This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author taleinat
Recipients georg.brandl, larry, loewis, nadeem.vawda, serhiy.storchaka, taleinat
Date 2015-05-16.11:04:48
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <>
Attached in a conversion patch for Modules/_hashopenssl.c.

Since it appear that zlib has already been converted, that's the last file to convert in this batch!

All tests pass on my OSX 10.10 (after running "touch Modules/hashlib.h" to get make to recompile things that use _hashopenssl).


In HASH.__init__, the "name" parameter must be a string. However, the current code accepts any object and checks something else before checking whether the given object is a string. Therefore, changing the type from "object" to "str" would slightly change the parameter checking behavior. For now I've left it as it was. Also, since HASH.__new__ mirrors this, I've left it as-is as well.

I slightly changed the first line of the HASH object's doc-string, to make it fit in one line. It was: "A hash represents the object used to calculate a checksum of a string of information." I've changed the beginning to "A hash is an object used to calculate ..."

pbkdb_hmac accepts an optional "dklen" parameter which may be either a long or None. I left this as an "object", since long(accept={int, NoneType}) gives "long_converter: default value None for field dklen_obj is not of type int".
Date User Action Args
2015-05-16 11:04:50taleinatsetrecipients: + taleinat, loewis, georg.brandl, larry, nadeem.vawda, serhiy.storchaka
2015-05-16 11:04:50taleinatsetmessageid: <>
2015-05-16 11:04:50taleinatlinkissue20182 messages
2015-05-16 11:04:49taleinatcreate