[Python-Dev] PEP 269 versus 283.

[Greg Ewing]

Jeremy Hylton <jeremy@alum.mit.edu>:

> I lobbied in favor of PEP 269 earlier, because it is mostly exposing
> functionality that already exists but is difficult to reuse.  That
> seems like a good thing even if some people want to use other parser
> generators.

There's a downside that you'd then be committed to supporting
it, even if Python stopped using pgen itself some time in the
future.

If that's not a worry, then fine -- just pointing out that
exposing previously unexposed functionality isn't necessarily
without cost.

Greg Ewing, Computer Science Dept, +--------------------------------------+
University of Canterbury,	   | A citizen of NewZealandCorp, a	  |
Christchurch, New Zealand	   | wholly-owned subsidiary of USA Inc.  |
greg@cosc.canterbury.ac.nz	   +--------------------------------------+