Info about available CVS/Arch/Git branches

Info about available CVS/Arch/Git branches

[Stefan Monnier]

We'd like to add to http://savannah.gnu.org/projects/emacs
some info about the available/relevant branches in the CVS and
Arch repositories.  Basically something like

  CVS
  ---
    trunk : branch of the next major revision, to become 23.X.
            Bidirectionally sync'd with the Gnus stable branch.
            Changes from EMACS_22_BASE get auto-added here as well.
    EMACS_22_BASE: branch where we prepare the next minor revisions 22.X.
    emacs-unicode-2: branch where the inernel text representation has been
                     changed to be based on utf-8.  Slated to be merged into
                     the trunk before 23.1.
                     Changes from the trunk get auto-added here as well.
    multi-tty: branch where we added support for multi-ttys (via emacsclient).
               Slated to be merged into the trunk before 23.1
               Changes from the trunk get auto-added here as well.
    emacs-bidi: branch for development of bidirectional support.
                Very preliminary.

  Arch
  ----
    emacs--devo--0: bidirectionally sync'd with CVS's trunk.
    emacs--unicode--0: bidirectionally sync'd with CVS's emacs-unicode-2.
    emacs--multi-tty--0: bidirectionally sync'd with CVS's multi-tty.
    emacs--lexbind--0: branch where we add support for lexical scoping.


I could see this kind of info added directly to
http://savannah.gnu.org/cvs/?group=emacs or a separate page (in which case
it'd make sense to have a single page for all backends rather than one per
backend).

It could be structured (with fileds like "branch name", "branch status",
"description", "slated for merge into", "parent branch", "sync'd with", ...)
or be just plain text.


        Stefan

Re: Info about available CVS/Arch/Git branches

[Reiner Steib]

On Thu, Aug 09 2007, Stefan Monnier wrote:

>   CVS
>   ---
>     trunk : branch of the next major revision, to become 23.X.
>             Bidirectionally sync'd with the Gnus stable branch.

While "Gnus stable branch" is true at the moment, our aim is to change
this to "Gnus trunk" soon.

I'm not sure if the information about Gnus is useful here in the first
place.  At least this line is more important for most Emacs user an
should precede the info on Gnus:

>             Changes from EMACS_22_BASE get auto-added here as well.

>     EMACS_22_BASE: branch where we prepare the next minor revisions 22.X.
>     emacs-unicode-2: branch where the inernel text representation has been

s/inernel/internal/?

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: Info about available CVS/Arch/Git branches

[Michael Olson]

[-- Attachment #1.1: Type: text/plain, Size: 804 bytes --]

Stefan Monnier <monnier@iro.umontreal.ca> writes:

> We'd like to add to http://savannah.gnu.org/projects/emacs
> some info about the available/relevant branches in the CVS and
> Arch repositories.  Basically something like
>
> [snip]

I'm intrigued by the mention of "git" in the subject line.  Do we have a
bidirectionally-synced git repo for Emacs yet?  I noticed
http://git.savannah.gnu.org/gitweb/?p=emacs.git;a=summary
but I wasn't sure if it was two-way, since there is no mention of it
on the Emacs Savannah project page.

-- 
       Michael Olson -- FSF Associate Member #652     |
 http://mwolson.org/ -- Jabber: mwolson_at_hcoop.net  |  /` |\ | | |
            Sysadmin -- Hobbies: Lisp, GP2X, HCoop    | |_] | \| |_|
Projects: Emacs, Muse, ERC, EMMS, ErBot, DVC, Planner |

[-- Attachment #1.2: Type: application/pgp-signature, Size: 188 bytes --]

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

Re: Info about available CVS/Arch/Git branches

[Andreas Schwab]

Michael Olson <mwolson@gnu.org> writes:

> I'm intrigued by the mention of "git" in the subject line.  Do we have a
> bidirectionally-synced git repo for Emacs yet?  I noticed
> http://git.savannah.gnu.org/gitweb/?p=emacs.git;a=summary
> but I wasn't sure if it was two-way, since there is no mention of it
> on the Emacs Savannah project page.

You'll have to ask Jim Meyering about it.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Info about available CVS/Arch/Git branches

[David Kastrup]

Andreas Schwab <schwab@suse.de> writes:

> Michael Olson <mwolson@gnu.org> writes:
>
>> I'm intrigued by the mention of "git" in the subject line.  Do we have a
>> bidirectionally-synced git repo for Emacs yet?  I noticed
>> http://git.savannah.gnu.org/gitweb/?p=emacs.git;a=summary
>> but I wasn't sure if it was two-way, since there is no mention of it
>> on the Emacs Savannah project page.
>
> You'll have to ask Jim Meyering about it.

It would appear that this repo is made from CVS.  If it is feasible,
it might me nicer to do this from the arch repo since that would
presumably keep track of the merges (information that is missing from
CVS).  I've had mediocre success using git archimport however.

Maybe it is also possible to extract just the merge info (which commit
was merged into which branch) in textual form from arch?  I could then
try my hand at figuring out a way to inject this info into the git
repository.

