Error converting P4 repo to Mercurial

Steve Borho steve at borho.org
Tue Jul 14 10:00:34 CDT 2009


On Tue, Jul 14, 2009 at 3:06 AM, Ron Gery<rong at gamani.com> wrote:
> I am new to Mercurial and know nothing about Python. I am running on Windows
> XP with the P4 depot hosted on the local machine. I'm trying to convert my
> P4 repo to Mercurial, and regardless of what variations I've tried for
> specifying the source (local path, depot address, changelist limitations), I
> soon get the following error message (about 20 changelists, ending with the
> most recent one, are shown preceding the error):
>
> ** unknown exception encountered, details follow
> ** report bug details to http://mercurial.selenic.com/bts/
> ** or mercurial at selenic.com
> ** Mercurial Distributed SCM (version 1.3)
> ** Extensions loaded: extdiff, convert
> Traceback (most recent call last):
>   File "hg", line 38, in <module>
>   File "mercurial\dispatch.pyc", line 16, in run
>   File "mercurial\dispatch.pyc", line 27, in dispatch
>   File "mercurial\dispatch.pyc", line 43, in _runcatch
>   File "mercurial\dispatch.pyc", line 449, in _dispatch
>   File "mercurial\dispatch.pyc", line 317, in runcommand
>   File "mercurial\dispatch.pyc", line 501, in _runcommand
>   File "mercurial\dispatch.pyc", line 454, in checkargs
>   File "mercurial\dispatch.pyc", line 448, in <lambda>
>   File "mercurial\util.pyc", line 370, in check
>   File "hgext\convert\__init__.pyc", line 232, in convert
>   File "hgext\convert\convcmd.pyc", line 369, in convert
>   File "hgext\convert\convcmd.pyc", line 54, in convertsource
>   File "hgext\convert\p4.pyc", line 51, in __init__
>   File "hgext\convert\p4.pyc", line 109, in _parse
> EOFError: EOF read where object expected
>
> My invocation looked like:
> hg convert //depot/gmg/... e:\test
>
> I would simply love to be able to bring my history with me as I move to
> Mercurial, but I'm out of my scope (and available time) in trying to track
> this problem down. Is the above something other people have run into, or is
> this cause by some sort of corruption in my P4 repo?

I've noticed that p4 conversion on Windows is fairly unstable.  I
didn't get a chance to look at it too closely, but it looked like the
output of p4 commands was being truncated without hg being aware of
it.   Doing the same conversions on a Linux box has always worked for
me.

--
Steve Borho



More information about the Mercurial mailing list