Message8226

Author jeff.allen
Recipients jeff.allen
Date 2014-01-12.23:20:58
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1389568859.2.0.938486969725.issue2109@psf.upfronthosting.co.za>
In-reply-to
Content
I'd like to make minimal changes so we test the same things, but in a way compatible with the BFE on Windows. In view of the experience above, I think this means:

test_request_line_trimming: the test fundamentally involves a malformed request, with a verb the server doesn't support. I propose to change the verb to one the BFE doesn't know either (like 'XYZ').

test_version_none: the intent is to check that a potentially valid verb is rejected when submitted using "no version protocol" (HTTP/0.9), for which only GET should be valid. Options to fix are: 1. choose a verb the server actually supports (and the BFE doesn't know) like 'CUSTOM'; 2. keep 'PUT' but add a Content-Length header. Of these two, option 1. meets the intent better than the original test.
History
Date User Action Args
2014-01-12 23:20:59jeff.allensetmessageid: <1389568859.2.0.938486969725.issue2109@psf.upfronthosting.co.za>
2014-01-12 23:20:59jeff.allensetrecipients: + jeff.allen
2014-01-12 23:20:59jeff.allenlinkissue2109 messages
2014-01-12 23:20:58jeff.allencreate