Mark Phippard wrote:
> On Mon, May 19, 2008 at 4:55 PM, Karl Fogel <kfogel_at_red-bean.com> wrote:
>> "Mark Phippard" <markphip_at_gmail.com> writes:
> Then I guess I'd be in favor of releasing 1.5.0 based off RC5 and get
> the 1.5.1 queued up to be ready to go shortly thereafter (say 3-4
> weeks later).
I don't see what the large rush is to get it out. I would rather not put out a
release that distributers are just going to put in patches that will be in
1.5.1. Also, I would not want to see a release based off rc5.
These are the changes that were merged in that I would like to see in a release.
Some are pretty minor. The most important ones for me are the improvements to
svnmerge-migrate-history.py and the Python bindings memory leak fix. Some of
these just seem like they should be in a top notch release. I don't want to be
like Oracle, as our DBA says, always wait for the .1 release.
r31114 | hwright | 2008-05-09 07:47:36 -0700 (Fri, 09 May 2008) | 7 lines
Merge r31107 from trunk:
ra_serf: Fix API bug when fetched_rev is non-NULL and revision is invalid.
+1: epg, jerenkrantz, kfogel
r31147 | hwright | 2008-05-12 21:50:16 -0700 (Mon, 12 May 2008) | 12 lines
Merge r31055, r31066 from trunk:
* r31055 (subversion/bindings/swig), r31066
Make svn_wc_diff_callbacks2_t work in Python (not a new interface,
it was just broken for Python until r31055).
'svn merge -c 31055 ^/trunk/subversion/bindings/swig
subversion/bindings/swig' cleanly merges the Python change.
r31148 | hwright | 2008-05-12 21:51:24 -0700 (Mon, 12 May 2008) | 12 lines
Merge r31145 from trunk:
Fix memory leaks in Python bindings.
Memory leaks are bad.
Depends on r31055 and r31066 (see above)
r31164 | hwright | 2008-05-14 07:44:27 -0700 (Wed, 14 May 2008) | 9 lines
Merge r31081, r31082, r31153, r31155 from trunk:
* r31081, r31082, r31153, r31155
Various improvements to svnmerge-migrate-history.py for dealing
with the kinds of problems caused by svnmerge.py's total ignorance
of object history.
+1: cmpilato, dlr
r31182 | hwright | 2008-05-14 20:53:40 -0700 (Wed, 14 May 2008) | 7 lines
Merge r31162 from trunk:
Fix an OBOE in svn_path_splitext().
+1: cmpilato, kfogel, danielsh
r31183 | hwright | 2008-05-14 20:56:17 -0700 (Wed, 14 May 2008) | 10 lines
Merge r31131 from trunk:
Restore compatibility of svn_client_add3()'s non-recursive behavior.
This fixes an API incompatibility. But in practice it's a minor
problem, and could probably wait until 1.5.1 if necessary.
+1: kfogel, epg, cmpilato
r31184 | hwright | 2008-05-14 20:57:18 -0700 (Wed, 14 May 2008) | 7 lines
Merge r31056 from trunk:
Enable svn.diff in swig-py on Windows.
+1: epg, arfrever, cmpilato
The changes that have not yet been merged in that I feel should are:
* r31059, 31060, 31061, 31075, 31151
Fix issue #3157, Merging a change from a path's natural history
creates self-referential mergeinfo
r31075 and 31151 are the core fixes for the issue, the other revisions
are a new test for it (and a few minor follow-ups to that test).
The addition of the new merge test in r31059 trivally conflicts
because several other new merge tests are not backported yet.
+1: pburba, cmpilato
* r31243, r31246, r31249, r31250, r31251, r31271
Fix issue #3200, API Ickiness: svn_client_ctx_t shouldn't carry
"extra revprops"; the svn_client APIs should
If we *don't* backport this in 1.5.0, we need to undo the changes
or otherwise rework them to use newly-revved svn_client APIs.
This is not quite finished -- need feedback from Ruby, Perl, and
JavaHL bindings maintainers.
Given the changes that have been merged into the branch, I think we should merge
in these two sets of changes, cut an RC6 by Wednesday if possible. If the Ruby,
Perl and JavaHL bindings are not completely up to snuff, we can extend them in
1.5.1. Better the C API be clean.
To unsubscribe, e-mail: dev-unsubscribe_at_subversion.tigris.org
For additional commands, e-mail: dev-help_at_subversion.tigris.org
Received on 2008-05-19 23:48:08 CEST