Discussion:
Magit release 1.1.0 and future plans
Yann Hodique
13 years ago
Permalink
Hi all,

it's been 9 months since the previous official announcement, and some of
our users are craving for a new release. Since I had a some spare time
recently, I took the liberty to wrap version 1.1.0 as a Christmas
present !

This version is available for download on the github page:
https://github.com/magit/magit/downloads

Also see http://magit.github.com/magit/magit.html for an online version
of the updated manual.

Although there is no huge feature or UI change in this release, it's
quite big already, with more than 126 bugfixes (only counting the
reported ones :)) and tons of cosmetic changes.

A brief list of noticeable changes since 1.0.0:

- [Pieter Praet] behavior of `magit-rewrite-start' has been slighty
changed, now including the base commit in the rewrite list for easier
manipulations in the "status" buffer. See variable
`magit-rewrite-inclusive' for customizing that behavior.

- [Ramkumar Ramachandra] for people that don't like the 1.0.0 approach
of having submenus for advanced options access, a magit-simple-keys.el
is provided in the contrib/ directory. This is not part of the
official magit distribution, but some might find it a lifesaver.

- [Hannu Koivisto] as per popular request, the window management feature
of Magit are now a bit more flexible (although still not perfect). See
variable `magit-status-buffer-switch-function'

- [Lluís Vilanova] a new extension is available for interacting
with stgit.

- [Peter J. Weisberg] former Magit "submodes" are now major modes
deriving from `magit-mode'. See for example `magit-status-mode',
`magit-log-mode'...
This is conceptually much cleaner, but might require some adjustments
for existing customization code.

- [Yann Hodique] shipped extensions are now implemented as minor modes,
which means they can easily be toggled (manually or automatically) for
a given repository. Again, this is conceptually cleanre but will
require some adjustments in user configuration, as `require'-ing them
is no longer sufficient. See user manual of details.

- ... and of course a lot more, by many contributors. Many thanks to you
all !



Now, about the future.

Looking back, many (many !) of the fixes that are released today should
have been part of a much earlier maintenance release. Actually I've
even been willing to do that, but the state of "master" didn't make it
easy to pick the relevant fixes.

So I'd like to take the opportunity to propose that we adopt a new
branching model that would allow us to release subminor versions
more often.

Since I hate reinventing the wheel, I propose we mainly stick to the Git
way of maintaining software
http://repo.or.cz/w/git.git?a=blob;f=Documentation/howto/maintain-git.txt
with the exception of the "pu" branch, which I think we don't need at
this point (and anyway branch resets would probably not work very well
with a group of maintainers). To paraphrase that document:

- 'master' branch is used to prepare for the next feature release. In
other words, at some point, the tip of 'master' branch is tagged with
vX.Y.0.

- 'maint' branch is used to prepare for the next maintenance release.
After the feature release vX.Y.0 is made, the tip of 'maint' branch is
set to that release, and bugfixes will accumulate on the branch, and
at some point, the tip of the branch is tagged with vX.Y.1, vX.Y.2,
and so on.

- 'next' branch is used to publish changes (both enhancements and fixes)
that (1) have worthwhile goal, (2) are in a fairly good shape suitable
for everyday use, (3) but have not yet demonstrated to be regression
free. New changes are tested in 'next' before merged to 'master'.

The benefit is double: first it allows disruptive changes to be staged
outside 'master' until they're proven ready, and then 'maint' would be
a good place for many bugfixes, allowing for shorter release cycle.

Any comments welcome.



On another note, I'm starting having trouble living without a proper
test suite for Magit. As we get bigger and more complex, it doesn't seem
reasonable to continue testing by hand (especially since everyone of us
has its own configuration and subset of use cases he's using all the
time. Thinking of extensions, which are still quite easy to break
inadvertently...)

I've started some very preliminary work in branch 'unit-tests'
(leveraging the ERT library that's now shipping with Emacs), and expect
to come up with something decent by the time 1.2 is out...

