Sorry, folks. I'm a guilty party here, too. I got cranky about a prior
commit that made bogus assumptions about the state of the database,
ultimately annoyin^H^H^H^H^H^H^Hencouraging Bert to make this format bump.
On 04/19/2011 05:37 PM, Greg Stein wrote:
> Bert:
>
> I think you should check with the community before doing a format
> bump. You're cutting us out of the conversation.
>
> -g
>
> On Tue, Apr 19, 2011 at 17:05, <rhuijben_at_apache.org> wrote:
>> Author: rhuijben
>> Date: Tue Apr 19 21:05:28 2011
>> New Revision: 1095214
>>
>> URL: http://svn.apache.org/viewvc?rev=1095214&view=rev
>> Log:
>> Bump our working copy format version to 28 and fixup the remaining MD5
>> references in the NODES table to SHA1. (This also fixes the reference counts)
>>
>> On any working copy checked out in +- the last half year (and probably longer)
>> this only bumps the working copy format number.
>>
>> * subversion/libsvn_wc/upgrade.c
>> (bump_to_28): New function.
>> (svn_wc__upgrade_sdb): Add bump to 28.
>>
>> * subversion/libsvn_wc/wc-metadata.sql
>> (STMT_UPGRADE_TO_28): Add the bump to format 28 query.
>>
>> * subversion/libsvn_wc/wc.h
>> (SVN_WC__VERSION): Document format 28 and switch to 28.
>>
>> * subversion/libsvn_wc/wc_db_pristine.c
>> (svn_wc__db_pristine_read): Revert r1095194, which reverted some parts of
>> r1091381.
>>
>> Modified:
>> subversion/trunk/subversion/libsvn_wc/upgrade.c
>> subversion/trunk/subversion/libsvn_wc/wc-metadata.sql
>> subversion/trunk/subversion/libsvn_wc/wc.h
>> subversion/trunk/subversion/libsvn_wc/wc_db_pristine.c
>> ...
--
C. Michael Pilato <cmpilato_at_collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
Received on 2011-04-20 03:42:02 CEST