From: Alan van den Bosch
Date: October 31 2000 3:32am
Subject: ha_berkeley.[ch] files
List-Archive: http://lists.mysql.com/internals/84
Message-Id: <00103114321504.13992@odin>
MIME-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

Hi,

I read in one of the mysql mailling lists a message regarding transaction=
=20
support posted by Monty. IIRC the message was posted around 2000-10-11 an=
d=20
suggested that transaction support would be available in downloadable=20
binaries (at that time) in two to four weeks, this was due to the sleepyc=
at=20
tables always requiring a primary key and some heavy inhouse testing.

I Downloaded the src tarballs for mysql-3.23.27-beta-win-src.zip and=20
db-3.2.3b.tar.gz had had a poke around slq/ha_berkeley.[ch] and the db.h=20
files. I've the impression that transaction support has been successfully=
=20
compiled and tested to a degree. However, there are references to a membe=
r of=20
the DB type (__DB struct) that doesnt exist in that structure, but is a=20
member of the DBT type ( does app_private ring any bells? ). An expample =
of=20
this is the berkley_cmp_packed_key() function in ha_berkeley.c

Is there a more recent version of the ha_berkeley.[ch] files that will=20
compile ? I've tried to use bitkeeper to get the dev source tree, followi=
ng=20
the instructions in the mysql manual, and ended up with lots of revision=20
control/log related files - but none of the source files. I've also tried=
 the=20
sourceforge CVS repository but it appears not to be up and running (acces=
s=20
denied) for joe public.


--=20
Regards,

Alan van den Bosch
Sanguis Pty Ltd


/* All generalizations are false. */