Speaking of that, I'd really like to put in place a Continuous
Integration solution for Magit. I started looking at
http://continuous.io but suggestions are highly welcome. Ideally we
should be able to validate Magit with different combinations of Emacs
and Git, and that definitely requires some automation.



I guess that will be all for now. Thanks for reading this far, Merry
Xmas, Happy New Year, and have fun using (Ma)git !



Thanks,

Yann.
Timo Juhani Lindfors
13 years ago
Permalink
Post by Yann Hodique
it's been 9 months since the previous official announcement, and some of
our users are craving for a new release. Since I had a some spare time
recently, I took the liberty to wrap version 1.1.0 as a Christmas
present !
Thank you! If no major issues are found I'll upload this new version to
Debian in a week or two.
Post by Yann Hodique
- [Ramkumar Ramachandra] for people that don't like the 1.0.0 approach
of having submenus for advanced options access, a magit-simple-keys.el
is provided in the contrib/ directory. This is not part of the
official magit distribution, but some might find it a lifesaver.
So should I include this in the Debian package or not?

-Timo
Yann Hodique
13 years ago
Permalink
On Sat, Dec 24, 2011 at 12:46 PM, Timo Juhani Lindfors
Post by Timo Juhani Lindfors
Post by Yann Hodique
it's been 9 months since the previous official announcement, and some of
our users are craving for a new release. Since I had a some spare time
recently, I took the liberty to wrap version 1.1.0 as a Christmas
present !
Thank you! If no major issues are found I'll upload this new version to
Debian in a week or two.
Great, thanks !
Post by Timo Juhani Lindfors
Post by Yann Hodique
- [Ramkumar Ramachandra] for people that don't like the 1.0.0 approach
  of having submenus for advanced options access, a magit-simple-keys.el
  is provided in the contrib/ directory. This is not part of the
  official magit distribution, but some might find it a lifesaver.
So should I include this in the Debian package or not?
I'm not sure what the usual debian way is in such situations. For
sure, some people will appreciate it, so I guess it should be made
available somehow. But once installed, magit behavior is significantly
different from the one described in the manual, so I guess it should
probably not be in /usr/share/emacs/site-lisp/magit (or more generally
it should not be in the default load-path).
Also, in a later version I'd like to render it obsolete by providing
something equivalent (but better integrated) in magit itself.

Maybe somewhere in /usr/share/doc/magit or a similar place ?

Yann.
Ramkumar Ramachandra
13 years ago
Permalink
Hi,
...
Hm. It's not auto-loaded so users will have to "(require
'magit-simple-keys)" in their .emacs explicitly to use it anyway.
Sure, putting it a place like /usr/share/doc/magit would work, but I'm
wondering if we should push it more aggressively to get feedback.
Then again, I'm biased as the author: I don't know how many people
actually (would like to) use it.

Maintainability shouldn't be a major concern: it's just a few lines of
code that won't break unless magit changes significantly.
Post by Yann Hodique
Also, in a later version I'd like to render it obsolete by providing
something equivalent (but better integrated) in magit itself.
Mainly, I don't want magit (or atleast my local copy) to deviate from
the command-line interface too much: I have to use the command-line
interface most of the time as a git contributor, and I see magit as a
nice shortcut for common tasks.

Thanks.

-- Ram
Yann Hodique
13 years ago
Permalink
Post by Ramkumar Ramachandra
Hm. It's not auto-loaded so users will have to "(require
'magit-simple-keys)" in their .emacs explicitly to use it anyway.
Sure, putting it a place like /usr/share/doc/magit would work, but I'm
wondering if we should push it more aggressively to get feedback.
Then again, I'm biased as the author: I don't know how many people
actually (would like to) use it.
Sure, my point was just about the fact that we don't document it
anywhere, hence once it's loaded (which users might be more tempted to
do unknowingly if it has the same visibility as the rest) the manual
isn't so helpful anymore.
So like I said, it's more of a policy thing. Personally I don't have any
strong feeling.
Post by Ramkumar Ramachandra
Mainly, I don't want magit (or atleast my local copy) to deviate from
the command-line interface too much: I have to use the command-line
interface most of the time as a git contributor, and I see magit as a
nice shortcut for common tasks.
Yep, I tend to agree with that, and sometimes I wish there was no
annoying popup coming in the way (especially when it breaks my
layout). But on the other hand, I find the magit-simple-keys.el approach
a bit too radical :)
While I do simple things most of the time (and would enjoy short
bindings then), it's good to be able to do complex stuff from magit as
well, from time to time.
I think there could be 2 modes: simple and complex (and in complex mode,
we could probably even make the popup optional). And one could probably
toggle the mode, maybe with a C-- prefix, or something like this.

