# StGit: kha/{stable,safe,experimental} updated 6 messages from 2008-04-13 to 2008-04-15. Participants: Karl Hasselström, Samuel Tardieu. Thread: https://gitlist.dev/t/13094 ## Karl Hasselström, 2008-04-13 19:11 Subject: StGit: kha/{stable,safe,experimental} updated Message-ID: <20080413191140.GA1912@diana.vm.bytemark.co.uk> URL: https://gitlist.dev/e/20080413191140.GA1912%40diana.vm.bytemark.co.uk ``` I've rebased my branches on top of Catalin's latest master and stable. They include all the patches I've sent to the list recently. I've added a "stable" branch, for patches intended for Catalin's "stable". In the experimental branch, I've updated the reset command a bit -- it should now be quite usable. Bug reports and feature requests and general comments welcome. (Note to those who already run my experimental branch: I've changed the stack log format, so you'll have to nuke your .stgit branches. StGit will remind you if you forget.) -+- The following changes since commit bd3f01319a03fc883d391f24639176f0f65727e5: Catalin Marinas (1): Release 0.14.2 are available in the git repository at: git://repo.or.cz/stgit/kha.git stable Karl Hasselström (1): Fix problems in t1000-branch-create t/t1000-branch-create.sh | 56 +++++++++++++++++++++++++++++---------------- 1 files changed, 36 insertions(+), 20 deletions(-) -+- The following changes since commit 719309aa54c8fa43bd8700f09b1e293923ed4e0a: Catalin Marinas (1): Merge branch 'stable' are available in the git repository at: git://repo.or.cz/stgit/kha.git safe Karl Hasselström (5): Fix problems in t1000-branch-create Merge branch 'stable' into stable-master-merge Don't both change directory and set GIT_WORK_TREE Log environment and cwd as well as the actual command emacs mode: handle "stg status" output with # contrib/stgit.el | 2 +- stgit/lib/git.py | 14 +++++++---- stgit/run.py | 6 +++++ t/t1000-branch-create.sh | 56 +++++++++++++++++++++++++++++---------------- 4 files changed, 52 insertions(+), 26 deletions(-) -+- The following changes since commit 999dcbc241e377f8b39f0064dd8cc1670e667f18: Karl Hasselström (1): emacs mode: handle "stg status" output with # are available in the git repository at: git://repo.or.cz/stgit/kha.git experimental Karl Hasselström (12): Disable patchlog test for "stg new" Convert "stg new" to the new infrastructure Refactor --author/--committer options Let "stg new" support more message options Emacs mode: use "stg new --file" Convert "stg delete" to the new infrastructure Emacs mode: delete patches Add property with a list of all patch names Library functions for tree and blob manipulation Write to a stack log when stack is modified Add utility function for reordering patches New command: stg reset contrib/stgit.el | 33 +++++++--- stgit/commands/branch.py | 18 ++++-- stgit/commands/clone.py | 3 +- stgit/commands/coalesce.py | 3 +- stgit/commands/commit.py | 6 +- stgit/commands/common.py | 33 +++------- stgit/commands/delete.py | 70 ++++++--------------- stgit/commands/edit.py | 26 +------- stgit/commands/float.py | 3 + stgit/commands/fold.py | 3 +- stgit/commands/imprt.py | 3 +- stgit/commands/init.py | 3 +- stgit/commands/new.py | 97 ++++++++++++++++------------- stgit/commands/pick.py | 4 +- stgit/commands/pop.py | 3 +- stgit/commands/pull.py | 3 +- stgit/commands/push.py | 6 +- stgit/commands/rebase.py | 4 +- stgit/commands/refresh.py | 4 +- stgit/commands/rename.py | 4 +- stgit/commands/reset.py | 