subversion-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Joe Schaefer <joe_schae...@yahoo.com>
Subject Re: Coniguring 301/302 redirects to track an fspath rename
Date Sun, 03 Feb 2013 21:40:11 GMT
>________________________________
> From: Daniel Shahaf <d.s@daniel.shahaf.name>
>To: Joe Schaefer <joe_schaefer@yahoo.com> 
>Cc: Greg Stein <gstein@gmail.com>; "dev@subversion.apache.org" <dev@subversion.apache.org>

>Sent: Sunday, February 3, 2013 4:15 PM
>Subject: Re: Coniguring 301/302 redirects to track an fspath rename
> 
>I already told you Joe: go over existing callers of svn_ra_open4() and
>see what value each of them could provide as the ?p= argument.


Thanks for barking more orders at me- I'd prefer collaboration where
possible.   It is far easier for me to get some discussion out of the
experts in the room than it is for me to go grepping for things in
the source that I barely understand.  I certainly didn't expect you
to write the http module to support this, even tho it's your idea;
please be respectful of the fact that I don't have enough knowledge
about the subversion codebase to provide a desirable patch at this
juncture.  Thanks.


>checkout will provide NULL (or maybe SVN_INVALID_REVNUM, if we don't
>make that mean "HEAD").


Pretty sure I've checked out specific revisions of trees using pegrevs
in the past.


>
>cat will provide the operative or peg revision.
>
>update/switch will provide the BASE revision of the update root, or
>something like that (compare the "target revision" concept in the
>editor).


>
>Someone needs to fill in the rest...


I think Greg was trying to point out that the OPTIONS request
is oblivious to the eventual svn operations desired- the only
thing I know for certain is that it knows the proper url to
throw the initial request at.  If it knows that much, perhaps
from recalling initial command line args, perhaps the remainder
of those args involving revisions can be provided in the OPTIONS request?
All I really need is some ballpark number to provide meaningful support
for walking through the history of a graduated podling;
right now the tools available to a http server admin trying
to work with the fledgling redirect support in newer svn
clients are inadequate for the ASF's use case, as I'm sure you
know Daniel.

Mime
View raw message