It's not very clear in my head yet, but I'm quite sure we can improve
the current situation to satisfy everybody :)

Yann.
--
The individual is the key, the final effective unit of all biological processes.

-- PARDOT KYNES
Ramkumar Ramachandra
13 years ago
Permalink
Hi Yann,
...
Yeah, me too. Let the Debian policy people decide what to do :)
Post by Yann Hodique
While I do simple things most of the time (and would enjoy short
bindings then), it's good to be able to do complex stuff from magit as
well, from time to time.
Agreed: I just think of magit-simple-keys as a stopgap: hopefully, we
won't get too comfortable/ lazy to write a new interface.
Post by Yann Hodique
I think there could be 2 modes: simple and complex (and in complex mode,
we could probably even make the popup optional). And one could probably
toggle the mode, maybe with a C-- prefix, or something like this.
It's not very clear in my head yet, but I'm quite sure we can improve
the current situation to satisfy everybody :)
I was hoping to make it some kind of clean fallback from one mode to
the other instead of two separate user interfaces. Gah, we'll talk
more when we have some more code demonstrating this :)

-- Ram
Timo Juhani Lindfors
13 years ago
Permalink
Post by Ramkumar Ramachandra
Yeah, me too. Let the Debian policy people decide what to do :)
magit 1.1.0-1 is now in debian unstable. I did not call "make
install_contrib" there mostly because I have not tested
magit-simple-keys.el or magit-classic-theme.el. This means I did not
feel very comfortable at writing documentation on how to enable these
either at this time.
Ramkumar Ramachandra
13 years ago
Permalink
Hi Timo,
Post by Timo Juhani Lindfors
Yeah, me too.  Let the Debian policy people decide what to do :)
magit 1.1.0-1 is now in debian unstable. I did not call "make
install_contrib" there mostly because I have not tested
magit-simple-keys.el or magit-classic-theme.el. This means I did not
feel very comfortable at writing documentation on how to enable these
either at this time.
That's alright; I'll extend it a bit and write some documentation
before the next release.

Thanks.

-- Ram
Rémi Vanicat
13 years ago
Permalink
Post by Timo Juhani Lindfors
Post by Ramkumar Ramachandra
Yeah, me too. Let the Debian policy people decide what to do :)
magit 1.1.0-1 is now in debian unstable. I did not call "make
install_contrib" there mostly because I have not tested
magit-simple-keys.el or magit-classic-theme.el. This means I did not
feel very comfortable at writing documentation on how to enable these
either at this time.
From faa7463500945b61fafc60f11fa12eb8c5e9c79a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Vanicat?= <vanicat-8fiUuRrzOP0dnm+***@public.gmane.org>
Date: Mon, 9 Jan 2012 17:42:36 +0100
Subject: [PATCH] Install the contribs in the Debian package.

