))
(prog1 count
(if (zerop count)
(message "No marked branches")
(message "%d marked %s"
count
(if (> count 1)
"branches" "branch")))))))
;;; Mode map.
(defvar gited-mode-map
(let ((map (make-keymap)))
(define-key map (kbd "t") 'gited-toggle-marks)
(define-key map (kbd "(") 'gited-hide-details-mode)
(define-key map (kbd "u") 'gited-unmark)
(define-key map (kbd "=") 'gited-diff-with-branch)
(define-key map (kbd "j") 'gited-goto-branch)
(define-key map (kbd "DEL") 'gited-unmark-backward)
(define-key map (kbd "U") 'gited-unmark-all-marks)
(define-key map (kbd "M-DEL") 'gited-unmark-all-branches)
(define-key map (kbd "x") 'gited-do-flagged-delete)
;; moving
(define-key map (kbd "<") 'gited-goto-first-branch)
(define-key map (kbd ">") 'gited-goto-last-branch)
(define-key map (kbd "M-<") 'gited-goto-first-branch)
(define-key map (kbd "M->") 'gited-goto-last-branch)
(define-key map (kbd "n") 'gited-next-line)
(define-key map (kbd "SPC") 'next-line)
(define-key map (kbd "p") 'gited-prev-line)
(define-key map (kbd "M-}") 'gited-next-marked-branch)
(define-key map (kbd "M-{") 'gited-prev-marked-branch)
;; immediate operations
(define-key map (kbd "a") 'gited-apply-patch)
(define-key map (kbd "A") 'gited-add-patched-files)
(define-key map (kbd "B") 'gited-bisect)
(define-key map (kbd "C-c c") 'gited-commit)
(define-key map (kbd "w") 'gited-copy-branchname-as-kill)
(define-key map (kbd "e") 'gited-extract-patches)
(define-key map (kbd "T") 'gited-sync-with-trunk)
(define-key map (kbd "M") 'gited-merge-branch)
(define-key map (kbd "c") 'gited-checkout-branch)
(define-key map (kbd "v") 'gited-visit-branch-sources)
(define-key map (kbd "f") 'gited-visit-branch-sources)
(define-key map (kbd "s") 'gited-show-commit)
(define-key map (kbd "S") 'gited-status)
(define-key map (kbd "RET") 'gited-visit-branch-sources)
(define-key map (kbd "g") 'gited-update)
(define-key map "\C-x\C-g" 'gited-list-branches)
(define-key map (kbd "k") 'gited-do-kill-lines)
(define-key map (kbd "P") 'gited-pull)
(define-key map (kbd "r") 'gited-reset-branch)
(define-key map (kbd "* p") 'gited-set-branch-upstream)
(define-key map (kbd "* <") 'gited-pull)
(define-key map (kbd "* >") 'gited-push)
(define-key map (kbd "o") 'gited-origin)
(define-key map (kbd "l") 'gited-log)
(define-key map (kbd "L") 'gited-log-last-n-commits)
;; marking banches
(define-key map (kbd "m") 'gited-mark)
(define-key map (kbd "% n") 'gited-mark-branches-regexp)
(define-key map (kbd "% c") 'gited-mark-branches-containing-commit)
(define-key map (kbd "% g") 'gited-mark-branches-containing-regexp)
(define-key map (kbd "% m") 'gited-mark-merged-branches)
(define-key map (kbd "% M") 'gited-mark-unmerged-branches)
(define-key map (kbd "d") 'gited-flag-branch-deletion)
;; Git stash
(define-key map (kbd "* s") 'gited-stash)
(define-key map (kbd "* a") 'gited-stash-apply)
(define-key map (kbd "* A") 'gited-stash-pop)
(define-key map (kbd "* b") 'gited-stash-branch)
(define-key map (kbd "* d") 'gited-stash-drop)
;; marked operations
(define-key map (kbd "* N") 'gited-number-marked)
(define-key map (kbd "R") 'gited-rename-branch)
(define-key map (kbd "C") 'gited-copy-branch)
(define-key map (kbd "D") 'gited-do-delete)
(define-key map (kbd "?") 'gited-summary)
map))
;;;###autoload
(defun gited-list-branches (&optional pattern other-window update)
"List all branches with the time of its last commit."
