You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@subversion.apache.org by Tim Noell <tn...@gmail.com> on 2007/03/02 21:42:52 UTC

CHANGES file typo, or what (for svn 1.4.2)

Hi svn users:

In the CHANGES file, under User-visible changes, Client, it says:

"* fixed: 'svn status -u' fails in a read-only working copy (r21904, -19)"

Note the issue ID is "-19".  Is this a typo?

I'd like to understand the scenarios that this fix affects.

Can someone point me at the issue, and/or explain?  (is this Issue# 2583?)

We use Linux RHEL3 and RHEL4, svn version 1.3.2, and are considering an upgrade.

Does version 1.4.2 no longer use .svn/tmp for storing temp files
during translations on svn st, svn diff?  IOW,
does this fis the problem where when a user does an svn st on a
working copy owned by some other user,
and they get permission denited on .xvn/tmp/foo Can't open foo?
when permissions oare 750 on .svn/tmp?

Or is this some other scenario that was fixed?

Thx,
Tim Noell

-- 
// "Only dead fish go with the flow"

Re: CHANGES file typo, or what (for svn 1.4.2)

Posted by Blair Zajac <bl...@orcaware.com>.
Tim Noell wrote:
> Hi svn users:
> 
> In the CHANGES file, under User-visible changes, Client, it says:
> 
> "* fixed: 'svn status -u' fails in a read-only working copy (r21904, -19)"
> 
> Note the issue ID is "-19".  Is this a typo?  

No, it means r21904 with 04 replaced with 19, so you don't have to write 
  21919.

> I'd like to understand the scenarios that this fix affects.
> 
> 
> Can someone point me at the issue, and/or explain?  (is this Issue# 2583?)
> 
> We use Linux RHEL3 and RHEL4, svn version 1.3.2, and are considering an upgrade.
> 
> Does version 1.4.2 no longer use .svn/tmp for storing temp files during translations on svn st, svn diff?  IOW, 
> 
> does this fis the problem where when a user does an svn st on a working copy owned by some other user,
> and they get permission denited on .xvn/tmp/foo Can't open foo?
> when permissions oare 750 on .svn/tmp?
> 
> 
> Or is this some other scenario that was fixed?

Well, it's not totally clear without reviewing the diffs, but here's the 
log messages:

------------------------------------------------------------------------
r21904 | dionisos | 2006-10-12 11:11:08 -0700 (Thu, 12 Oct 2006) | 8 lines

Stop requiring write access to the working copy when running status -u.


* subversion/libsvn_client/status.c
   (svn_client_status2): Don't necessarily use the adm dir for
   temporary files (in the case of ra_dav: the file to spool the
   server response into).

------------------------------------------------------------------------
r21919 | dionisos | 2006-10-13 15:00:22 -0700 (Fri, 13 Oct 2006) | 15 lines

Respect the fact that the working copy was to be used read only.


* subversion/libsvn_client/client.h
   (struct svn_client__callback_baton_t.read_only_wc): New. Signals whether
    we should treat the working copy as writeable or not.


* subversion/libsvn_client/ra.c
   (svn_client__open_ra_session_internal): Initialize the new
    callback baton field.
   (open_tmp_file): Only open a file in the working copy root
    iff the working copy is to be treated writeable.  Fall back to
    a temp dir otherwise.

------------------------------------------------------------------------

Regards,
Blair

-- 
Blair Zajac, Ph.D.
<bl...@orcaware.com>
Subversion training, consulting and support
http://www.orcaware.com/svn/

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org