"Daniel L. Rall" <dlr@finemaltcoding.com> wrote on 09/28/2005 12:58:20 AM:
> On Mon, 26 Sep 2005, Paul Burba wrote:
>
> > Here is the latest patch for improved out of date info when using svn
> > status -u.
> >
> > The background on all of this can be found here:
> > http://svn.haxx.se/dev/archive-2005-08/0257.shtml
> ...
> > [[[
> > Provide more complete server-side information about out of date items
when
> > using svn status with -u option.
> >
> > * subversion/include/svn_wc.h
> > (svn_wc_status2_t): Add new fields to store the most recent
revision,
> > author, and commit date for an out of date item, as well as its
url
> > and node type.
>
> This portion of the log should enumerate the field names. ^
Done, see below.
> > * subversion/libsvn_wc/status.c
> > include svn_time.h
> > (dir_baton, file_baton): Add new fields mirroring those in
> > svn_wc_status2_t.
> > (assemble_status): Initialize new svn_wc_status2_t members.
> > (tweak_statushash): Changed arguments, all callers changed.
> > Copy out of date info into svn_wc_status2_t struct.
> > (make_dir_baton, make_file_baton): Initialize new dir_baton and
> > file_baton members respectively.
> > (change_dir_prop, change_file_prop): Temporarily store out of
> > date info in dir_baton and file_baton respectively.
>
> Regarding the "ood_" prefix, I was wondering if the "repos_" prefix
already
> in use for the lock tokens, text status, and prop status wouldn't be a
more
> appropriate prefix. Are those fields also only set when the editor has
> changes which are new to the working copy to report?
I'm not married to the "ood_" prefix and am quite willing to change it,
but to answer your second question, the repos_* fields can have values not
driven by changes new to the working copy. For example, if your WC is up
to date and you then modify some file in it and run svn status -u -v, that
file's repos_text_status == svn_wc_status_modified. But unless someone
else has committed a new change to that file in the meantime, its not out
of date with regard to the repository and the ood_* fields would not be
set. So the ood_* fields are independent of the repos_* fields.
I'm not saying the preceding is a compelling argument for the ood_*
prefix; being new to this I'm quite happy to defer on field names - if
that's the biggest problem my patches have I'll be a happy man :-)
Anyway, thanks for taking a look at this, here's the new log message and
the patch from before:
[[[
Provide more complete server-side information about out of date items when
using svn status with -u option.
* subversion/include/svn_wc.h
(svn_wc_status2_t): Add new fields to store the most recent revision,
author, and commit date for an out of date item, as well as its url
and node type: ood_last_cmt_rev, ood_last_cmt_author,
ood_last_cmt_date,
ood_url, and ood_kind.
* subversion/libsvn_wc/status.c
include svn_time.h
(dir_baton, file_baton): Add new fields mirroring those in
svn_wc_status2_t.
(assemble_status): Initialize new svn_wc_status2_t members.
(tweak_statushash): Changed arguments, all callers changed.
Copy out of date info into svn_wc_status2_t struct.
(make_dir_baton, make_file_baton): Initialize new dir_baton and
file_baton members respectively.
(change_dir_prop, change_file_prop): Temporarily store out of
date info in dir_baton and file_baton respectively.
]]]
_____________________________________________________________________________
Scanned for SoftLanding Systems, Inc. and SoftLanding Europe Plc by IBM Email Security Management Services powered by MessageLabs.
_____________________________________________________________________________
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Wed Sep 28 16:02:51 2005