devtools/fix_po_encoding
author Julien Jehannet <julien.jehannet@logilab.fr>
Fri, 28 Jan 2011 16:08:40 +0100
branchstable
changeset 6919 8fd6921f3e7c
parent 0 b97547f5f1fa
permissions -rwxr-xr-x
[selectors] modify workflow selectors: is_in_state, on_transition - factorize `is_on_state` selector - add new `on_transition` selector Especially useful to match pending transitions to enable notifications when your workflow allows several transition to the same states. Note that if workflow `change_state` adapter method is used, this selector will not be triggered. In debug mode: These both selectors will check against the entity current workflow if expected values given in selector argument are valid. ValueError exception will be raised for unmatching state/transition names against the current workflow (generic etype workflow). (check against custom workflow is not implemented)

#!/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)