On Jun 16, 2009, at 11:51 PM, Michele Petrazzo - Unipex wrote:
First of all, thanks for the module.
Two questions:
- I look around, but I don't find a reply to the question if you
want to
"backport" the py-postgresql to the python 2.X line. Have us some hope
to use it in our "old" projects?
Well, there is pg_proboscis that py-postgresql was based on.
It is quite lacking by comparison, but functional.
But no, I don't plan to backport it.
- Making some tests with a test db, I see that, like the other
python-pg
drivers, the cur.execute("select * from table"), load all the data in
memory. Can be useful, if possible of course, to not use this lot of
ram? With psycopg2 devels can use the "named cursor" for solve. Any
chances to add this type of feature?
It shouldn't be loading it all into memory by default... If it is,
it's a bug.
However, It should be grabbing some pretty big chunks when it needs
more data to supply a fetchXXX request, so if you're looking at a wide
result set, that could use a large amount of memory.
_______________________________________________
DB-SIG maillist - DB-SIG@python.org
http://mail.python.org/mailman/listinfo/db-sig
