On 04/28/2016 08:09 PM, Mike Bonnet wrote:
Hi! As an exercise, I looked in to what it would take to port Koji from PyGreSQL to psycopg2. Turns out, it didn't take much at all:
Hurray for PEP 249!
Well, actually the spec could be better/clearer, but at least it's there ;)
https://pagure.io/fork/mikeb/koji/c/676db88a8bd4c729278ef4594c9174be5b554b2f
Now, do we think this is a good idea? I haven't done any performance testing, but in general it seems like pycopg2 is better supported. Thoughts?
Googling psycopg2 vs pygresql gives a number of hits, but not very helpful.
There is a licensing difference. PyGreSQL is MIT, psychopg2 is GPL (though one post describes it as "a bizarre hacked-up license"). https://github.com/psycopg/psycopg2/blob/master/LICENSE
A number of posts talk about PyGreSQL being unmaintained, but this no longer seems to be the case. Their svn has been pretty active the last couple years. Otoh, one post complains that the psycopg2 website (though not the code) is unmaintained. http://it.toolbox.com/blogs/database-soup/postgres-needs-a-new-python-driver...
PyGreSQL doesn't support all postgres data types (or didn't?). We're used to that, though.