(interactive
(progn
(unless (gited-dir-under-Git-control-p)
(error "No Git repository in current directory"))
(let* ((opts '("local" "remote" "tags"))
(patt (completing-read
"List (local, remote, tags): "
opts nil t nil nil "local")))
(list patt current-prefix-arg nil))))
(if (and (buffer-live-p gited-buffer)
(not update)
(or (not pattern)
(equal pattern gited-ref-kind)))
(switch-to-buffer gited-buffer)
(unless (gited-dir-under-Git-control-p)
(error "No Git repository in current directory"))
(let ((buf (or (and (buffer-live-p gited-buffer) gited-buffer)
(setq gited-buffer (generate-new-buffer gited-buffer-name)))))
(unless (equal pattern gited-ref-kind)
(setq gited-ref-kind pattern))
(if other-window
(switch-to-buffer-other-window buf)
(switch-to-buffer buf))
(or gited-mode (gited-mode))
;; Set `gited-toplevel-dir' if not set yet.
(unless gited-toplevel-dir
(setq gited-toplevel-dir
(with-temp-buffer
(gited-git-command '("rev-parse" "--show-toplevel")
(current-buffer))
(file-name-as-directory
(buffer-substring 1 (1- (point-max)))))))
(setq tabulated-list-use-header-line gited-use-header-line
gited-buffer buf
gited-ref-kind pattern
gited-current-branch (gited-current-branch)
tabulated-list-printer #'gited-print-entry)
;; Ignore dired-hide-details-* value of invisible text property by default.
(when (eq buffer-invisibility-spec t)
(setq buffer-invisibility-spec (list t)))
(gited-tabulated-list-entries)
(tabulated-list-print)
(gited-goto-branch gited-current-branch)
(gited-fontify-current-branch)
(unless gited--hide-details-set
(or gited-verbose (gited-hide-details-mode 1))
(setq gited--hide-details-set t)))))
(defun gited-print-entry (id cols)
"Insert a Gited entry at point.
ID is a Lisp object identifying the entry to print, and COLS is a vector
of column descriptors."
(let ((beg (point))
(x (max tabulated-list-padding 0))
(ncols (length tabulated-list-format))
(inhibit-read-only t))
(if (> tabulated-list-padding 0)
(insert (make-string x ?\s)))
(dotimes (n ncols)
(let ((pos (point)))
(setq x (tabulated-list-print-col n (aref cols n) x))
(cond
((= n gited-author-idx)
(add-text-properties
pos (point)
`(invisible gited-hide-details-author
font-lock-face ,gited-author-face)))
((= n gited-date-idx)
(add-text-properties
pos (point)
`(invisible gited-hide-details-date
font-lock-face ,gited-date-time-face)))
((= n gited-branch-idx)
(put-text-property
pos (point)
'font-lock-face gited-branch-name-face))
((= n gited-commit-idx)
(put-text-property
pos (point)
'font-lock-face gited-commit-msg-face))
(t nil))
))
(insert ?\n)
;; Ever so slightly faster than calling `put-text-property' twice.
(add-text-properties
beg (point)
`(tabulated-list-id ,id tabulated-list-entry ,cols))))
(defun gited-tabulated-list-entries ()
(let ((alist (gited--fill-branch-alist gited-ref-kind))
mark author date branch commit id values result)
(while alist
(setq id (caar alist)
values (cdr (car alist))
mark (nth 0 values)
author (nth 3 values)
date (nth 1 values)
branch (nth 2 values)
commit (nth 4 values)
alist (cdr alist))
(push (list id (vector mark author date branch commit)) result))
(setq tabulated-list-entries (nreverse result))
(tabulated-list-init-header)))
;;; Define minor mode.
