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 serhiy.storchaka
Recipients Jonathan Kamens, Paolo Veglia, benjamin.peterson, jmadden, marcjofre, martin.panter, pitrou, pje, serhiy.storchaka, stutzbach, tarek
Date 2016-06-02.19:21:17
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1464895278.03.0.516887927231.issue24291@psf.upfronthosting.co.za>
In-reply-to
Content
It looks to me that there is two opposite ways to resolve this issue.

1. Document the requirement for write() and raise an exception if write() returned not the size of full data. But I think it would be more user-friendly first emit a warning and continue calling write() until full data is written.

2. Don't document any requirement, but just silently handle partial writing.

wsgiref-only.patch contains parts from both ways. It documents new requirement, but doesn't deprecate partial writing. What way is chosen?
History
Date User Action Args
2016-06-02 19:21:18serhiy.storchakasetrecipients: + serhiy.storchaka, pje, pitrou, benjamin.peterson, tarek, stutzbach, martin.panter, jmadden, Jonathan Kamens, marcjofre, Paolo Veglia
2016-06-02 19:21:18serhiy.storchakasetmessageid: <1464895278.03.0.516887927231.issue24291@psf.upfronthosting.co.za>
2016-06-02 19:21:18serhiy.storchakalinkissue24291 messages
2016-06-02 19:21:17serhiy.storchakacreate