Giorgos Keramidas <keramida@ceid.upatras.gr> [Sun, 18 Mar 2007 22:49:05 +0200] rev 4253
tests: expand test-commit a bit to test 'hg commit .' too
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Tue, 20 Mar 2007 22:09:51 -0300] rev 4252
fix hg commit . in the repo root
Brendan Cully <brendan@kublai.com> [Tue, 20 Mar 2007 13:13:39 -0700] rev 4251
transplant: fix ignoring empty changesets (eg after filter)
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Mon, 19 Mar 2007 19:16:35 -0300] rev 4250
Merge with crew-stable.
Brendan Cully <brendan@kublai.com> [Mon, 19 Mar 2007 09:36:06 -0700] rev 4249
Add hg-relink script to contrib
Thomas Arendsen Hein <thomas@intevation.de> [Sun, 18 Mar 2007 20:44:58 +0100] rev 4248
merge with crew-stable
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Mon, 19 Mar 2007 19:07:40 -0300] rev 4247
debugsetparents: wlock the repo before updating the dirstate
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Mon, 19 Mar 2007 19:07:39 -0300] rev 4246
avoid _wsgioutputfile <-> _wsgirequest circular reference
We use the _wsgirequest object itself as the output file object.
To avoid a "self.out = self" which would create another circular
reference, we make the "out" attribute a trivial property.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Mon, 19 Mar 2007 19:07:38 -0300] rev 4245
avoid wsgiapplication <-> MercurialHTTPServer circular reference
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Mon, 19 Mar 2007 19:07:37 -0300] rev 4244
hgwebdir: break templater -> templater circular reference
This is essentially another instance of the same problem fixed
by the parent changeset. See its commit message for the details.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Mon, 19 Mar 2007 19:07:35 -0300] rev 4243
hgweb: break templater -> templater circular reference
The problem were some functions passed in the "defaults" argument
during the templater creation which use "self.t" directly. This
creates the cycle:
hgweb object
-> templater object
-> defaults dict
-> footer function
-> hgweb object
Instead of completely avoding the cycle, we break it after using
the templater.
Thomas Arendsen Hein <thomas@intevation.de> [Sun, 18 Mar 2007 20:39:25 +0100] rev 4242
Fixed synopsis for some mq commands
Thomas Arendsen Hein <thomas@intevation.de> [Sun, 18 Mar 2007 20:37:47 +0100] rev 4241
hg qseries -m: guards file was not ignored
Thomas Arendsen Hein <thomas@intevation.de> [Sun, 18 Mar 2007 19:41:28 +0100] rev 4240
Tests for qapplied/qunapplied fixes (417c2068cb92 and ce6c364ebb2a)
Thomas Arendsen Hein <thomas@intevation.de> [Sun, 18 Mar 2007 12:20:15 +0100] rev 4239
Simplified qseries and hg qapplied to fix some bugs caused by optimization:
- hg qapplied -v now works consistendly to hg qunapplied -v, i.e. showing
guarded (or unapplied because they were guarded during hg qpush) patches.
- hg qapplied <patchname> now works again
Thomas Arendsen Hein <thomas@intevation.de> [Sat, 17 Mar 2007 18:46:52 +0100] rev 4238
Fix issue443: inconsistent output of "hg qunapplied -v"
1. Don't skip over unpushable (guarded) unapplied patches.
2. Don't display unpushable patches, unless -v is given, otherwise
guarded and unguarded patches can't be distinguished.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 22:58:49 -0300] rev 4237
Merge some matcher fixes/cleanups and add some more hg locate tests
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 22:48:23 -0300] rev 4236
util._matcher: remove superfluous variable
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 22:48:22 -0300] rev 4235
add some more tests to hg locate
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 22:48:21 -0300] rev 4234
make the output of test-locate more readable
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 22:48:20 -0300] rev 4233
util._matcher: use "." as the root of empty {rel,}path patterns
Should fix issue332. Really.
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 00:45:18 -0300] rev 4232
Merge with crew-stable
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 00:22:59 -0300] rev 4231
avoid a traceback with hg branch newbranch; hg up
Alexis S. L. Carvalho <alexis@cecm.usp.br> [Fri, 16 Mar 2007 00:22:58 -0300] rev 4230
Fix handling of paths when run outside the repo.
The main problem was that dirstate.getcwd() returned just "",
which was interpreted as "we're at the repo root". It now returns
an absolute path.
The util.pathto function was also changed to deal with the "cwd is
an absolute path" case.