AssertionError("%d != %d" % (len(tview), tview_len)) in delta.py, line 57, in apply_txdelta_window

[Sergey]

There are Mercurial, hgsubversion plug-in, and subvertpy involved here, but I'm filing this in subvertpy bug tracker because an assertion was hit.

The full trace is attached. On Windows XP, I was trying to use hg 1.8.4 and hgsubversion (fairly recent) with subvertpy 0.8 to clone a Subversion 1.6.15 repository (on RHEL 5). The repository is structured as multiple projects, each with its own branches/tags/trunk. There are 10+ projects currently, all but two I could clone successfully. The remaining two have the same error -- assertion in subvertpy.

If you could take a look at it and let me know what seems to be the problem I'd be grateful. If there's additional information needed please let me know. I had to replace some unique identifiers in the trace to protect my employer.

[Jelmer Vernooij (jelmer)]

This generally happens when trying to apply a delta against an incorrect base; in all cases I've seen so far this has happened because an invalid delta or invalid fulltext were passed in.

[Sergey (hvmptydvmpty)]

Cross-filed https://bitbucket.org/durin42/hgsubversion/issue/294/subvertpy-fails-with-an-assertion-on-hg.

[Jelmer Vernooij (jelmer)]

Have you tried reproducing the issue while using hgsubversion on the SWIG bindings?

[Sergey (hvmptydvmpty)]

Yes, using SWIG binding works.

[Jelmer Vernooij (jelmer)]

On Wed, 2011-07-27 at 02:12 +0000, Sergey wrote:
> Yes, using SWIG binding works.
Hmm, interesting. I guess that means there is a bug in the subvertpy
implementation. Thanks for checking.

  status triaged
  importance high

Cheers,

Jelmer

[Sergey (hvmptydvmpty)]

If there are any checks that can help determine the root cause I'm ready to try them out.

[Jelmer Vernooij (jelmer)]

On Thu, 2011-07-28 at 02:31 +0000, Sergey wrote:
> If there are any checks that can help determine the root cause I'm ready
> to try them out.
What would really help is having the parameters to apply_txdelta_window,
as that would make it possible to reproduce the problem here locally.

Cheers,

Jelmer

[David Coles (dcoles)]

At least for me the parameters that I see the crash on are:
apply_txdelta_window(sbuf, (0L, 6563, 6562, 2, [(0, 0, 927), (0, 928, 5635)], '')

(where 'sbuff' is the full contents of one of the files stored in the repository).

[Jelmer Vernooij (jelmer)]

On Fri, 2011-07-29 at 16:52 +0000, David Coles wrote:
> At least for me the parameters that I see the crash on are:
> apply_txdelta_window(sbuf, (0L, 6563, 6562, 2, [(0, 0, 927), (0, 928, 5635)], '')
>
> (where 'sbuff' is the full contents of one of the files stored in the
> repository).
Can you provide sbuf as well? The combination of these things is what makes it useful.

Cheers,

Jelmer

[Dennis Schridde (devurandom)]

I have the same issue, also using hgsubversion.

$ hg pull -u
pulling from svn+http://svn.enlightenment.org/svn/e
[r25195] raster: todo++
[…]

Complete backtrace attached.

Parameters to apply_txdelta_window were:
(0L, 6599, 6703, 2, [(0, 0, 1471), (2, 0, 97), (0, 1464, 5135)], 'needs way to replace or extend the right click menu (disable/enable\n rename/delete/refresh etc.)')

The contents of sbuf are attached.

This is the code that generated the output:
    f = open("/tmp/hgsvn-params", "w")
    f.write(str((sview_offset, sview_len, tview_len, src_ops, ops, new_data)))
    f.close()
    f = open("/tmp/hgsvn-sbuf", "w")
    f.write(sbuf)
    f.close()
    if len(tview) != tview_len:
        [...]

hgsubversion: e252f9355933
Subversion: 1.6.17
bindings: Subvertpy 0.8.10