Remove cl usage from aifc

[brettcannon]

BPO	2847
Nosy	@brettcannon, @birkenfeld
Files	aifc_3.0.patch

^{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 = 'https://github.com/birkenfeld'
closed_at = <Date 2008-06-07.19:02:22.225>
created_at = <Date 2008-05-14.00:57:13.662>
labels = ['library', 'release-blocker']
title = 'Remove cl usage from aifc'
updated_at = <Date 2008-06-07.19:02:22.186>
user = 'https://github.com/brettcannon'

bugs.python.org fields:

activity = <Date 2008-06-07.19:02:22.186>
actor = 'georg.brandl'
assignee = 'georg.brandl'
closed = True
closed_date = <Date 2008-06-07.19:02:22.225>
closer = 'georg.brandl'
components = ['Library (Lib)']
creation = <Date 2008-05-14.00:57:13.662>
creator = 'brett.cannon'
dependencies = []
files = ['10506']
hgrepos = []
issue_num = 2847
keywords = ['patch']
message_count = 5.0
messages = ['66808', '67496', '67611', '67650', '67814']
nosy_count = 3.0
nosy_names = ['brett.cannon', 'georg.brandl', 'quentin.gallet-gilles']
pr_nums = []
priority = 'release blocker'
resolution = 'accepted'
stage = None
status = 'closed'
superseder = None
type = None
url = 'https://bugs.python.org/issue2847'
versions = ['Python 3.0']

[brettcannon]

The cl module has been removed from Python 3.0, but the aifc module still
imports it in multiple locations. The module needs to be updated (with
tests hopefully) so as to not use the cl module.

[quentingallet-gilles]

I'm working on this one.

[quentingallet-gilles]

The attached patch removes all cl references from the aifc module and
updates it to make it usable in 3.0. It also supports more compression
types because audioop has been enhanced throughout the years (i.e.
support for alaw encoding since 2.5). The changes are the following :

• replace strings with bytes for frames read from an aiff/aifc file.
• replace / by // to have integer division back
• added ulaw (as an alternative of ULAW) and alaw/ALAW compression
  because audioop supports them (I've tested them successfully, using the
  samples from this URL :
  http://www-mmsp.ece.mcgill.ca/Documents/AudioFormats/AIFF/Samples.html )
• removed all tests when trying to import audioop : they can't fail anymore.
• a bit of PEP-8 cleanup

I'm not an AIFF/AIFC expert, so criticism is welcome!

[quentingallet-gilles]

Updated the patch with the following corrections/improvements :

• corrected a missed str -> bytes
• replace % formatting occurrences with str.format()
• more PEP-8 conformance

[birkenfeld]

I found a few more bytes/str problems, committed patch in r64023.