devtools/fix_po_encoding
author Denis Laxalde <denis.laxalde@logilab.fr>
Wed, 26 Feb 2014 14:50:13 +0100
branchstable
changeset 9590 048d6de92b63
parent 0 b97547f5f1fa
permissions -rwxr-xr-x
Make EditController edit_entity method always return an eid In cases the entity was to be created or copied, eid was None hence the method returned None despite the dosctring says the method should always return an eid. Now if the eid variable is None, it is assigned to the newly created entity eid. Closes #3593029.

#!/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)