(define-derived-mode gited-mode tabulated-list-mode "Gited"
"Toggle gited-mode.
Interactively with no argument, this command toggles the mode.
A positive prefix argument enables the mode, any other prefix
argument disables it. From Lisp, argument omitted or nil enables
the mode, `toggle' toggles the state.
Mode to edit Git branches as Dired."
(unless (gited-buffer-p)
(error "Gited mode cannot be enabled in this buffer"))
(gited--list-format-init)
(setq tabulated-list-format gited-list-format)
(add-hook 'tabulated-list-revert-hook 'gited-tabulated-list-entries nil t)
(setq tabulated-list-sort-key '("Date")))
(provide 'gited)
;;; gited.el ends here
--8<-----------------------------cut here---------------end--------------->8---
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-05-21 3:49 ` Jean-Christophe Helary
@ 2017-05-22 12:15 ` Richard Stallman
0 siblings, 0 replies; 51+ messages in thread
From: Richard Stallman @ 2017-05-22 12:15 UTC (permalink / raw)
To: Jean-Christophe Helary; +Cc: emacs-devel
[[[ To any NSA and FBI agents reading my email: please consider ]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> Are there references that document the _recommended_ way ?
For this issue, the recommended way is whatever developers recommend
to the public -- whether it is in the manual or not. If the manual
says, "Do X", and developers tell people, "Do Y instead, that's
better," then Y is in effect a recommended way.
--
Dr Richard Stallman
President, Free Software Foundation (gnu.org, fsf.org)
Internet Hall-of-Famer (internethalloffame.org)
Skype: No way! See stallman.org/skype.html.
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2017-04-27 16:16 ` Ted Zlatanov
@ 2018-07-23 13:14 ` Tino Calancha
2018-07-27 13:48 ` Toon Claes
0 siblings, 1 reply; 51+ messages in thread
From: Tino Calancha @ 2018-07-23 13:14 UTC (permalink / raw)
To: emacs-devel; +Cc: Ted Zlatanov
Ted Zlatanov <tzz@lifelogs.com> writes:
> On Wed, 26 Apr 2017 21:51:09 +0300 Eli Zaretskii <eliz@gnu.org> wrote:
>
>>> From: Ted Zlatanov <tzz@lifelogs.com>
>>> Date: Wed, 26 Apr 2017 14:45:24 -0400
>>>
>>> I've pushed a .gitlab-ci.yml file to the branch scratch/tzz/gitlab
>>>
>>> Merging that branch and letting me and Toon maintain that file would
>>> make it much easier to support https://gitlab.com/emacs-ci/emacs and
>>> further evaluate its viability as a CI service.
Sorry if I missed some announcement (I have not followed this thread
recently).
The last build of this CI was 2 months ago.
I see a message about some difficulties to update the
mirror (https://gitlab.com/emacs-ci/emacs)
Is there a plan to fix that problem and keep running this CI?
Tino
^ permalink raw reply [flat|nested] 51+ messages in thread
* Re: GitLab CI setup file in scratch/tzz/gitlab
2018-07-23 13:14 ` Tino Calancha
@ 2018-07-27 13:48 ` Toon Claes
0 siblings, 0 replies; 51+ messages in thread
From: Toon Claes @ 2018-07-27 13:48 UTC (permalink / raw)
To: Tino Calancha; +Cc: Ted Zlatanov, emacs-devel
[-- Attachment #1: Type: text/plain, Size: 1126 bytes --]
Tino Calancha <tino.calancha@gmail.com> writes:
> The last build of this CI was 2 months ago.
> I see a message about some difficulties to update the
> mirror (https://gitlab.com/emacs-ci/emacs)
>
> Is there a plan to fix that problem and keep running this CI?
That project was set up as a test, to see if GitLab CI would be
workable.
There was decided to set up GitLab on GNU/FSF servers, and start using
that instead.
