| Submitter | Sietse Brouwer |
|---|---|
| Date | Dec. 3, 2015, 1:22 a.m. |
| Message ID | <CAF=dkzx8CBYXVve9r-9qgqZBWFrmpdxw_VrYc9bnyOmVDOWfEg@mail.gmail.com> |
| Download | mbox | patch |
| Permalink | /patch/11778/ |
| State | Accepted |
| Headers | show |
Comments
On 12/02/2015 05:22 PM, Sietse Brouwer wrote: > This fixes issue4983 [1]: post-init and post-clone hooks always crash > (after hg init / hg clone) Thanks, you patch had some bad damage from the your email client. Consider using the patchbomb extensions or the pushgate to send a patch next week. I've pushed this to the clowncopter. > I realise that Matt's e-mail backlog is terrifying right now [2]; but > the bug is a regression, and the fix is a simple one, so I thought I'd > send it in anyway. Augie is the goaltender this week, so the important backlog is the patchwork one. The regression is indeed pretty bad. (and the scheduled bugfixe version was 2 day ago…)
> Thanks, you patch had some bad damage from the your email client. Consider > using the patchbomb extensions or the pushgate to send a patch next week. Noted. I thought sending as plain text would keep Gmail from mangling it, but nooo. Have updated the wiki to make that clear. Excited to try out the pushgate in future. Thank you very much for the repair work. https://www.mercurial-scm.org/wiki/ContributingChanges#Experimental_submission_via_push_gate >> Matt's e-mail backlog is terrifying > Augie is the goaltender this week, so the important backlog is the patchwork one. That's good news. Pity about missing 3.6.2, yes... ah well, such is life. Cheers, Sietse On 3 December 2015 at 10:20, Pierre-Yves David <pierre-yves.david@ens-lyon.org> wrote: > > > On 12/02/2015 05:22 PM, Sietse Brouwer wrote: >> >> This fixes issue4983 [1]: post-init and post-clone hooks always crash >> (after hg init / hg clone) > > > Thanks, you patch had some bad damage from the your email client. Consider > using the patchbomb extensions or the pushgate to send a patch next week. > > I've pushed this to the clowncopter. > >> I realise that Matt's e-mail backlog is terrifying right now [2]; but >> the bug is a regression, and the fix is a simple one, so I thought I'd >> send it in anyway. > > > Augie is the goaltender this week, so the important backlog is the patchwork > one. > > The regression is indeed pretty bad. (and the scheduled bugfixe version was > 2 day ago…) > > -- > Pierre-Yves David
On 12/03/2015 02:10 AM, Sietse Brouwer wrote: >> Thanks, you patch had some bad damage from the your email client. Consider >> using the patchbomb extensions or the pushgate to send a patch next week. > > Noted. I thought sending as plain text would keep Gmail from mangling > it, but nooo. Have updated the wiki to make that clear. Excited to try > out the pushgate in future. Thank you very much for the repair work. > > https://www.mercurial-scm.org/wiki/ContributingChanges#Experimental_submission_via_push_gate > >>> Matt's e-mail backlog is terrifying > >> Augie is the goaltender this week, so the important backlog is the patchwork one. > > That's good news. Pity about missing 3.6.2, yes... ah well, such is life. That said, you still caught an exception and are eligible for a Mercurial T-shirt. Can you send me the following information in private, T-Shirt size (Male or Female too), Name, Full address, Phone number I'll ship you one.
> That said, you still caught an exception and are eligible for a Mercurial T-shirt. Can you send me the following information in private, Private e-mail sent. I think it's very cool that this project (a) does not just deal in online activities, and (b) has so few regressions that you can generally afford to send out T-shirts for them. That puts Mercurial's reliability somewhere in the same league as TeX, which is amazing. Cheers, Sietse -- Sietse Brouwer -- sbbrouwer@gmail.com -- +31 6 13.456.848 Korte Langestraat 2 -- 2312 SK Leiden -- the Netherlands On 3 December 2015 at 11:14, Pierre-Yves David <pierre-yves.david@ens-lyon.org> wrote: > > > On 12/03/2015 02:10 AM, Sietse Brouwer wrote: >>> >>> Thanks, you patch had some bad damage from the your email client. >>> Consider >>> using the patchbomb extensions or the pushgate to send a patch next week. >> >> >> Noted. I thought sending as plain text would keep Gmail from mangling >> it, but nooo. Have updated the wiki to make that clear. Excited to try >> out the pushgate in future. Thank you very much for the repair work. >> >> >> https://www.mercurial-scm.org/wiki/ContributingChanges#Experimental_submission_via_push_gate >> >>>> Matt's e-mail backlog is terrifying >> >> >>> Augie is the goaltender this week, so the important backlog is the >>> patchwork one. >> >> >> That's good news. Pity about missing 3.6.2, yes... ah well, such is life. > > > That said, you still caught an exception and are eligible for a Mercurial > T-shirt. Can you send me the following information in private, > > T-Shirt size (Male or Female too), > Name, > Full address, > Phone number > > I'll ship you one. > > -- > Pierre-Yves David
On 12/04/2015 12:54 AM, Sietse Brouwer wrote: >> That said, you still caught an exception and are eligible for a Mercurial T-shirt. Can you send me the following information in private, > > Private e-mail sent. I think it's very cool that this project (a) does > not just deal in online activities, and (b) has so few regressions > that you can generally afford to send out T-shirts for them. That puts > Mercurial's reliability somewhere in the same league as TeX, which is > amazing. Even if we try hard, we are unfortunately not that regression free. This is why our T-Shirt program does not plan for exponential rewards.
