[Python-Dev] PEP 3147 ready for pronouncement and merging

Barry Warsaw barry at python.org
Thu Apr 15 03:10:15 CEST 2010

On Apr 15, 2010, at 08:33 AM, Nick Coghlan wrote:

>Brett Cannon wrote:
>> And just a quick suggestion: can we standardize what
>> imp.source_to_path() and friend are supposed to return if the
>> interpreter doesn't support bytecode? I will probably have to rely on
>> that for something so it would be best to say now whether it should be
>> None or raise an exception so there is no divergence on this between VMs.
>Returning None sounds like the most straightforward option. "__cached__
>= None" will just mean "for whatever reason, we have no cached filename
>for this file". It may be the cached file doesn't exist, or the
>interpreter simply wasn't in a position to figure it out in a user
>visible way.

I completely agree.  The PEP already leaves __cached__ up to the
implementation, but I'll update it to be clear that None is an acceptable
return value from imp.cached_from_source() (which is the one I think you
mean), and also what __cached__=None means.

Thanks Brett and Nick.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-dev/attachments/20100414/b0b50d4d/attachment.pgp>

More information about the Python-Dev mailing list