| Thu, 20 Jun 2002 23:52:35 +0000 |
Richard Jones |
More informative error message
|
| Wed, 19 Jun 2002 03:07:19 +0000 |
Richard Jones |
Moved the file storage commit into blobfiles where it belongs.
|
| Sat, 25 May 2002 07:16:25 +0000 |
Roche Compaan |
Merged search_indexing-branch with HEAD
|
| Wed, 15 May 2002 06:21:21 +0000 |
Richard Jones |
node caching now works, and gives a small boost in performance
|
| Thu, 02 May 2002 13:09:11 +0000 |
Roche Compaan |
Brought search_indexing-branch up to date with latest changes in HEAD.
search_indexing-0-4-2-branch
|
| Wed, 24 Apr 2002 10:38:26 +0000 |
Roche Compaan |
All database files are now created group readable and writable.
|
| Mon, 15 Apr 2002 23:25:15 +0000 |
Richard Jones |
node ids are now generated from a lockable store - no more race conditions
|
| Wed, 03 Apr 2002 11:55:57 +0000 |
Roche Compaan |
Added feature [SF#526730] - search for messages capability
search_indexing-0-4-2-branch
|
| Wed, 03 Apr 2002 05:54:31 +0000 |
Richard Jones |
Fixed serialisation problem...
|
| Wed, 27 Feb 2002 03:40:59 +0000 |
Richard Jones |
Ran it through pychecker, made fixes
|
| Mon, 25 Feb 2002 14:34:31 +0000 |
Engelbert Gruber |
use blobfiles in back_anydbm which is used in back_bsddb.
|
| Sat, 16 Feb 2002 09:14:17 +0000 |
Richard Jones |
[SF#514854] History: "User" is always ticket creator
|
| Wed, 06 Feb 2002 04:05:55 +0000 |
Richard Jones |
Brought the config branch up to date with HEAD
config-0-4-0-branch
|
| Tue, 22 Jan 2002 07:21:13 +0000 |
Richard Jones |
fixed back_bsddb so it passed the journal tests
|
| Tue, 22 Jan 2002 05:18:38 +0000 |
Roche Compaan |
last_set_entry was referenced before assignment
|
| Tue, 22 Jan 2002 05:06:09 +0000 |
Roche Compaan |
We need to keep the last 'set' entry in the journal...
|
| Mon, 21 Jan 2002 16:33:20 +0000 |
Roche Compaan |
You can now use the roundup-admin tool to pack the database
|
| Fri, 18 Jan 2002 04:32:04 +0000 |
Richard Jones |
Rollback was breaking...
|
| Mon, 14 Jan 2002 02:20:15 +0000 |
Richard Jones |
changed all config accesses...
|
| Wed, 02 Jan 2002 02:31:38 +0000 |
Richard Jones |
Fix breakage caused by transaction changes.
|
| Tue, 18 Dec 2001 15:30:34 +0000 |
Roche Compaan |
Fixed bugs:
|
| Mon, 17 Dec 2001 03:52:48 +0000 |
Richard Jones |
Implemented file store rollback.
|
| Sun, 16 Dec 2001 10:53:38 +0000 |
Richard Jones |
take a copy of the node dict...
|
| Fri, 14 Dec 2001 23:42:57 +0000 |
Richard Jones |
yuck, a gdbm instance tests false :(
|
| Wed, 12 Dec 2001 03:23:14 +0000 |
Richard Jones |
Cor blimey this anydbm/whichdb stuff is yecchy.
|
| Wed, 12 Dec 2001 02:30:51 +0000 |
Richard Jones |
I fixed the problems with anydbm using the dbm module at the backend.
|
| Mon, 10 Dec 2001 22:20:01 +0000 |
Richard Jones |
Enabled transaction support in the bsddb backend.
|
| Sun, 02 Dec 2001 05:06:16 +0000 |
Richard Jones |
Eliminate database close method by using weakrefs.
|
| Sat, 01 Dec 2001 07:17:50 +0000 |
Richard Jones |
Database transactions.
|
| Wed, 21 Nov 2001 02:34:18 +0000 |
Richard Jones |
Added a target version field to the extended issue schema
|
| Tue, 09 Oct 2001 23:58:10 +0000 |
Richard Jones |
Moved the data stringification up into the hyperdb.Class class's...
|
| Tue, 09 Oct 2001 07:25:59 +0000 |
Richard Jones |
Added the Password property type.
|
| Sat, 29 Sep 2001 13:27:00 +0000 |
Richard Jones |
CGI interfaces now spit up a top-level index of all instances they can serve.
|
| Sun, 12 Aug 2001 06:32:36 +0000 |
Richard Jones |
using isinstance(blah, Foo) now instead of isFooType
|
| Tue, 07 Aug 2001 00:24:43 +0000 |
Richard Jones |
stupid typo
|
| Tue, 07 Aug 2001 00:15:51 +0000 |
Richard Jones |
Added the copyright/license notice to (nearly) all files...
|
| Mon, 30 Jul 2001 01:41:36 +0000 |
Richard Jones |
Makes schema changes mucho easier.
|
| Wed, 25 Jul 2001 01:23:07 +0000 |
Richard Jones |
Added the Roundup spec to the new documentation directory.
|
| Mon, 23 Jul 2001 08:20:44 +0000 |
Richard Jones |
Moved over to using marshal in the bsddb and anydbm backends.
|
| Mon, 23 Jul 2001 07:22:13 +0000 |
Richard Jones |
*sigh* some databases have _foo.so as their underlying implementation.
|