Patch
diff --git a/mercurial/hook.py b/mercurial/hook.py --- a/mercurial/hook.py +++ b/mercurial/hook.py @@ -120,10 +120,11 @@ env = {} # make in-memory changes visible to external process - tr = repo.currenttransaction() - repo.dirstate.write(tr) - if tr and tr.writepending(): - env['HG_PENDING'] = repo.root + if repo is not None: + tr = repo.currenttransaction() + repo.dirstate.write(tr) + if tr and tr.writepending(): + env['HG_PENDING'] = repo.root for k, v in args.iteritems(): if callable(v): diff --git a/tests/test-hook.t b/tests/test-hook.t --- a/tests/test-hook.t +++ b/tests/test-hook.t @@ -681,10 +681,19 @@ ...skipping... - tr = repo.currenttransaction() - repo.dirstate.write(tr) - if tr and tr.writepending(): - env['HG_PENDING'] = repo.root + if repo is not None: + tr = repo.currenttransaction() + repo.dirstate.write(tr) + if tr and tr.writepending(): + env['HG_PENDING'] = repo.root for k, v in args.iteritems(): if callable(v): diff --git a/tests/test-hook.t b/tests/test-hook.t --- a/tests/test-hook.t +++ b/tests/test-hook.t @@ -681,10 +681,19 @@ $ hg tag -f foo ['a', 'foo', 'tip'] +post-init hooks must not crash (issue4983) +This also creates the `to` repo for the next test block. + + $ cd ..
This fixes issue4983 [1]: post-init and post-clone hooks always crash (after hg init / hg clone) I realise that Matt's e-mail backlog is terrifying right now [2]; but the bug is a regression, and the fix is a simple one, so I thought I'd send it in anyway. @FUJIWARA Katsunori: this patch touches dirstate code. Does it look all right to you? Kind regards, Sietse Sietse Brouwer [1] http://bz.mercurial-scm.org/show_bug.cgi?id=4983 [2] https://www.mercurial-scm.org/wiki/mpm/flow # HG changeset patch # User Sietse Brouwer <sbbrouwer@gmail.com> # Date 1449103101 -3600 # Thu Dec 03 01:38:21 2015 +0100 # Branch stable # Node ID b5b96dcd27a0e89c0161f5863f4dc58c06a9519c # Parent 7e1fac6c0a9ce6afd3edeed5e47bcca343155d8a dirstate: don't write repo.currenttransaction to repo.dirstate if repo is None (issue4983) Some hooks, such as post-init and post-clone, do not get a repo parameter in their environment. If there is no repo, there is no repo.currenttransaction(); attempting to retrieve it anyway was causing crashes. Now currenttransaction is only retrieved and written if the repo is not None. + $ cat << EOF >> hgrc-with-post-init-hook + > [hooks] + > post-init = printenv.py post-init + > EOF + $ HGRCPATH=hgrc-with-post-init-hook hg init to + post-init hook: HG_ARGS=init to HG_OPTS={'insecure': None, 'remotecmd': '', 'ssh': ''} HG_PATS=['to'] HG_RESULT=0 + new commits must be visible in pretxnchangegroup (issue3428) - $ cd .. - $ hg init to $ echo '[hooks]' >> to/.hg/hgrc $ echo 'prechangegroup = hg --traceback tip' >> to/.hg/hgrc $ echo 'pretxnchangegroup = hg --traceback tip' >> to/.hg/hgrc