67 ++++++++++++++++++++ stgit/commands/resolved.py | 4 +- stgit/commands/sink.py | 3 +- stgit/commands/sync.py | 5 +- stgit/commands/uncommit.py | 2 +- stgit/lib/git.py | 147 ++++++++++++++++++++++++++++++++++++-------- stgit/lib/log.py | 130 ++++++++++++++++++++++++++++++++++++++ stgit/lib/stack.py | 1 + stgit/lib/transaction.py | 17 +++++- stgit/main.py | 2 + stgit/utils.py | 50 +++++++++++++++ t/t1400-patch-history.sh | 2 - t/t1600-delete-one.sh | 8 +- 33 files changed, 559 insertions(+), 208 deletions(-) create mode 100644 stgit/commands/reset.py create mode 100644 stgit/lib/log.py -- Karl Hasselström, kha@treskal.com www.treskal.com/kalle ``` ## Samuel Tardieu, 2008-04-14 06:44 Subject: Re: StGit: kha/{stable,safe,experimental} updated Message-ID: <2008-04-14-08-44-30+trackit+sam@rfc1149.net> URL: https://gitlist.dev/e/2008-04-14-08-44-30%2Btrackit%2Bsam%40rfc1149.net In-Reply-To: <20080413191140.GA1912@diana.vm.bytemark.co.uk> ``` >>>>> "Karl" == Karl Hasselström writes: Karl> In the experimental branch, I've updated the reset command a bit Karl> -- it should now be quite usable. Bug reports and feature Karl> requests and general comments welcome. (Note to those who Karl> already run my experimental branch: I've changed the stack log Karl> format, so you'll have to nuke your .stgit Karl> branches. StGit will remind you if you forget.) Karl, I was using the StGit in Debian/unstable and just switched to your experimental branch. When doing "stg pull", I get, at the end: Pushing patch "pr20822" ... done Pushing patch "pr22387" ... done Pushing patch "pr30614" ... done Error: Unhandled exception: Traceback (most recent call last): File "/usr/local/lib/python2.4/site-packages/stgit/main.py", line 280, in main ret = command.func(parser, options, args) File "/usr/local/lib/python2.4/site-packages/stgit/commands/pull.py", line 109, in func log.log_entry(log.default_stack(), 'stg pull') File "/usr/local/lib/python2.4/site-packages/stgit/lib/log.py", line 45, in log_entry stack.repository.refs.get(ref))] File "/usr/local/lib/python2.4/site-packages/stgit/lib/log.py", line 94, in __init__ self.unapplied, patches = pl('unapplied') File "/usr/local/lib/python2.4/site-packages/stgit/lib/log.py", line 89, in pl return ([pn for sha1, pn in patches], ValueError: need more than 0 values to unpack Note that I have no unapplied patch in case it matters, only hidden ones (and applied ones of course). ``` ## Karl Hasselström, 2008-04-14 07:03 Subject: Re: StGit: kha/{stable,safe,experimental} updated Message-ID: <20080414070348.GA9135@diana.vm.bytemark.co.uk> URL: https://gitlist.dev/e/20080414070348.GA9135%40diana.vm.bytemark.co.uk In-Reply-To: <2008-04-14-08-44-30+trackit+sam@rfc1149.net> ``` On 2008-04-14 08:44:29 +0200, Samuel Tardieu wrote: > I was using the StGit in Debian/unstable and just switched to your > experimental branch. > > When doing "stg pull", I get, at the end: > > Pushing patch "pr20822" ... done > Pushing patch "pr22387" ... done > Pushing patch "pr30614" ... done > Error: Unhandled exception: > Traceback (most recent call last): > File "/usr/local/lib/python2.4/site-packages/stgit/main.py", line > 280, in main > ret = command.func(parser, options, args) > File > "/usr/local/lib/python2.4/site-packages/stgit/commands/pull.py", > line 109, in func > log.log_entry(log.default_stack(), 'stg pull') > File "/usr/local/lib/python2.4/site-packages/stgit/lib/log.py", line > 