[PATCH 0/2] Series short description

[PATCH 0/2] Series short description

[Yann Dirson]

The following series allow to pick from all over the tree, even if your tree
is full of merge commits, and if you use grafts extensively.

The series is not really complete, it lacks a fix for a nasty bug that
crashes stgit when "pick" encounters a file-creation conflict.  But
nevertheless, I would not have been able to do my job today without them :)
-- 
Yann Dirson    <ydirson@altern.org> |
Debian-related: <dirson@debian.org> |   Support Debian GNU/Linux:
                                    |  Freedom, Power, Stability, Gratis
     http://ydirson.free.fr/        | Check <http://www.debian.org/>

[PATCH 1/2] Add a --parent flag to "stgit pick".

[Yann Dirson]

This is useful to pick changes from a non-stgit branch, that were
recorded as a merge commit, while giving control on the parent to
use as "bottom" for the patch (for some reason stgit selects the
first parent by default).

Incidentally, it can be used to pick changes between arbitrary
trees in the revision graph.  Since that can be useful as well,
I did not implement the is_parent check I originally meant to add.

Signed-off-by: Yann Dirson <ydirson@altern.org>
---

 stgit/commands/pick.py |   11 +++++++++--
 1 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/stgit/commands/pick.py b/stgit/commands/pick.py
index 1aa83d0..2916b6e 100644
--- a/stgit/commands/pick.py
+++ b/stgit/commands/pick.py
@@ -37,6 +37,8 @@ options = [make_option('-n', '--name',
            make_option('-r', '--reverse',
                        help = 'reverse the commit object before importing',
                        action = 'store_true'),
+           make_option('-p', '--parent',
+                       help = 'use COMMITID as parent'),
            make_option('--fold',
                        help = 'fold the commit object into the current patch',
                        action = 'store_true'),
@@ -73,12 +75,17 @@ def func(parser, options, args):
             if not patch:
                 raise CmdException, 'Unknown patch name'
 
+    if options.parent:
+        parent = options.parent
+    else:
+        parent = commit.get_parent()
+
     if not options.reverse:
-        bottom = commit.get_parent()
+        bottom = parent
         top = commit_id
     else:
         bottom = commit_id
-        top = commit.get_parent()
+        top = parent
 
     if options.fold:
         print 'Folding commit %s...' % commit_id,

[PATCH 2/2] Look for a commit's parents in the standard way.

[Yann Dirson]

This has the direct effect of taking info/grafts into account, since
ignoring it only causes confusion.

My original implementation was pasted from the same fix applied to
cogito some time ago.  That one is hopefully more pythonic, but it
looks like split() is deprecated for some reason, and I don't know
what should be used instead.

Signed-off-by: Yann Dirson <ydirson@altern.org>
---

 stgit/git.py |   10 ++++++----
 1 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/stgit/git.py b/stgit/git.py
index c8b7b8f..68b547c 100644
--- a/stgit/git.py
+++ b/stgit/git.py
@@ -19,6 +19,7 @@ Foundation, Inc., 59 Temple Place, Suite
 """
 
 import sys, os, popen2, re, gitmergeonefile
+from string import split
 
 from stgit import basedir
 from stgit.utils import *
@@ -47,12 +48,13 @@ class Commit:
             field = line.strip().split(' ', 1)
             if field[0] == 'tree':
                 self.__tree = field[1]
-            elif field[0] == 'parent':
-                self.__parents.append(field[1])
-            if field[0] == 'author':
+            #elif field[0] == 'parent':
+                #self.__parents.append(field[1])
+            elif field[0] == 'author':
                 self.__author = field[1]
-            if field[0] == 'committer':
+            elif field[0] == 'committer':
                 self.__committer = field[1]
+        self.__parents = split(_output_lines('git-rev-list --parents --max-count=1 %s' % id_hash)[0])[1:]
         self.__log = ''.join(lines[i+1:])
 
     def get_id_hash(self):

Re: [PATCH 2/2] Look for a commit's parents in the standard way.