-- 
David Kastrup

Re: Info about available CVS/Arch/Git branches

[Jim Meyering]

David Kastrup <dak@gnu.org> wrote:
> Andreas Schwab <schwab@suse.de> writes:
>> Michael Olson <mwolson@gnu.org> writes:
>>> I'm intrigued by the mention of "git" in the subject line.  Do we have a
>>> bidirectionally-synced git repo for Emacs yet?  I noticed
>>> http://git.savannah.gnu.org/gitweb/?p=emacs.git;a=summary
>>> but I wasn't sure if it was two-way, since there is no mention of it
>>> on the Emacs Savannah project page.
>>
>> You'll have to ask Jim Meyering about it.
>
> It would appear that this repo is made from CVS.  If it is feasible,
> it might me nicer to do this from the arch repo since that would
> presumably keep track of the merges (information that is missing from
> CVS).  I've had mediocre success using git archimport however.
>
> Maybe it is also possible to extract just the merge info (which commit
> was merged into which branch) in textual form from arch?  I could then
> try my hand at figuring out a way to inject this info into the git
> repository.

You might want to look at how that is handled in the git <-> p4 mirroring
code in git's contrib/fast-import/*.

BTW, in the last week, I changed things so that branches in the git
repository are now pushed into the public emacs.git.  Before, I'd
been pushing only the trunk ("master").

Re: Info about available CVS/Arch/Git branches

[David Kastrup]

Jim Meyering <jim@meyering.net> writes:

> David Kastrup <dak@gnu.org> wrote:
>
> You might want to look at how that is handled in the git <-> p4
> mirroring code in git's contrib/fast-import/*.
>
> BTW, in the last week, I changed things so that branches in the git
> repository

Which repository are you talking about right now?

> are now pushed into the public emacs.git.  Before, I'd been pushing
> only the trunk ("master").q

I presume that emacs.git is just a mirror, and nothing gets committed
back into CVS either via Miles' arch or directly?

-- 
David Kastrup

Re: Info about available CVS/Arch/Git branches

[dhruva]

Hi,
 Well, with GIT, users on windows are left high and dry...
Fortunately, there is a mercurial repo of HEAD (I wish UNICODE branch
on mercurial was there too somewhere)...

-dhruva

-- 
Dhruva Krishnamurthy
Contents reflect my personal views only!

Re: Info about available CVS/Arch/Git branches

[David Kastrup]

dhruva <dhruvakm@gmail.com> writes:

> Hi,
>  Well, with GIT, users on windows are left high and dry...

<URL:http://code.google.com/p/msysgit/>

It is work in progress but purportedly becoming usable.  And I want to
stress that this project is mainly carried forward by volunteers that
don't use Windows themselves, detracting from other development
priorities.

One would expect that people who are crying for a _source_ _code_
_management_ system on their preferred platform would be able to
actually help with coding and testing.

-- 
David Kastrup

Re: Info about available CVS/Arch/Git branches

[Jim Meyering]

David Kastrup <dak@gnu.org> wrote:
> Jim Meyering <jim@meyering.net> writes:
>
>> David Kastrup <dak@gnu.org> wrote:
>>
>> You might want to look at how that is handled in the git <-> p4
>> mirroring code in git's contrib/fast-import/*.
>>
>> BTW, in the last week, I changed things so that branches in the git
>> repository
>
> Which repository are you talking about right now?

AFAIK, there is only one public git mirror of the emacs CVS repository:

  http://git.sv.gnu.org/gitweb/?p=emacs.git
  git-clone git://git.sv.gnu.org/emacs.git

>> are now pushed into the public emacs.git.  Before, I'd been pushing
>> only the trunk ("master").q
>
> I presume that emacs.git is just a mirror, and nothing gets committed
> back into CVS either via Miles' arch or directly?

Yep.  That's what "mirror" means to me, too.

However, if you make changes on a private git branch,
it's easy (git-cvsexportcommit) to mirror them back into a cvs commit.

Re: Info about available CVS/Arch/Git branches

[Andreas Schwab]

Jim Meyering <jim@meyering.net> writes:

> AFAIK, there is only one public git mirror of the emacs CVS repository:
>
>   http://git.sv.gnu.org/gitweb/?p=emacs.git
>   git-clone git://git.sv.gnu.org/emacs.git

There is a problem with this repo: for some time all commits have been
made to a subdirectory emacs.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Info about available CVS/Arch/Git branches

[Jim Meyering]

Andreas Schwab <schwab@suse.de> wrote:
> Jim Meyering <jim@meyering.net> writes:
>
>> AFAIK, there is only one public git mirror of the emacs CVS repository:
>>
>>   http://git.sv.gnu.org/gitweb/?p=emacs.git
>>   git-clone git://git.sv.gnu.org/emacs.git
>
> There is a problem with this repo: for some time all commits have been
> made to a subdirectory emacs.

Thanks, I know.
I'm working on it now.

Re: Info about available CVS/Arch/Git branches

[Randal L. Schwartz]

>>>>> "Jim" == Jim Meyering <jim@meyering.net> writes:

>> There is a problem with this repo: for some time all commits have been
>> made to a subdirectory emacs.

