Py3K compatibility #88
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
We use this to shim up some of the modules that changed locations between Python 2 and Python 3.
Breakage warning: this won't work on Python versions < 2.6.
To allow Python 3 compatibility, Django 1.5 renamed smart_unicode in django.utils.encoding as smart_text (as Python 3's str type is unicode by default). To accomodate running on earlier Django versions, we do an import of the old name; if we fail, we import smart_text as smart_unicode instead (with the assumption that Django is actually in the path).
urlparse in Python 3 has been subsumed into urllib; to support both Python 2 and Python 3, we import urlparse via Six. This commit also fixes a bug in the tests. There is an incorrect assumption that parse_qs parses just the query string of a full URL, when given a full URL; it doesn't, and instead parse_qs assumes the caller passes in just the query string part of the URL. When given a full URL, the rest of the URL including the first attribute in the query string is assigned as the first key of the resulting dictionary returned by parse_qs. Because the 'code' key incidentally ends up being the second attribute in the query string, the test passes. However, because there was a change in the ordering of the query string, no 'code' key could be found when running in Python 3.
Because Python 3 is strict when it comes to byte strings vs. text strings, and because Python 3 does not do any automatic coercion between the two types, we need to be more explicit. In particular, when using response.content (which is in really a byte string), we need to convert it to a particular character encoding before treating it as a text string. We assume UTF-8 for the tests, which *will* break in other encodings, but since these are the tests, we don't mind assuming UTF-8. As well, because encode() in the str type in Python 3 no longer supports non-character encodings such as Base64, we need to use codecs to do the conversion. We also fix the assumption that shortuuid.uuid() returns a byte string (which it doesn't) when passing to hashlib.
This breaks in Python 3, as modifying a dictionary while iterating through its keys raises an exception, because dict.keys() now returns a view instead of a materialized list. We instead pull out the session keys as a list and iterate through that instead, which works for both Python 3 and Python 2 (although in Python 2 this creates an additional list).
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit. This suggestion is invalid because no changes were made to the code. Suggestions cannot be applied while the pull request is closed. Suggestions cannot be applied while viewing a subset of changes. Only one suggestion per line can be applied in a batch. Add this suggestion to a batch that can be applied as a single commit. Applying suggestions on deleted lines is not supported. You must change the existing code in this line in order to create a valid suggestion. Outdated suggestions cannot be applied. This suggestion has been applied or marked resolved. Suggestions cannot be applied from pending reviews. Suggestions cannot be applied on multi-line comments. Suggestions cannot be applied while the pull request is queued to merge. Suggestion cannot be applied right now. Please check back later.
Various fixes to allow running on Python 3 as well as Python 2.