Stowage/cpython
2
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-10-30 21:21:22 +00:00
Code Issues Projects Releases Packages Wiki Activity

Fix gb18030 codec's bug that doesn't map two-byte characters on

Browse source 
GB18030 extension in encoding. (bug reported by Bjorn Stabell)
This commit is contained in:
Hye-Shik Chang 2007-08-04 04:10:18 +00:00
parent 766d880a2f
commit f3e93a0268
4 changed files with 26 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

16
Lib/test/test_multibytecodec_support.py
View file

@ -5,7 +5,7 @@
#
import sys, codecs, os.path
import unittest
import unittest, re
from test import test_support
from StringIO import StringIO
@ -272,6 +272,12 @@ def open_mapping_file(self):
return test_support.open_urlresource(self.mapfileurl)
def test_mapping_file(self):
if self.mapfileurl.endswith('.xml'):
self._test_mapping_file_ucm()
else:
self._test_mapping_file_plain()
def _test_mapping_file_plain(self):
unichrs = lambda s: u''.join(map(unichr, map(eval, s.split('+'))))
urt_wa = {}
@ -303,6 +309,14 @@ def test_mapping_file(self):
self._testpoint(csetch, unich)
def _test_mapping_file_ucm(self):
ucmdata = self.open_mapping_file().read()
uc = re.findall('<a u="([A-F0-9]{4})" b="([0-9A-F ]+)"/>', ucmdata)
for uni, coded in uc:
unich = unichr(int(uni, 16))
codech = ''.join(chr(int(c, 16)) for c in coded.split())
self._testpoint(codech, unich)
def test_mapping_supplemental(self):
for mapping in self.supmaps:
self._testpoint(*mapping)