True, we can potentially see those sorts of problems where import is used to test for a feature. But it's worth it for getting it into trunk and making ctypes a clearly achievable goal for 2.5.2. Thanks so much for your work!

- Jim

On Tue, Feb 2, 2010 at 6:44 PM, Wayne Meissner <> wrote:
On 3 February 2010 11:23, Philip Jenvey <> wrote:
> On Jan 30, 2010, at 2:40 AM, Wayne Meissner wrote:
>> Is there any reason I shouldn't merge the ctypes work I've done into
>> trunk?  Its far (very, very far) from complete, but some basic stuff
>> (scalar params/return types) is working, and I think it would be
>> better to have it in trunk where others can hack on it, rather than
>> having it languish in a branch where it only gets hit'n'run attention.
> +1, we'll just make sure the unit tests are disabled so the buildbots are happy

Ok, I merged it in.  In Theory[tm], the only things that might break,
is something that imported ctypes and had a case of either
fully-working, or not-available-at-all - given its only partially
working, those types of tests may fail now.

All the low level jffi stuff was already in trunk, and only needed a
minor change to add some apis I used in ctypes.

The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
Jython-dev mailing list

Jim Baker