[Catalin Marinas]

On 14/08/06, Yann Dirson <ydirson@altern.org> wrote:
> This has the direct effect of taking info/grafts into account, since
> ignoring it only causes confusion.

I don't know the difference but is there any between parsing the
commit file and using git-rev-list --parents?

> +        self.__parents = split(_output_lines('git-rev-list --parents --max-count=1 %s' % id_hash)[0])[1:]

Instead of using the split() method, you could call
_output_lines('git-rev-list --parents --max-count=1 %s' %
id_hash)[0].split()[1:]. Maybe that's why they might deprecate the
global split method.

Setting self.__parents by calling get-rev-list would have a
performance impact on the push operation. I think we could remove the
__parents variable and only call git-rev-parse in get_parents() or
get_parent().

-- 
Catalin

Re: [PATCH 2/2] Look for a commit's parents in the standard way.

[Junio C Hamano]

"Catalin Marinas" <catalin.marinas@gmail.com> writes:

> On 14/08/06, Yann Dirson <ydirson@altern.org> wrote:
>> This has the direct effect of taking info/grafts into account, since
>> ignoring it only causes confusion.
>
> I don't know the difference but is there any between parsing the
> commit file and using git-rev-list --parents?

If you mean parsing "cat-file commit" output vs "rev-list
--parents" output, the former shows the true parents while the
latter shows a list of parents adjusted with information in the
grafts file.

If you are using pathspec then the latter further simplifies the
parent list, but you know that already and you are not using
that form here.

Re: [PATCH 2/2] Look for a commit's parents in the standard way.

[Yann Dirson]

On Sun, Aug 20, 2006 at 11:43:22AM +0100, Catalin Marinas wrote:
> On 14/08/06, Yann Dirson <ydirson@altern.org> wrote:
> >This has the direct effect of taking info/grafts into account, since
> >ignoring it only causes confusion.
> 
> I don't know the difference but is there any between parsing the
> commit file and using git-rev-list --parents?

Yes, git-rev-list at least takes info/grafts into account.

> >+        self.__parents = split(_output_lines('git-rev-list --parents 
> >--max-count=1 %s' % id_hash)[0])[1:]
> 
> Instead of using the split() method, you could call
> _output_lines('git-rev-list --parents --max-count=1 %s' %
> id_hash)[0].split()[1:]. Maybe that's why they might deprecate the
> global split method.

Ah, OK.

Best regards,
-- 
Yann.

-- 
VGER BF report: U 0.500235

[EGIT PATCH 00/10] Fixes and improvement on Eclipse plugin

[Robin Rosenberg]

Here is a mix of bug fixes and performance improvements. The most
important ones are that the history panel now shows the right
history (yea, I'm an SWT/JFace n00b so there may be more suble bugs
there). Actually, now that I come to think of it; the rows in the
history window sometimes get the wrong height. The workaround is
to press ALT-F4 and then cancel. 

The other big thing is that Eclipse now caches commits so browsing
the resorces updates the history view immediately instead of an
annoying second or two later, provided the Link with Editor button
is pushed.

As for JDK, I recommend JDK 1.5.11 or JDK 1.6 for running because
of improvements and bug fixes made by Sun wrt GC and memory mapping.

-- robin

[PATCH 0/2] Series short description

[Robin Rosenberg]

The following series implements...

-- 
Signature

Re: [PATCH 0/2] Series short description

[Shawn O. Pearce]

Robin Rosenberg <robin.rosenberg@dewire.com> wrote:
> The following series implements...

Exciting new features?  ;-)
 
-- 
Shawn.

Re: [PATCH 0/2] Series short description

[Robin Rosenberg]

söndag 11 mars 2007 21:19 skrev Shawn O. Pearce:
> Robin Rosenberg <robin.rosenberg@dewire.com> wrote:
> > The following series implements...
> 
> Exciting new features?  ;-)

Indeed.  I thought it'd be better if you read the patch rather
than a misleading subject. :)

Actually, I exited vi with :q! and stacked git didn't realize this. :( 

-- robin