[Python-Dev] constness of interpreter data

"Martin v. Löwis" martin at v.loewis.de
Mon Sep 8 08:35:18 CEST 2008

> So, er, is this a remotely sane thing to be doing, and does anyone have
> suggestions? :)

Not for 2.6/3.0 - but after that: sounds good.

OTOH, I don't see a clear technical *need* for these data to be const.
In general, const-ness helps correctness and sharing across processes,
and the cases you provide are correct and provide sharing even without
being const.


More information about the Python-Dev mailing list