---
debian/README.Debian | 26 ++++++++++++++++++++++++++
debian/changelog | 6 ++++++
debian/magit.emacsen-install | 2 +-
debian/rules | 1 +
4 files changed, 34 insertions(+), 1 deletions(-)

diff --git a/debian/README.Debian b/debian/README.Debian
index 039b898..432c731 100644
--- a/debian/README.Debian
+++ b/debian/README.Debian
@@ -1,6 +1,31 @@
magit for Debian
================

+Using magit magit simple keys
+-----------------------------
+
+Older version of magit used a simpler mapping for key, that didn't
+give access to all magit capabilities. You can reenable those simpler
+binding by adding
+
+ (require 'magit-simple-keys)
+
+to your .emacs.
+
+ -- Rémi Vanicat <vanicat-8fiUuRrzOP0dnm+***@public.gmane.org>, Mon, 9 Jan 2012 17:40:37 +0100
+
+Using magit classic theme
+-------------------------
+
+Magit comme with a color theme that try to make magit look like older
+version of magit. You can load it by using
+
+ (load-theme 'magit-classic)
+
+to your .emacs
+
+ -- Rémi Vanicat <vanicat-8fiUuRrzOP0dnm+***@public.gmane.org>, Mon, 9 Jan 2012 17:37:08 +0100
+
Colors used by magit are invisible?
-----------------------------------

@@ -14,3 +39,4 @@ to ~/.emacs or by using M-x customize-variable RET
frame-background-mode RET.

-- Timo Juhani Lindfors <timo.lindfors-***@public.gmane.org>, Fri, 25 Feb 2011 19:05:11 +0200
+
diff --git a/debian/changelog b/debian/changelog
index 3606e00..0edf9ad 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,9 @@
+magit (1.1.0-2) unstable; urgency=low
+
+ * Install contrib
+
+ -- Rémi Vanicat <vanicat-8fiUuRrzOP0dnm+***@public.gmane.org> Mon, 09 Jan 2012 17:29:56 +0100
+
magit (1.1.0-1) unstable; urgency=low

* New upstream release.
diff --git a/debian/magit.emacsen-install b/debian/magit.emacsen-install
index 94b65bf..9c832d9 100644
--- a/debian/magit.emacsen-install
+++ b/debian/magit.emacsen-install
@@ -6,7 +6,7 @@ PACKAGE=magit
FLAVOR=$1

byte_compile_options="-batch -f batch-byte-compile"
-el_files="magit-topgit.el magit.el magit-key-mode.el magit-svn.el magit-bisect.el magit-stgit.el"
+el_files="magit-topgit.el magit.el magit-key-mode.el magit-svn.el magit-bisect.el magit-stgit.el magit-classic-theme.el magit-simple-keys.el"
el_dir=/usr/share/emacs/site-lisp/${PACKAGE}/
elc_dir=/usr/share/${FLAVOR}/site-lisp/${PACKAGE}/

diff --git a/debian/rules b/debian/rules
index e4ab221..4890ec2 100755
--- a/debian/rules
+++ b/debian/rules
@@ -4,6 +4,7 @@
dh $@

override_dh_auto_install:
+ make install_contrib DESTDIR=$(CURDIR)/debian/magit PREFIX=/usr
dh_auto_install -- PREFIX=/usr
mkdir -p $(CURDIR)/debian/magit/usr/share/emacs/site-lisp/magit
mv $(CURDIR)/debian/magit/usr/share/emacs/site-lisp/*.el $(CURDIR)/debian/magit/usr/share/emacs/site-lisp/magit
--
1.7.8.2
Timo Juhani Lindfors
13 years ago
Permalink
Hi,
Idea sounds fine, thanks.
Post by Rémi Vanicat
+Using magit classic theme
+-------------------------
+
+Magit comme with a color theme that try to make magit look like older
s/comme/comes/
Post by Rémi Vanicat
+version of magit. You can load it by using
+
+ (load-theme 'magit-classic)
+
+to your .emacs
s/using/adding/ or s/to/in/?
Post by Rémi Vanicat
+
How about having just one date in the bottom instead of one per each
entry? Usually people just want to see if anything in the document has
changed since they last read it so they expect one global datestamp.
Post by Rémi Vanicat
+ make install_contrib DESTDIR=$(CURDIR)/debian/magit PREFIX=/usr
Should this use $(MAKE)?

