Tinkering with ElementTree encoding detection

etree-eval.py

# -*- coding: utf-8 -*-
from xml.etree import ElementTree as ET
import itertools
import StringIO
import traceback

XML=r"""<?xml version='1.0' encoding='%s'?><response><msg>АБВГДЕЖЗСКЛМНОПРСТУФХЦЧШЩЫЪЭЬЮЯ</msg></response>"""
# XXX: for some unknown reason Ё isn't supported


PARSING_STYLE = ["string", "file"]
BYTE_ENCODING = ["unicode", "utf-8", "windows-1251"]
PROLOG_ENCODING = [u"utf-8", u"windows-1251"]

if __name__ == "__main__":
    r = itertools.product(PARSING_STYLE, BYTE_ENCODING, PROLOG_ENCODING)
    for parse, byte_enc, prolog_enc in r:
    	result = "OK"
    	xml = unicode(XML, "utf-8") % (prolog_enc)
    	if byte_enc != u"unicode":
    		xml = xml.encode(byte_enc)

    	if parse == "string":
    		try:
    			doc = ET.fromstring(xml)
    			#print doc.find("msg").text
    		except:
    			result = "F" # traceback.format_exc()
    	elif parse == "file":
    		try:
    			doc = ET.parse(StringIO.StringIO(xml))
    		except:
    			result = "F" # traceback.format_exc()

    	print "%s %s %s: %s" % (parse, byte_enc, prolog_enc, result)