Backport r59049:
Fix for #1444: utf_8_sig.StreamReader was (indirectly through decode())
calling codecs.utf_8_decode() with final==True, which falled with incomplete
byte sequences. Fix and test by James G. Sack.
diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index aacf19f..5233434 100644
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -435,6 +435,50 @@
s = u"spam"
self.assertEqual(d.decode(s.encode("utf-8-sig")), s)
+ def test_stream_bom(self):
+ unistring = u"ABC\u00A1\u2200XYZ"
+ bytestring = codecs.BOM_UTF8 + "ABC\xC2\xA1\xE2\x88\x80XYZ"
+
+ reader = codecs.getreader("utf-8-sig")
+ for sizehint in [None] + range(1, 11) + \
+ [64, 128, 256, 512, 1024]:
+ istream = reader(StringIO.StringIO(bytestring))
+ ostream = StringIO.StringIO()
+ while 1:
+ if sizehint is not None:
+ data = istream.read(sizehint)
+ else:
+ data = istream.read()
+
+ if not data:
+ break
+ ostream.write(data)
+
+ got = ostream.getvalue()
+ self.assertEqual(got, unistring)
+
+ def test_stream_bare(self):
+ unistring = u"ABC\u00A1\u2200XYZ"
+ bytestring = "ABC\xC2\xA1\xE2\x88\x80XYZ"
+
+ reader = codecs.getreader("utf-8-sig")
+ for sizehint in [None] + range(1, 11) + \
+ [64, 128, 256, 512, 1024]:
+ istream = reader(StringIO.StringIO(bytestring))
+ ostream = StringIO.StringIO()
+ while 1:
+ if sizehint is not None:
+ data = istream.read(sizehint)
+ else:
+ data = istream.read()
+
+ if not data:
+ break
+ ostream.write(data)
+
+ got = ostream.getvalue()
+ self.assertEqual(got, unistring)
+
class EscapeDecodeTest(unittest.TestCase):
def test_empty(self):
self.assertEquals(codecs.escape_decode(""), ("", 0))