vc-annotate fails

vc-annotate fails

[Robert J. Chassell]

Today's GNU Emacs CVS snapshot, Fri, 2005 Jul  1  10:39 UTC
GNU Emacs 22.0.50.34 (i686-pc-linux-gnu, GTK+ Version 2.6.8)
started with

    src/emacs -Q -D

The lisp/loadhist.el file is 232 lines long.  However, after visiting
it, and running `vc-annotate' (C-x v g) to produce

    *Annotate loadhist.el (rev 1.33)*

only a 67 line buffer appears -- and that buffer, which includes both
the beginning and end of loadhist.el is missing part of the middle,
including the defun for `unload-feature'.

--
    Robert J. Chassell
    bob@rattlesnake.com                         GnuPG Key ID: 004B4AC8
    http://www.rattlesnake.com                  http://www.teak.cc

Re: vc-annotate fails

[Kim F. Storm]

"Robert J. Chassell" <bob@rattlesnake.com> writes:

> Today's GNU Emacs CVS snapshot, Fri, 2005 Jul  1  10:39 UTC
> GNU Emacs 22.0.50.34 (i686-pc-linux-gnu, GTK+ Version 2.6.8)
> started with
>
>     src/emacs -Q -D
>
> The lisp/loadhist.el file is 232 lines long.  However, after visiting
> it, and running `vc-annotate' (C-x v g) to produce
>
>     *Annotate loadhist.el (rev 1.33)*
>
> only a 67 line buffer appears -- and that buffer, which includes both
> the beginning and end of loadhist.el is missing part of the middle,
> including the defun for `unload-feature'.

Sounds like the cvs / ssh problem described in INSTALL.CVS.


-- 
Kim F. Storm <storm@cua.dk> http://www.cua.dk

Re: vc-annotate fails

[Robert J. Chassell]

Kim F. Storm wrote

    Sounds like the cvs / ssh problem described in INSTALL.CVS.

Yes.  However, the bug was discussed a long time ago.
vc-annotate (or cvs or ssh) should have been fixed by now.  

-- 
    Robert J. Chassell                         
    bob@rattlesnake.com                         GnuPG Key ID: 004B4AC8
    http://www.rattlesnake.com                  http://www.teak.cc

Re: vc-annotate fails

[Stefan Monnier]

> Kim F. Storm wrote
>     Sounds like the cvs / ssh problem described in INSTALL.CVS.

> Yes.  However, the bug was discussed a long time ago.
> vc-annotate (or cvs or ssh) should have been fixed by now.

IIUC the general "consensus" is that it's a problem in SSH but that the SSH
people don't accept it as a bug.  It can't be fixed elsewhere, only
worked-around.  Last I heard recent development versions of CVS contain
a workaround (similar to the one proposed in etc/PROBLEMS).


        Stefan

Re: vc-annotate fails

[Richard M. Stallman]

    Yes.  However, the bug was discussed a long time ago.
    vc-annotate (or cvs or ssh) should have been fixed by now.  

I am trying to get the bug fixed.  Maybe I am getting close.

Re: vc-annotate fails

[Kevin Rodgers]

Richard M. Stallman wrote:
>     Yes.  However, the bug was discussed a long time ago.
>     vc-annotate (or cvs or ssh) should have been fixed by now.  
> 
> I am trying to get the bug fixed.  Maybe I am getting close.

If the bug is not fixed, should the workaround described in INSTALL.CVS
be implemented within Emacs itself?  That is, the script would be
installed as "cvs_rsh" in exec-directory by `make install` and pcl-cvs
would do (setenv "CVS_RSH" (expand-file-name "cvs_rsh" exec-directory)).

-- 
Kevin Rodgers

Re: vc-annotate fails

[Richard M. Stallman]

    If the bug is not fixed, should the workaround described in INSTALL.CVS
    be implemented within Emacs itself?  That is, the script would be
    installed as "cvs_rsh" in exec-directory by `make install` and pcl-cvs
    would do (setenv "CVS_RSH" (expand-file-name "cvs_rsh" exec-directory)).

This would be better than nothing, but nowhere near as good as
fixing the bug.  I am continuing to work for that.