Merged revisions 79723 via svnmerge from

benjaminp · benjaminp · commit 8daa49ee3466 · 2010-04-03T22:55:48.000Z
svn+ssh://pythondev@svn.python.org/python/trunk

........
  r79723 | benjamin.peterson | 2010-04-03 17:48:51 -0500 (Sat, 03 Apr 2010) | 1 line

  ensure that the locale does not affect the tokenization of identifiers
........
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -12,6 +12,8 @@ What's New in Python 3.2 Alpha 1?
 Core and Builtins
 -----------------
 
+- Ensure that tokenization of identifiers is not affected by locale.
+
 - Issue #1222585: Added LDCXXSHARED for C++ support. Patch by Arfrever.
 
 - Raise a TypeError when trying to delete a T_STRING_INPLACE struct member.
diff --git a/Parser/tokenizer.c b/Parser/tokenizer.c
@@ -179,6 +179,16 @@ decode_str(const char *str, int exec_input, struct tok_state *tok)
 
 #else /* PGEN */
 
+/* Ensure that the locale does not interfere with tokenization. */
+
+static int
+ascii_isalnum(int c)
+{
+	return (('a' <= c && c <= 'z') ||
+		('A' <= c && c <= 'Z') ||
+		('0' <= c && c <= '9'));
+}
+
 static char *
 error_ret(struct tok_state *tok) /* XXX */
 {
@@ -245,7 +255,7 @@ get_coding_spec(const char *s, Py_ssize_t size)
 			} while (t[0] == '\x20' || t[0] == '\t');
 
 			begin = t;
-			while (isalnum(Py_CHARMASK(t[0])) ||
+			while (ascii_isalnum(Py_CHARMASK(t[0])) ||
 			       t[0] == '-' || t[0] == '_' || t[0] == '.')
 				t++;
 
