[ZODB-Dev] ZODB on ironPython?!

[Jim Fulton]

On Wed, Oct 26, 2011 at 2:09 PM, Tamer Higazi <th982a at googlemail.com> wrote:
> dear Jim,
> Thank your for your quick answer.
>
> Am 26.10.2011 19:34, schrieb Jim Fulton:
>> On Wed, Oct 26, 2011 at 12:59 PM, Tamer Higazi <th982a at googlemail.com> wrote:
>>> Hey Jim!
>>> Hey just cool down.
>>
>> /me cools down
>>
>>> I love the ZODB database, and this toy is just
>>> beautiful. I iam willing to offer myself to support you porting the
>>> database in pure python, if you need volunteers.
>>
>> Thanks.  I'll keep that in mind.
>>
>>> I really didn't mean, to come anybody stupid.
>>
>> I hope you don't mind me playfully amplifying your !s. :)
>>
>>> I am just thinking porting
>>> a customer project from python to iron-python where ZODB plays a role.
>>
>> If this is a customer project, I'd recommend dropping either
>> iron python or ZODB at this stage.
> Why that?! As I know, ironpython creates with the JIT compiler bytecode
> in realtime.

Because this is going to be pretty new code and you'll be a trail blazer.

>>> Is there much work to do?!
>>
>> Not sure. I suspect not.  Tres created a Python implementation of
>> persistence and timestamp.   I've got a BTrees patch to review.
>>
>> Keep in mind that these will run a lot slower than what you're used to.
>> If a performance is an issue, you may find yourself wanting to write
>> c# versions of these.
>>
> what about pypy ?!

It will be interesting to see how much of the performance pypy
can gain back.

Jim

-- 
Jim Fulton
http://www.linkedin.com/in/jimfulton