cubicweb/devtools/fix_po_encoding
author Sylvain Thénault <sylvain.thenault@logilab.fr>
Fri, 06 Oct 2017 16:59:10 +0200
changeset 12206 cfd25da225c2
parent 11057 0b59724cb3f2
child 12779 44147dab9d27
permissions -rwxr-xr-x
[test] Pin some test dependencies We currently have CI failures because cubes used as test dependencies have been updated to new-style cube layout. To avoid this, pin them to previous released. Those dependencies should be removed but in the mean time this should be enough (and backported in all active branches).

#!/usr/bin/python

"""usage: fix-po-encodings [filename...]
change the encoding of the po files passed as arguments to utf-8
"""
import sys
import re
import codecs

def change_encoding(filename, target='UTF-8'):
    fdesc = open(filename)
    data = fdesc.read()
    fdesc.close()
    encoding = find_encoding(data)
    if encoding == target:
        return
    data = fix_encoding(data, target)
    data = unicode(data, encoding)
    fdesc = codecs.open(filename, 'wb', encoding=target)
    fdesc.write(data)
    fdesc.close()

def find_encoding(data):
    regexp = re.compile(r'"Content-Type:.* charset=([a-zA-Z0-9-]+)\\n"', re.M)
    mo = regexp.search(data)
    if mo is None:
        raise ValueError('No encoding declaration')
    return mo.group(1)

def fix_encoding(data, target_encoding):
    regexp = re.compile(r'("Content-Type:.* charset=)(.*)(\\n")', re.M)
    return regexp.sub(r'\1%s\3' % target_encoding, data)
    


for filename in sys.argv[1:]:
    print filename
    change_encoding(filename)