45, in log_entry > stack.repository.refs.get(ref))] > File "/usr/local/lib/python2.4/site-packages/stgit/lib/log.py", line > 94, in __init__ > self.unapplied, patches = pl('unapplied') > File "/usr/local/lib/python2.4/site-packages/stgit/lib/log.py", line > 89, in pl > return ([pn for sha1, pn in patches], > ValueError: need more than 0 values to unpack Euh! Thanks for the report, I'll look into it tonight after work. > Note that I have no unapplied patch in case it matters, only hidden > ones (and applied ones of course). Yeah, that might be a clue. Though I was under the impression that I'd tested it with a fully applied stack ... I guess the test suite needs an addition or two. -- Karl Hasselström, kha@treskal.com www.treskal.com/kalle ``` ## Karl Hasselström, 2008-04-14 08:11 Subject: Re: StGit: kha/{stable,safe,experimental} updated Message-ID: <20080414081139.GB9135@diana.vm.bytemark.co.uk> URL: https://gitlist.dev/e/20080414081139.GB9135%40diana.vm.bytemark.co.uk In-Reply-To: <20080414070348.GA9135@diana.vm.bytemark.co.uk> ``` On 2008-04-14 09:03:48 +0200, Karl Hasselström wrote: > Though I was under the impression that I'd tested it with a fully > applied stack ... I guess the test suite needs an addition or two. The real proble turned out to be a certain developer who remembers running the test suite before pushing out his branches, even though no such thing occurred. Ahem. This fixes the bug for me; I'll be squashing it into one of the other patches tonight. Sorry for the breakage. diff --git a/stgit/lib/log.py b/stgit/lib/log.py index 083a546..1830cbd 100644 --- a/stgit/lib/log.py +++ b/stgit/lib/log.py @@ -84,7 +84,7 @@ class Log(object): def pl(name): patches = [x.split() for x in self.commit.data.tree.data.entries[name][1] - .data.str.strip().split('\n')] + .data.str.strip().split('\n') if x] # TODO: handle case where we don't have the commit object. return ([pn for sha1, pn in patches], dict((pn, repo.get_commit(sha1)) for sha1, pn in patches)) -- Karl Hasselström, kha@treskal.com www.treskal.com/kalle ``` ## Karl Hasselström, 2008-04-14 20:08 Subject: Re: StGit: kha/{stable,safe,experimental} updated Message-ID: <20080414200814.GA21039@diana.vm.bytemark.co.uk> URL: https://gitlist.dev/e/20080414200814.GA21039%40diana.vm.bytemark.co.uk In-Reply-To: <20080414081139.GB9135@diana.vm.bytemark.co.uk> ``` On 2008-04-14 10:11:39 +0200, Karl Hasselström wrote: > This fixes the bug for me; I'll be squashing it into one of the > other patches tonight. Sorry for the breakage. Done, and pushed out. -- Karl Hasselström, kha@treskal.com www.treskal.com/kalle ``` ## Karl Hasselström, 2008-04-15 00:26 Subject: Re: StGit: kha/{stable,safe,experimental} updated Message-ID: <20080415002613.GA24093@diana.vm.bytemark.co.uk> URL: https://gitlist.dev/e/20080415002613.GA24093%40diana.vm.bytemark.co.uk In-Reply-To: <20080413191140.GA1912@diana.vm.bytemark.co.uk> ``` On 2008-04-13 21:11:40 +0200, Karl Hasselström wrote: > I've rebased my branches on top of Catalin's latest master and > stable. They include all the patches I've sent to the list recently. Updated again with the newest stuff. No shortlog since it's hardly changed since last post. Get it here: git://repo.or.cz/stgit/kha.git stable git://repo.or.cz/stgit/kha.git safe git://repo.or.cz/stgit/kha.git experimental And this time I made _sure_ to run the test suite. -- Karl Hasselström, kha@treskal.com www.treskal.com/kalle ```