Patch #681152: Support escaped Unicode characters in classes. Fixes #612074.

loewis · loewis · commit 53d93adc46c7 · 2003-04-19T08:37:24.000Z
Will backport to 2.2.
diff --git a/Lib/sre_parse.py b/Lib/sre_parse.py
@@ -254,7 +254,7 @@ def _class_escape(source, escape):
             if len(escape) != 2:
                 raise error, "bogus escape: %s" % repr("\\" + escape)
             return LITERAL, atoi(escape, 16) & 0xff
-        elif str(escape[1:2]) in OCTDIGITS:
+        elif escape[1:2] in OCTDIGITS:
             # octal escape (up to three digits)
             while source.next in OCTDIGITS and len(escape) < 5:
                 escape = escape + source.get()
diff --git a/Lib/test/test_sre.py b/Lib/test/test_sre.py
@@ -96,6 +96,10 @@ def test(expression, result, exception=None):
 # non-simple '*?' still recurses and hits the recursion limit
 test(r"""sre.search('(a|b)*?c', 10000*'ab'+'cd').end(0)""", None, RuntimeError)
 
+# bug 612074
+pat=u"["+sre.escape(u"\u2039")+u"]"
+test(r"""sre.compile(pat) and 1""", 1, None)
+
 if verbose:
     print 'Running tests on sre.sub'
 