We've been provisioned a machine by the FSF sysadmins, but we haven't
set up GitLab on it yet. We've started an Ansible playbook to set it up:
https://gitlab.com/emacs-ci/emba-ansible
But I've ran into a problem with the installation:
https://gitlab.com/emacs-ci/emba-ansible/merge_requests/1#note_84988269
I should pick this up again, and get it working. But I've been really
busy at my job, so it has low priority.
Also Ted has been really busy with other stuff if I'm not mistaken. And
for now we're the only two volunteers to maintain that instance.
But yeah, you're right. We should give this some attention, and make CI
for Emacs operational again.
-- Toon
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]
^ permalink raw reply [flat|nested] 51+ messages in thread
end of thread, other threads:[~2018-07-27 13:48 UTC | newest]
Thread overview: 51+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-04-26 18:45 GitLab CI setup file in scratch/tzz/gitlab Ted Zlatanov
2017-04-26 18:51 ` Eli Zaretskii
2017-04-26 20:55 ` dptd dptd
2017-04-27 5:01 ` Lars Brinkhoff
2017-04-27 14:23 ` Eli Zaretskii
2017-04-27 14:37 ` Dmitry Gutov
2017-04-27 14:59 ` Eli Zaretskii
2017-04-27 15:16 ` Dmitry Gutov
2017-04-28 21:33 ` Richard Stallman
2017-04-29 0:41 ` Ted Zlatanov
2017-04-29 15:58 ` Richard Stallman
2017-04-29 17:44 ` Dmitry Gutov
2017-04-30 22:19 ` Richard Stallman
2017-04-30 23:55 ` Dmitry Gutov
2017-05-01 1:53 ` Ted Zlatanov
2017-05-01 15:54 ` Richard Stallman
2017-05-02 12:42 ` Richard Stallman
2017-05-03 0:06 ` Dmitry Gutov
2017-05-03 19:15 ` Richard Stallman
2017-05-14 2:25 ` Ted Zlatanov
2017-05-15 1:44 ` Richard Stallman
2017-05-15 20:08 ` Ted Zlatanov
2017-05-18 1:12 ` Perry E. Metzger
2017-05-19 1:19 ` Richard Stallman
2017-05-19 1:53 ` Perry E. Metzger
2017-05-21 3:25 ` Richard Stallman
2017-05-21 3:49 ` Jean-Christophe Helary
2017-05-22 12:15 ` Richard Stallman
2017-05-21 15:19 ` Tino Calancha
2017-04-29 18:22 ` Lars Brinkhoff
2017-04-29 21:11 ` Ted Zlatanov
2017-04-30 22:20 ` Richard Stallman
2017-04-29 1:29 ` Mike Gerwitz
2017-04-27 17:43 ` Lars Brinkhoff
2017-04-27 20:16 ` Dmitry Gutov
2017-04-27 20:42 ` Eli Zaretskii
2017-04-27 23:20 ` Dmitry Gutov
2017-04-28 6:49 ` Eli Zaretskii
2017-04-28 8:33 ` Dmitry Gutov
2017-04-28 9:45 ` Eli Zaretskii
2017-04-28 9:58 ` Dmitry Gutov
2017-04-28 13:17 ` Eli Zaretskii
2017-04-28 13:35 ` Dmitry Gutov
2017-04-28 12:47 ` Ted Zlatanov
2017-04-28 13:24 ` Lars Brinkhoff
2017-04-28 14:16 ` Ted Zlatanov
2017-04-27 20:27 ` Eli Zaretskii
2017-04-28 21:34 ` Richard Stallman
2017-04-27 16:16 ` Ted Zlatanov
2018-07-23 13:14 ` Tino Calancha
2018-07-27 13:48 ` Toon Claes
Code repositories for project(s) associated with this external index
https://git.savannah.gnu.org/cgit/emacs.git
https://git.savannah.gnu.org/cgit/emacs/org-mode.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.