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.

Title: Test failure: Lib/test/ line 851, "topic?key=def"
Type: behavior Stage: resolved
Components: Versions: Python 3.4
Status: closed Resolution: duplicate
Dependencies: Superseder: 3.4.1rc1 test_pydoc fails: pydoc_data.topics.topics values are type "bytes" not "str"
View: 21431
Assigned To: Nosy List: georg.brandl, larry, ned.deily, r.david.murray, zach.ware
Priority: deferred blocker Keywords:

Created on 2014-09-22 13:14 by larry, last changed 2022-04-11 14:58 by admin. This issue is now closed.

Messages (9)
msg227267 - (view) Author: Larry Hastings (larry) * (Python committer) Date: 2014-09-22 13:14
I get a test failure in the regression test suite.  This appears to be the important bit:

    Traceback (most recent call last):
      File "/tmp/Python-3.4.2rc1/Lib/test/", line 851, in test_url_requests
        self.assertEqual(result, title, text)
    AssertionError: 'Pydoc: Error - topic?key=def' != 'Pydoc: KEYWORD def'
    - Pydoc: Error - topic?key=def
    + Pydoc: KEYWORD def

I can ship 3.4.2rc1 like this, but I'd really like this fixed before 3.4.2 final.

Does anybody own pydoc?  There's no "expert" listed on the Python Experts page.

(Adding you, Georg, because you're the DE.)
msg227268 - (view) Author: Georg Brandl (georg.brandl) * (Python committer) Date: 2014-09-22 13:19
I have no idea about that code, and I can't reproduce the failure.

(Could the buildbots?)
msg227270 - (view) Author: R. David Murray (r.david.murray) * (Python committer) Date: 2014-09-22 13:42
I can't reproduce it either, and none of the failing stable buildbots show this error.  Unfortunately we can only look at tip since we can't see your tag yet.  But I doubt that's the issue...the last commit to pydoc or its tests was on the 17th, and was a one line change in an unrelated area.  Last change before that was in June.
msg227271 - (view) Author: Larry Hastings (larry) * (Python committer) Date: 2014-09-22 13:46
FWIW, 3.4.2rc1 is based on 7af0315bdfe0.  (The release process creates a couple additional changesets.)

The failure is on my laptop, Ubuntu 14.04 x64.
msg227272 - (view) Author: R. David Murray (r.david.murray) * (Python committer) Date: 2014-09-22 13:58
No failure running test_pydoc for me on gentoo linux with that changeset.
msg227286 - (view) Author: Zachary Ware (zach.ware) * (Python committer) Date: 2014-09-22 17:38
R. David Murray wrote (on python-committers):
> So...Larry broke it, but it is not obvious how.  Could it be something
> wrong with the pydoc topics update for the release?

Indeed, it looks like somehow the pydoc-topics update turned all of the values in the topics dict into bytes literals rather than plain strings.

Unfortunately, I don't have the right setup currently to look into it further, but my best guess would be that perhaps the pydoc-topics Sphinx builder needs an update for Python 3?
msg227287 - (view) Author: Georg Brandl (georg.brandl) * (Python committer) Date: 2014-09-22 17:42
That is very likely the reason. So far nobody has run the builder with Python 3.
msg227288 - (view) Author: Ned Deily (ned.deily) * (Python committer) Date: 2014-09-22 17:44
Duplicate of Issue21431?
msg227289 - (view) Author: Georg Brandl (georg.brandl) * (Python committer) Date: 2014-09-22 18:01
Date User Action Args
2022-04-11 14:58:08adminsetgithub: 66651
2014-09-23 02:22:33berker.peksagsetstage: needs patch -> resolved
2014-09-22 18:01:32georg.brandlsetstatus: open -> closed
superseder: 3.4.1rc1 test_pydoc fails: pydoc_data.topics.topics values are type "bytes" not "str"
resolution: duplicate
messages: + msg227289
2014-09-22 17:44:49ned.deilysetnosy: + ned.deily
messages: + msg227288
2014-09-22 17:42:03georg.brandlsetmessages: + msg227287
2014-09-22 17:38:47zach.waresetnosy: + zach.ware
messages: + msg227286
2014-09-22 13:58:28r.david.murraysetmessages: + msg227272
2014-09-22 13:46:55larrysetmessages: + msg227271
2014-09-22 13:42:11r.david.murraysetnosy: + r.david.murray
messages: + msg227270
2014-09-22 13:19:05georg.brandlsetmessages: + msg227268
2014-09-22 13:14:17larrycreate