Hi. I've been working on some refactoring of the Python bindings for sqlite. I now have a working Python binding for sqlite 3 which is fairly different than the bindings for sqlite 2. [...] I've been doing development in our Conary CVS repository. You can browse the repository here:
http://cvs.specifixinc.com/viewcvs/viewcvs.cgi/conary/pysqlite3/
I'm not trying to hijack the work by Gerhard - but our project needed new features that sqlite 3 brings, and we couldn't wait for a few months while new support was added.
I welcome feedback, and am more than happy to throw this work away if a better Python binding for sqlite emerges.
Nice work, Matt :-)
As far as I could see in a glance, you based your work on the current PySQLite codebase. What I currently have here is a proof-of-concept of a PySQLite rewrite, completely in C. Most design decisions are already formed in my head, and only need implementation ;-) The benefits compared to the current PySQLite would be significant improvements in performance and memory consumption and a lean and mean API.
From this proof-of-concept on I planned to develop the next generation of PySQLite. Perhaps I could reuse some of your code as well.
Btw. as I'm currently unemployed I would accept sponsorship of the PySQLite rewrite and could help with other Python or Python extension programming projects :-)
-- Gerhard