-Timo
Rémi Vanicat
13 years ago
Permalink
Timo Juhani Lindfors <timo.lindfors-***@public.gmane.org> writes:

[...]
Post by Timo Juhani Lindfors
s/comme/comes/
[...]
Post by Timo Juhani Lindfors
s/using/adding/ or s/to/in/?
probably, I'm no native speaker, so...
Post by Timo Juhani Lindfors
Post by Rémi Vanicat
+
How about having just one date in the bottom instead of one per each
entry? Usually people just want to see if anything in the document has
changed since they last read it so they expect one global datestamp.
I didn't wanted to sign all the text under my name, but you are probably
correct.
Post by Timo Juhani Lindfors
Post by Rémi Vanicat
+ make install_contrib DESTDIR=$(CURDIR)/debian/magit PREFIX=/usr
Should this use $(MAKE)?
my mistake.
--
Rémi Vanicat
Timo Juhani Lindfors
13 years ago
Permalink
Hi,
Post by Rémi Vanicat
+ (load-theme 'magit-classic)
I added your patch (with some modifications) to git and built
preliminary packages:

http://anonscm.debian.org/gitweb/?p=collab-maint/magit.git

http://lindi.iki.fi/lindi/magit/debian/

When I tested the package I noticed a small problem:

1) If I put the suggested "(load-theme 'magit-classic)" to ~/.emacs then
I can only use emacs on machines where magit is installed! My $HOME is
shared with NFS so ~/.emacs is visible on many machines where I do not
have magit 1.1.0 or later installed. The error I get is

Warning (initialization): An error occurred while loading
`/home/lindi/.emacs':

File error: Cannot open load file, magit-classic-theme

To ensure normal operation, you should investigate and remove the
cause of the error in your initialization file. Start Emacs with
the `--debug-init' option to view a complete error backtrace.


Any suggested fixes?

-Timo
Rémi Vanicat
13 years ago
Permalink
...
obviously. The same would be true if you added anything that call magit
in your .emacs.
Post by Timo Juhani Lindfors
Warning (initialization): An error occurred while loading
File error: Cannot open load file, magit-classic-theme
(condition-case
(load-theme 'magit-classic)
(error nil))

is a not very clean way to fix your problem.
--
Rémi Vanicat
Moritz Bunkus
13 years ago
Permalink
Hey,

what about something like

(eval-after-load 'magit
'(progn
(load-theme 'magit-classic)))

Kind regards,
mo
Timo Juhani Lindfors
13 years ago
Permalink
Post by Moritz Bunkus
what about something like
(eval-after-load 'magit
'(progn
(load-theme 'magit-classic)))
Sounds better. However, this does not permit me to use magit in debian
stable anymore: M-x magit-status RET prints just

Cannot open load file: magit-classic-theme

and won't let me continue. I regularly use both stable and unstable
installations so it'd be nice to have README.Debian document a way to
use the "new" classic theme in unstable but also keep stable working.
Rémi Vanicat
13 years ago
Permalink
Post by Timo Juhani Lindfors
Post by Moritz Bunkus
what about something like
(eval-after-load 'magit
'(progn
(load-theme 'magit-classic)))
Sounds better. However, this does not permit me to use magit in debian
stable anymore: M-x magit-status RET prints just
Cannot open load file: magit-classic-theme
and won't let me continue. I regularly use both stable and unstable
installations so it'd be nice to have README.Debian document a way to
use the "new" classic theme in unstable but also keep stable working.
I don't believe those information belong to Debian Readme: unstable (and
testing) are still mostly tool to build the next stable, and such
documentation is not useful for stable user (and might become stall
before new magit's package reach stable).
--
Rémi Vanicat
Timo Juhani Lindfors
13 years ago
Permalink
Post by Rémi Vanicat
I don't believe those information belong to Debian Readme: unstable (and
testing) are still mostly tool to build the next stable, and such
Sure but I'd hate to have wheezy README.Debian suggest things that will
break if people still have some squeeze machines in use. Many people
share their ~/.emacs to all their computers.
Pieter Praet
13 years ago
Permalink
Post by Timo Juhani Lindfors
Post by Rémi Vanicat
I don't believe those information belong to Debian Readme: unstable (and
testing) are still mostly tool to build the next stable, and such
Sure but I'd hate to have wheezy README.Debian suggest things that will
break if people still have some squeeze machines in use. Many people
share their ~/.emacs to all their computers.
#+begin_src emacs-lisp
(eval-after-load 'magit
'(progn
(ignore-errors
(load-theme 'magit-classic))))
#+end_src



