Mercurial 3.2 breaks git-remote-hg

[alchemicalhydra]

Upgrading to Mercurial 3.2 appears to break git-remote-hg.

I get the following upon running git fetch:

Traceback (most recent call last):
  File "/home/tomxtobin/bin/git-remote-hg", line 1322, in <module>
    sys.exit(main(sys.argv))
  File "/home/tomxtobin/bin/git-remote-hg", line 1286, in main
    repo = get_repo(url, alias)
  File "/home/tomxtobin/bin/git-remote-hg", line 433, in get_repo
    repo.pull(peer, heads=None, force=True)
  File "/usr/lib/python2.7/site-packages/mercurial/repoview.py", line 307, in __getattr__
    return getattr(self._unfilteredrepo, attr)
AttributeError: 'localrepository' object has no attribute 'pull'

Downgrading to Mercurial 3.1.2 resolves the issue.

[tyrylu]

If you look at the mercurial changelog, they mention it, but unfortunately without any migration instructions. And, they also removed the LocalRepository.push method, but it fortunately appears not to be used anywhere.

[Lekensteyn]

Some of the changes are documented at http://mercurial.selenic.com/wiki/ApiChanges:

localrepo.pull method have been removed in favor of the exchange.pull function. It returns a pulloperation object. the former returned value is stored in the pullop.cgresult attribute.

getbundle to getchangegroup rename is not documented, but here it is:
http://selenic.com/hg/rev/e2806b8613ca

[djmitche]

With Mercuail 3.2.3 and the patch from #28 applied, I'm seeing

Traceback (most recent call last):
  File "/usr/bin/git-remote-hg", line 1307, in <module>
    sys.exit(main(sys.argv))
  File "/usr/bin/git-remote-hg", line 1271, in main
    repo = get_repo(url, alias)
  File "/usr/bin/git-remote-hg", line 429, in get_repo
    repo.pull(peer, heads=None, force=True)
  File "/usr/lib64/python2.7/site-packages/mercurial/repoview.py", line 307, in __getattr__
    return getattr(self._unfilteredrepo, attr)
AttributeError: 'mqrepo' object has no attribute 'pull'

Which is slightly different, but perhaps related?

[fingolfin]

This looks as if the patch is not actually applied -- perhaps you have multiple copies of git-remote-hg on your system and patched the wrong one?

[djmitche]

Yeah, you're right. Ebuild error on my part. Sorry for the noise.

[AdamWill]

#28 Works for me on Fedora Rawhide, FWIW (I'm backporting it to the Fedora package now).

[ChristopherKing42]

I just cloned right from github and got this:

git clone "hg::https://code.google.com/p/pychess/"
Cloning into 'pychess'...
Traceback (most recent call last):
File "/home/chris/bin/git-remote-hg", line 1322, in
sys.exit(main(sys.argv))
File "/home/chris/bin/git-remote-hg", line 1286, in main
repo = get_repo(url, alias)
File "/home/chris/bin/git-remote-hg", line 433, in get_repo
repo.pull(peer, heads=None, force=True)
File "/usr/lib/python2.7/site-packages/mercurial/repoview.py", line 307, in getattr
return getattr(self._unfilteredrepo, attr)
AttributeError: 'localrepository' object has no attribute 'pull'

[AdamWill]

@theking42 the fix has not been merged yet, you have to check out the branch from #28 or apply the fix to your master checkout.

[fingolfin]

@theking42 you could also just use my fork, which fixes this issue and a few other minor things: https://github.com/fingolfin/git-remote-hg

[nengxu]

@fingolfin I used your fork to solve this bug. Thanks.

[bukzor]

Please to merge.

[bukzor]

You can also use mercurial<3.2 as a workaround.

[DanielGGordon]

Is @fingolfin's changes not merged? I just pulled today and got the same error. Will try @fingolfin's fork

[pcn]

👍 for @fingolfin's fork. Is there a discussion of what needs to be done for a merge?

[fingolfin]

@pcn For a merge to happen, Felipe would have to do it, i.e. resurface. But it seems that currently he is busy with other things. Anyway, as long as my fork works for you, that should do it, right? (That said, I don't really use git-remote-hg myself very mich right now, so it indeed would be nice if there was somebody to properly maintain this project).

[fingolfin]

@DanielGGordon No it is not merged. Rest assured, if it ever gets merged, this issue will get a comment clearly stating so ;-)

[djmitche]

Worth noting that https://github.com/glandium/git-cinnabar may be a useful substitute given the apparent abandonment of this project. Dustin

[nengxu]

@djmitche git-cinnabar seems not quite active. The last commit is in April. Have you used it in real projects?

[djmitche]

I haven't as I've begun using hg directly, but I've heard good things from
the people using it. It's possible it hasn't needed any updates in the last
5 months.
On Sep 26, 2015 12:41 PM, "nengxu" notifications@github.com wrote:

@djmitche https://github.com/djmitche git-cinnabar seems not quite
active. The last commit is in April. Have you used it in real projects?

—
Reply to this email directly or view it on GitHub
#27 (comment)
.

[F30]

@nengxu The 'next' branch had its last commit on 28 August. I've used Cinnabar, it works fine but doesn't support merges at the moment, which is quite a limitation.

[nengxu]

@djmitche @F30 Thanks for sharing info. I may stay with git-remote-hg for now since it works fine except this issue:

#10

[nrolland]

same error as of today. thks @djmitche

[fingolfin]

@nrolland So, have you tried my fork, as suggested?

[nrolland]

no, I went to the one mentioned by @djmitche no deep reason though.. is this better to use this fork ? I find it absolutely idiotic that people would do "innovation" on the source control front, anything that makes hg get out of my way is good

[fingolfin]

Well, git-cinnabar is a completely different project, with its own sets of pros and cons... Anyway, if it works reasonably for you, I see no need to switch.

[nrolland]

for noobs like me you recommend this over cinnabar ?

On Fri, Oct 30, 2015 at 6:28 PM, Max Horn notifications@github.com wrote:

Well, git-cinnabar is a completely different project, with its own sets
of pros and cons... Anyway, if it works reasonably for you, I see no need
to switch.

—
Reply to this email directly or view it on GitHub
#27 (comment)
.

Nicolas Rolland
mobile : +33 6 62 88 42 92

[xrchz]

@felipec how come @fingolfin's patch is not merged yet?

[alexei-31]

@fingolfin is not patch, its actual fork against this, which depricated.
Author of @felipec is not update something here since 3 Jun 2014.

[fingolfin]

@nrolland sorry, didn't see your last question. I have never tried git-cinnabar, and so I can neither recommend for nor against it. My understanding is that using it is about exactly as easy / difficult as using this project.

What I can say is that my fork of this project has been working well for me for some years now. However, that doesn't mean it will work perfectly for you -- perhaps you will use it quite differently from the way I use it, uncovering problems I never noticed myself... who knows...

[xrchz]

@fingolfin I'm using your fork now, and it's working great :) Thanks

[felipec]

Fixed in 1e27907.

[felipec]

The home page clearly states:

Contributing

Send your patches to the mailing list git-fc@googlegroups.com (no need to subscribe).

Nobody sent any patches.