[Skunkweb-list] Re-2: Problem using PyDO 2.0 with SQLite (Python 2.4 + pysqlite2)

[<t....@ar...>]

Hello, Jacob!
thank's for the really fast response!

> I suppose this is a problem connected with pysqlite2 versions.  I happen to
> have 2.2.2 on the machine I'm sitting at right now, and here I can do this:
> 
>   smulloni@bracknell ~ $ ipython
> 
>   [1]:from pysqlite2 import dbapi2 as sqlite
> 
>   [2]:sqlite.converters
>   [2] {'date': <function convert_date at 0xb7ad048c>, 'timestamp': <
> function
>   convert_timestamp at 0xb7ad04c4>}
> 
>   [3]:sqlite.version
>   [3] '2.2.2'
>   
> Could you tell me what happens when you do the above?

Yes, your'e right:
 
*** Python 2.4.3 (#69, Apr 11 2006, 15:32:42) [MSC v.1310 32 bit (Intel)] on win32. ***
>>> from pysqlite2 import dbapi2 as sqlite
>>> sqlite.converters
{'DATE': <function convert_date at 0x01B6C430>, 'TIMESTAMP': <function convert_timestamp at 0x01B6C470>}
>>> sqlite.version
'2.3.3'
>>> 

obviously 2.3.3 uppercases the names automatically at registration!
Do you think it makes sense to patch PyDO or should I "downgrade" the sqlite lib ?

In response to your very detailed thoughts on the future of PyDO, well: 
coming from TurboGears - where it was integrated out of the box - I tried SQLObject: was nice, but then I ran into serious troubles with it's internal caching, so my multiuser client application didn't work properly (ending up in an academic discussion about "why an ORM library must do such chaching");
I tried SQLAlchemy - and it felt a bit clumsy; (can't remember the second point I decided to let it go, maybe it was SA that had the caching problem and not SO ?)

I like PyDO because its small, fast, easy to use, and one can do so much (with inheritance, projection, joins, "fetch" ...) that perhaps is not all real ORM, but helps a lot to do the things I need fast and easy.

Thanks for that!
Thomas