Is it just me, or is Sqlite3 goofy?

Magnus Lycka lycka at carmen.se
Mon Sep 11 10:25:44 CEST 2006


mensanator at aol.com wrote:
> I think an explanation of how Sqlite3 differs from SQL and
> a better set of examples is still warranted.

In general, Python standard library modules that are wrappers
for third party libraries are very thinly documented, and they
should probably remain that way, because it's really too much of
a burden on the Python developers to develop this documentation
and keep it up to date. The idea is to document the wrapper, not
the wrapped library.

If I had a choice concerning these wrapper libraries, I'd much
rather see more docs on the tkinter and xml libraries. There you
need to guess a lot. There is no shortage of Tcl/Tk docs, but it
doesn't look the same within Python. For the Python xml libraries,
I've had to experiment a lot, and I have this nagging feeling
that I don't do things the way I should. (From Python 2.5, we
have ElementTree, which is much better from this perspective,
but we've just started using 2.4 at work, and we try to keep
the amount of third party libraries to a minimum here.)

It seems to me that the sqlite3 module is fairly decent in
this regard, particularly since it's not even included in a
completed Python release yet. Concerning the API, I'm surprised
to see magic method naming such as __conform__ introduced in
a library like that. It seems to me that this is a violation
of PEP 8. I'm sure there are further details that could be worth
mentioning in the docs, but I think it's important that we don't
try to duplicate the SQLite docs in the Python docs.



More information about the Python-list mailing list