You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Ben Collins-Sussman <su...@collab.net> on 2002/07/23 06:37:17 UTC
Re: svn commit: rev 2637 - trunk/subversion/mod_dav_svn
sussman@tigris.org writes:
> + case DAV_PROPID_getlastmodified:
> case DAV_PROPID_creationdate:
> {
> + /* In subversion terms, the date attached to a file's CR is
> + the true "last modified" time. However, we're defining
> + creationdate in the same way. IMO, the "creationdate" is
> + really the date attached to the revision in which the item
> + *first* came into existence; this would found by tracing
> + back through the log of the file -- probably via
> + svn_fs_revisions_changed. gstein, is it a bad thing that
> + we're currently using 'creationdate' to mean the same thing
> + as 'last modified date'? */
IIRC, the 'DAV:creationdate' property is being interpreted as the
'svn:wc:committed-date' entryprop in the client, which eventually ends
up being stored as the 'committed-date' attribute in the entry.
If you agree with what I'm saying in the comment above, I think that
*instead* we should be using 'DAV:getlastmodified' for this purpose.
It should be paired with the CR and last-author entryprops. ((And
'DAV:creationdate' should truly trace back through many revisions.)
Does anyone agree with me? See what I'm saying? This is a matter of
changing how we map DAV concepts to svn concepts.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Re: svn commit: rev 2637 - trunk/subversion/mod_dav_svn
Posted by Greg Stein <gs...@lyra.org>.
The DAV:creationdate property states when the *resource* came into
existence. Since that is the version resource, then the creationdate
effectively means the revision date.
On the "public" resources, the creationdate would represent the
first-existence date (as you stated in the comment), and the getlastmodified
property would be the date on the created-rev.
Cheers,
-g
On Tue, Jul 23, 2002 at 01:37:17AM -0500, Ben Collins-Sussman wrote:
> sussman@tigris.org writes:
>
> > + case DAV_PROPID_getlastmodified:
> > case DAV_PROPID_creationdate:
> > {
> > + /* In subversion terms, the date attached to a file's CR is
> > + the true "last modified" time. However, we're defining
> > + creationdate in the same way. IMO, the "creationdate" is
> > + really the date attached to the revision in which the item
> > + *first* came into existence; this would found by tracing
> > + back through the log of the file -- probably via
> > + svn_fs_revisions_changed. gstein, is it a bad thing that
> > + we're currently using 'creationdate' to mean the same thing
> > + as 'last modified date'? */
>
> IIRC, the 'DAV:creationdate' property is being interpreted as the
> 'svn:wc:committed-date' entryprop in the client, which eventually ends
> up being stored as the 'committed-date' attribute in the entry.
>
> If you agree with what I'm saying in the comment above, I think that
> *instead* we should be using 'DAV:getlastmodified' for this purpose.
> It should be paired with the CR and last-author entryprops. ((And
> 'DAV:creationdate' should truly trace back through many revisions.)
>
> Does anyone agree with me? See what I'm saying? This is a matter of
> changing how we map DAV concepts to svn concepts.
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: svn-unsubscribe@subversion.tigris.org
> For additional commands, e-mail: svn-help@subversion.tigris.org
--
Greg Stein, http://www.lyra.org/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org