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 rbcollins
Recipients berker.peksag, michael.foord, pkoning, python-dev, quasipedia, rbcollins
Date 2015-07-23.13:42:41
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1437658962.07.0.622051144845.issue21750@psf.upfronthosting.co.za>
In-reply-to
Content
@Paul

So the problem is that its never been a high fidelity thing in that sense.... In that:
3.3 -> read() is a constant for the thing opened from the mock
3.4 -> read() works once and only once across all opened files from the mock
3.5 today -> read() works once for *each* opened file from the mock, but you *can't access* each file object (because the mock.returnvalue is replaced each time, which is unmocklike)

With this patch: -> read() works once for each opened file, as long as the sequence open -> read -> open -> read is followed, and mock.returnvalue is a constant, mocklike.
History
Date User Action Args
2015-07-23 13:42:42rbcollinssetrecipients: + rbcollins, michael.foord, python-dev, quasipedia, berker.peksag, pkoning
2015-07-23 13:42:42rbcollinssetmessageid: <1437658962.07.0.622051144845.issue21750@psf.upfronthosting.co.za>
2015-07-23 13:42:42rbcollinslinkissue21750 messages
2015-07-23 13:42:41rbcollinscreate