Peace
--
Pieter
Rémi Vanicat
13 years ago
Permalink
Yann Hodique <yann.hodique-***@public.gmane.org> writes:

[...]
Post by Yann Hodique
I'm not sure what the usual debian way is in such situations. For
sure, some people will appreciate it, so I guess it should be made
available somehow. But once installed, magit behavior is significantly
different from the one described in the manual, so I guess it should
probably not be in /usr/share/emacs/site-lisp/magit (or more generally
it should not be in the default load-path).
It's not a problem if it's in the load path if it's not loaded. At the
opposite, not putting it there cause more difficulties than needed for
those wanted to use it.

[...]
Post by Yann Hodique
Maybe somewhere in /usr/share/doc/magit or a similar place ?
You could just put there the description on how to activate it (in the
README.Debian).
--
Rémi Vanicat
Samuel Wales
13 years ago
Permalink
Great work.

So maint is a little like Debian stable and master is a little like unstable?

I get this in my status buffer:

Unpushed commits:
b70d6216 * Merge branch 'master' of git://github.com/magit/magit
e7c36aaf * Merge branch 'master' of git://github.com/magit/magit
c1c378f3 * Merge branch 'master' of git://github.com/magit/magit
151f586a * Merge branch 'master' of git://github.com/magit/magit
19fdad0a * Merge branch 'master' of git://github.com/magit/magit
33407d53 * Merge branch 'master' of git://github.com/magit/magit
69759253 * Merge branch 'master' of git://github.com/magit/magit
37a62291 * Merge branch 'master' of git://github.com/magit/magit
a09651f9 * Merge branch 'master' of git://github.com/magit/magit
5dae7cf6 * Merge branch 'master' of git://github.com/magit/magit
c5d965df * Merge branch 'master' of git://github.com/magit/magit
7e987435 * Merge branch 'master' of git://github.com/magit/magit
6854b6f8 * Merge branch 'master' of git://github.com/magit/magit
f0f8a443 * Merge branch 'master' of git://github.com/magit/magit
713f1856 * Merge branch 'master' of git://github.com/magit/magit
2d9cdfc1 * Merge branch 'master' of git://github.com/magit/magit
0961735f * Merge branch 'master' of git://github.com/magit/magit
c5e1644d * Merge branch 'master' of git://github.com/magit/magit
3e555a7d * Merge branch 'master' of git://github.com/magit/magit
8907f7d8 * Merge branch 'master' of git://github.com/magit/magit
da8184de * Merge branch 'master' of git://github.com/magit/magit
9bdb0ecb * Merge branch 'master' of git://github.com/magit/magit
5535baaf * Merge branch 'master' of git://github.com/magit/magit
289aac6b * Merge branch 'master' of git://github.com/magit/magit
41f22008 * Merge branch 'master' of git://github.com/magit/magit
57d5a1e2 * Merge branch 'master' of git://github.com/magit/magit
d45f04c2 * Merge branch 'master' of git://github.com/magit/magit
83744599 * Merge branch 'master' of git://github.com/magit/magit
3b03315e * Merge branch 'master' of git://github.com/magit/magit
51aba58a * Merge branch 'master' of git://github.com/magit/magit
143b07b3 * Merge branch 'master' of git://github.com/magit/magit
1f61f329 * Merge branch 'master' of git://github.com/magit/magit
5382725d * Don't try to call magit-refresh-all when the defcustoms are
first being initialized

Samuel
--
The Kafka Pandemic: http://thekafkapandemic.blogspot.com
===
Bigotry against people with serious diseases is still bigotry.
Yann Hodique
13 years ago
Permalink
Post by Samuel Wales
Great work.
So maint is a little like Debian stable and master is a little like unstable?
I guess master would be more like testing, and next would be unstable? :)

Yann.
--
"Once more the drama begins."

