PyArgs_ParseTupleAndKeywords() returns borrowed references for O format
starting in 2.0.1, prior to that it returns new references, so it's
necessary to test the Python version to determine whether or not we
should Py_INCREF() it. If we always Py_INCREF() it, this produces a
memory leak in versions prior to 2.0.1.
A plus or minus in front of the database name can force transactions on
or off.
Make use of MySQLdb.conversions. Only the SQL -> Python conversions are
actually used by ZMySQLDA.
A minor bug in MySQLdb-0.9.0c1 keeps it from working with this, only
seen with not using mx.DateTime. Fixed in CVS. Will release 0.9.0c2
before doing another ZMySQLDA release.
that name anyway.
Set has been extended somewhat. The comparision operators really only
work properly with Python 2.1, due to the limitations of __cmp__.
Set also uses the binary operators (&, |, ^), since these make somewhat
more sense than the arithmetic ones, though there is no good analog for
- (if only there were a nand operator...)
Bump the version to 0.9.0b3. This is not the actual 0.9.0b3 release yet,
however. I want to do some more insanity checking. But almost ready for
some candidate releases.
All the threading stuff is ripped out and burned.
Too much code for not enough benefit. Still thread-safe,
just don't share connections.
Made a nice Set class for SET columns.
Updated the docs.
