Author steve.dower
Recipients loewis, pitrou, steve.dower, tim.golden, twouters, vstinner, zach.ware
Date 2015-03-17.15:29:32
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1426606172.65.0.389438839251.issue23685@psf.upfronthosting.co.za>
In-reply-to
Content
I don't think we should be using PyMODINIT_FUNC for builtin modules, since that will make the init functions publicly available from python35.dll. That said, I do like being able to be consistent here... can we define PyMODINIT_FUNC differently when building pythoncore? That at least would keep the definitions in the same place if we ever change it (and since there's an open PEP regarding extension modules, it doesn't seem impossible).

The change for pyexpat looks right, since that's an external module. I have no ideas why PyMODINIT_FUNC wouldn't always be defined there.
History
Date User Action Args
2015-03-17 15:29:32steve.dowersetrecipients: + steve.dower, loewis, twouters, pitrou, vstinner, tim.golden, zach.ware
2015-03-17 15:29:32steve.dowersetmessageid: <1426606172.65.0.389438839251.issue23685@psf.upfronthosting.co.za>
2015-03-17 15:29:32steve.dowerlinkissue23685 messages
2015-03-17 15:29:32steve.dowercreate