test_httplib fails on the mac

[jackjansen]

BPO	661340
Nosy	@jackjansen

^{Note: these values reflect the state of the issue at the time it was migrated and might not reflect the current state.}

Show more details

GitHub fields:

assignee = None
closed_at = <Date 2003-01-23.18:08:28.000>
created_at = <Date 2003-01-02.22:13:45.000>
labels = ['interpreter-core']
title = 'test_httplib fails on the mac'
updated_at = <Date 2003-01-23.18:08:28.000>
user = 'https://github.com/jackjansen'

bugs.python.org fields:

activity = <Date 2003-01-23.18:08:28.000>
actor = 'jhylton'
assignee = 'jhylton'
closed = True
closed_date = None
closer = None
components = ['Interpreter Core']
creation = <Date 2003-01-02.22:13:45.000>
creator = 'jackjansen'
dependencies = []
files = []
hgrepos = []
issue_num = 661340
keywords = []
message_count = 3.0
messages = ['13822', '13823', '13824']
nosy_count = 2.0
nosy_names = ['jhylton', 'jackjansen']
pr_nums = []
priority = 'high'
resolution = 'fixed'
stage = None
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue661340'
versions = ['Python 2.3']

[jackjansen]

test_httplib fails on MacPython-OS9. This is because the output contains \r and \n characters, and this fails on MacOS9.

There are two problems really:

1. output/test_httplib is checked in as a text file, while it is a binary file (it contains some \r\n data) or even (brrr:-) mixed-mode (it also contains \n lineendings).
2. The output file is opened in text mode (when re-reading the data), so this can probably not be fixed by a simple cvs admin -kb.

I think that putting repr()s in the output in stead of raw strings is probably good enough, but you probably know better what is going on here.

[jhylton]

Logged In: YES
user_id=31392

I think the actual line-endings are uninteresting as far as
what the test is trying to do. I've changed the test so
that it doesn't contain \r\n pairs. I think the right thing
is to change the way debugging output is generated by
httplib, but that's a big project. So I'll just change the
test to capture stdout and clean it before printing it.

(Checking coming as soon as sourceforge responds.)

[jhylton]

Logged In: YES
user_id=31392

Fixed in rev 1.10 of test_httplib.py