cubicweb/devtools/fix_po_encoding
author Denis Laxalde <denis.laxalde@logilab.fr>
Wed, 29 Mar 2017 11:45:19 +0200
branch3.25
changeset 12112 0a54f49314f6
parent 11057 0b59724cb3f2
child 12779 44147dab9d27
permissions -rwxr-xr-x
[server] Introduce an `has_scheduler` method on Repository This is to be used by client application to determine if looping tasks may be registered in the current process. By checking this, one will avoid the warning in looping_task method when the repository has no scheduler.

#!/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)