Jim> Thanks, I know.
Jim> I'm working on it now.

Weird.  That was also the breakage on the version I was tracking at repo.or.cz
(which I ensured was turned off because it was broken and the "maintainer" is
AWOL).  Were they both from the same source, or both broken in the same way
because of problems with cvs2git?

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn@stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!

Re: Info about available CVS/Arch/Git branches

[Andreas Schwab]

David Kastrup <dak@gnu.org> writes:

> It would appear that this repo is made from CVS.  If it is feasible,
> it might me nicer to do this from the arch repo since that would
> presumably keep track of the merges (information that is missing from
> CVS).  I've had mediocre success using git archimport however.

The arch repo is missing too much information to be useful for import.

> Maybe it is also possible to extract just the merge info (which commit
> was merged into which branch) in textual form from arch?

You can use the commit message to detect merges, but you'll have to hack
the import script to map the arch names back to the CVS branch names.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Info about available CVS/Arch/Git branches

[Jim Meyering]

merlyn@stonehenge.com (Randal L. Schwartz) wrote:
>>>>>> "Jim" == Jim Meyering <jim@meyering.net> writes:
>>> There is a problem with this repo: for some time all commits have been
>>> made to a subdirectory emacs.
>
> Jim> Thanks, I know.
> Jim> I'm working on it now.
>
> Weird.  That was also the breakage on the version I was tracking at repo.or.cz
> (which I ensured was turned off because it was broken and the "maintainer" is
> AWOL).  Were they both from the same source, or both broken in the same way
> because of problems with cvs2git?

I don't use cvs2git.  Do you?  Would you recommend it?
git-cvsimport is the basis of my simple scripts.

This may have happened because the repo had denynonfastforwards = false
and because my script tries to work as well with sourceforge-layout
CVS repositories, which lack the extra $module/$module hierarchy
found in the savannah CVS repositories.  An early version may have
failed to adapt and imported both emacs and emacs/emacs.

In any case, I've moved emacs.git aside and re-pushed into a clean
repo, and that has fixed it.

Also, I've found that after running git-cvsimport, sometimes there is no
"master" branch.  I've manually run the few required commands to do what
git-cvsimport should have done.  Haven't yet found time to investigate;
part of the problem is that importing emacs takes several hours, and
it's only at the end that git-cvsimport might be misbehaving.

Re: Info about available CVS/Arch/Git branches

[Jim Meyering]

Jim Meyering <jim@meyering.net> wrote:
> Also, I've found that after running git-cvsimport, sometimes there is no
> "master" branch.  I've manually run the few required commands to do what
> git-cvsimport should have done.  Haven't yet found time to investigate;
> part of the problem is that importing emacs takes several hours, and
> it's only at the end that git-cvsimport might be misbehaving.

I traced this part of the problem to the fact that I'm using an
old version of git-cvsimport (1.5.2.2).  That bug has been fixed
in newer versions.

Re: Info about available CVS/Arch/Git branches

[Andreas Schwab]

dhruva <dhruvakm@gmail.com> writes:

>  Well, with GIT, users on windows are left high and dry...

http://git.or.cz/gitwiki/WindowsInstall

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany
PGP key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Info about available CVS/Arch/Git branches

[dhruva]

Hi,

On 8/18/07, Andreas Schwab <schwab@suse.de> wrote:
> dhruva <dhruvakm@gmail.com> writes:
>
> >  Well, with GIT, users on windows are left high and dry...
>
> http://git.or.cz/gitwiki/WindowsInstall

I gave it a try. I executed a command "git config --global user.name
dhruva" and nothing really happened for almost 10mins. Git was
consuming 50% CPU constantly. Anyway, I will setup a virtual PC
environment with GNU/Linux running on my windows box just to use Git.
I am quite prejudiced when it comes to using CYGWIN. Thank you for the
link, I will give it a try again at a later point.

-dky

-- 
Dhruva Krishnamurthy
Contents reflect my personal views only!

Re: Info about available CVS/Arch/Git branches

[Miles Bader]

dhruva <dhruvakm@gmail.com> writes:
> I am quite prejudiced when it comes to using CYGWIN.

What does that mean??

-miles

-- 
"Though they may have different meanings, the cries of 'Yeeeee-haw!' and
 'Allahu akbar!' are, in spirit, not actually all that different."

Re: Info about available CVS/Arch/Git branches

[dhruva]

Hi,

On 8/20/07, Miles Bader <miles.bader@necel.com> wrote:
> dhruva <dhruvakm@gmail.com> writes:
> > I am quite prejudiced when it comes to using CYGWIN.
>
> What does that mean??

Having used MKS toolkit and MinGW, I personally feel they are much
faster as they do not need an emulation layer (based on limited usage
and knowledge). Even AT&T UWIN out performed other equivalents. The
fork implementation in UWIN was far superior (similar to MS-SFU).

-dky

-- 
Dhruva Krishnamurthy
Contents reflect my personal views only!