-- The Emperor Paul Muad'dib on his ascension to the Lion Throne
theturingmachine
13 years ago
Permalink
For the CI, have you considered using travis-ci.org? It's very nice
and integrates with github.
...
Yann Hodique
13 years ago
Permalink
Post by theturingmachine
For the CI, have you considered using travis-ci.org? It's very nice
and integrates with github.
I didn't know about it, thanks. I just had a quick look, do you know if
it's flexible enough for emacs/magit requirements? They seem to be
focusing on some specific use cases, and at least don't emphasize custom
builders/runners.

Thanks,

Yann.
--
We tend to become like the worst in those we oppose.

-- Bene Gesserit Coda
Eli Barzilay
13 years ago
Permalink
Post by Yann Hodique
- 'next' branch is used to publish changes (both enhancements and
fixes) that (1) have worthwhile goal, (2) are in a fairly good
shape suitable for everyday use, (3) but have not yet demonstrated
to be regression free. New changes are tested in 'next' before
merged to 'master'.
Would this be the branch to submit requests against?
--
((lambda (x) (x x)) (lambda (x) (x x))) Eli Barzilay:
http://barzilay.org/ Maze is Life!
Yann Hodique
13 years ago
Permalink
Post by Eli Barzilay
Post by Yann Hodique
- 'next' branch is used to publish changes (both enhancements and
fixes) that (1) have worthwhile goal, (2) are in a fairly good
shape suitable for everyday use, (3) but have not yet demonstrated
to be regression free. New changes are tested in 'next' before
merged to 'master'.
Would this be the branch to submit requests against?
No, 'next' will just be an integration branch for topics that will live
for a while in their own topic branch anyway, so that probably nothing
will ever be merged directly into 'next'.

Requests should be submitted against 'master' (or 'maint' if the purpose
of the patch clearly belongs there), and they'll be re-routed correctly
(or merged directly if obviously correct).

Yann.
--
Only fools leave witnesses.

-- HASIMIR FENRING
Leslie Watter
13 years ago
Permalink
For continuous integration I suggest use of Jenkins (http://jenkins-ci.org )
which is pretty configurable and supports many custom builds.

My 2c. Cheers,

LEslie
...
--
Leslie H. Watter
Pieter Praet
13 years ago
Permalink
Post by Yann Hodique
[...]
- [Pieter Praet] behavior of `magit-rewrite-start' has been slighty
changed, now including the base commit in the rewrite list for easier
manipulations in the "status" buffer. See variable
`magit-rewrite-inclusive' for customizing that behavior.
[...]
Well, no, I actually made it *more* difficult to start rewrites
from `magit-status' by replacing the original behaviour of:

"start rewrite at REV~1"

with the way `git rebase' behaves:

"start rewrite at REV"

... which has since been replaced with a defcustom called
`magit-rewrite-inclusive' (default = original Magit behaviour),
allowing to always use either the original or the rebase-like
behaviour, or be asked whether to include the base commit every
time `magit-rewrite-start' is invoked.
Post by Yann Hodique
[...]
- ... and of course a lot more, by many contributors. Many thanks to you
all !
[...]
I second that, and many thanks to you as well, Yann!
...
+1; Every self-respecting project should follow this workflow IMO !
...
That's fantastic news!
Post by Yann Hodique
Speaking of that, I'd really like to put in place a Continuous
Integration solution for Magit. I started looking at
http://continuous.io but suggestions are highly welcome. Ideally we
should be able to validate Magit with different combinations of Emacs
and Git, and that definitely requires some automation.
I guess that will be all for now. Thanks for reading this far, Merry
Xmas, Happy New Year, and have fun using (Ma)git !
Thanks,
Yann.
Peace
--
Pieter
Loading...