diff -r 4be39be5230c Lib/idlelib/IOBinding.py --- a/Lib/idlelib/IOBinding.py Thu Aug 21 10:00:41 2014 +0300 +++ b/Lib/idlelib/IOBinding.py Sat Aug 23 10:44:07 2014 +0300 @@ -72,6 +72,7 @@ encoding = encoding.lower() coding_re = re.compile(r'^[ \t\f]*#.*coding[:=][ \t]*([-\w.]+)') +blank_re = re.compile(r'^[ \t\f]*(?:[#\r\n]|$)') class EncodingMessage(SimpleDialog): "Inform user that an encoding declaration is needed." @@ -130,6 +131,8 @@ match = coding_re.match(line) if match is not None: break + if not blank_re.match(line): + return None else: return None name = match.group(1) diff -r 4be39be5230c Lib/lib2to3/pgen2/tokenize.py --- a/Lib/lib2to3/pgen2/tokenize.py Thu Aug 21 10:00:41 2014 +0300 +++ b/Lib/lib2to3/pgen2/tokenize.py Sat Aug 23 10:44:07 2014 +0300 @@ -237,6 +237,7 @@ toks_append(tokval) cookie_re = re.compile(r'^[ \t\f]*#.*coding[:=][ \t]*([-\w.]+)') +blank_re = re.compile(r'^[ \t\f]*(?:[#\r\n]|$)') def _get_normal_name(orig_enc): """Imitates get_normal_name in tokenizer.c.""" @@ -309,6 +310,8 @@ encoding = find_cookie(first) if encoding: return encoding, [first] + if not blank_re.match(first): + return default, [first] second = read_or_stop() if not second: diff -r 4be39be5230c Lib/test/test_compile.py --- a/Lib/test/test_compile.py Thu Aug 21 10:00:41 2014 +0300 +++ b/Lib/test/test_compile.py Sat Aug 23 10:44:07 2014 +0300 @@ -412,9 +412,24 @@ l = lambda: "foo" self.assertIsNone(l.__doc__) - def test_unicode_encoding(self): + @test_support.requires_unicode + def test_encoding(self): + code = b'# -*- coding: badencoding -*-\npass\n' + self.assertRaises(SyntaxError, compile, code, 'tmp', 'exec') code = u"# -*- coding: utf-8 -*-\npass\n" self.assertRaises(SyntaxError, compile, code, "tmp", "exec") + code = 'u"\xc2\xa4"\n' + self.assertEqual(eval(code), u'\xc2\xa4') + code = u'u"\xc2\xa4"\n' + self.assertEqual(eval(code), u'\xc2\xa4') + code = '# -*- coding: latin1 -*-\nu"\xc2\xa4"\n' + self.assertEqual(eval(code), u'\xc2\xa4') + code = '# -*- coding: utf-8 -*-\nu"\xc2\xa4"\n' + self.assertEqual(eval(code), u'\xa4') + code = '# -*- coding: iso8859-15 -*-\nu"\xc2\xa4"\n' + self.assertEqual(eval(code), test_support.u(r'\xc2\u20ac')) + code = 'u"""\\\n# -*- coding: utf-8 -*-\n\xc2\xa4"""\n' + self.assertEqual(eval(code), u'# -*- coding: utf-8 -*-\n\xc2\xa4') def test_subscripts(self): # SF bug 1448804 diff -r 4be39be5230c Misc/NEWS --- a/Misc/NEWS Thu Aug 21 10:00:41 2014 +0300 +++ b/Misc/NEWS Sat Aug 23 10:44:07 2014 +0300 @@ -10,6 +10,13 @@ Core and Builtins ----------------- +- Issue #22221: The first line of Python script could be executed twice when + the source encoding was specified on the second line. Now the source encoding + declaration on the second line isn't effective if the first line contains + anything except a comment. 'python -x' works now again with files with the + source encoding declarations, and can be used to make Python batch files + on Windows. + - Issue #22023: Fix ``%S``, ``%R`` and ``%V`` formats of :c:func:`PyUnicode_FromFormat`. @@ -94,9 +101,19 @@ adds support for the Linux AArch64 and POWERPC ELF ABIv2 little endian architectures. +IDLE +---- + +- Issue #22221: IDLE now ignores the source encoding declaration on the second + line if the first line contains anything except a comment. + Tools/Demos ----------- +- Issue #22221: 2to3 and the findnocoding.py script now ignore the source + encoding declaration on the second line if the first line contains anything + except a comment. + - Issue #22201: Command-line interface of the zipfile module now correctly extracts ZIP files with directory entries. Patch by Ryan Wilson. diff -r 4be39be5230c Parser/tokenizer.c --- a/Parser/tokenizer.c Thu Aug 21 10:00:41 2014 +0300 +++ b/Parser/tokenizer.c Sat Aug 23 10:44:07 2014 +0300 @@ -259,11 +259,25 @@ char * cs; int r = 1; - if (tok->cont_line) + if (tok->cont_line) { /* It's a continuation line, so it can't be a coding spec. */ + tok->read_coding_spec = 1; return 1; + } cs = get_coding_spec(line, size); - if (cs != NULL) { + if (!cs) { + Py_ssize_t i; + for (i = 0; i < size; i++) { + if (line[i] == '#' || line[i] == '\n' || line[i] == '\r') + break; + if (line[i] != ' ' && line[i] != '\t' && line[i] != '\014') { + /* Stop checking coding spec after a line containing + * anything except a comment. */ + tok->read_coding_spec = 1; + break; + } + } + } else { tok->read_coding_spec = 1; if (tok->encoding == NULL) { assert(tok->decoding_state == 1); /* raw */ @@ -688,7 +702,7 @@ if (newl[0]) { if (!check_coding_spec(str, newl[0] - str, tok, buf_setreadl)) return error_ret(tok); - if (tok->enc == NULL && newl[1]) { + if (tok->enc == NULL && !tok->read_coding_spec && newl[1]) { if (!check_coding_spec(newl[0]+1, newl[1] - newl[0], tok, buf_setreadl)) return error_ret(tok); diff -r 4be39be5230c Tools/scripts/findnocoding.py --- a/Tools/scripts/findnocoding.py Thu Aug 21 10:00:41 2014 +0300 +++ b/Tools/scripts/findnocoding.py Sat Aug 23 10:44:07 2014 +0300 @@ -33,6 +33,7 @@ decl_re = re.compile(r'^[ \t\f]*#.*coding[:=][ \t]*([-\w.]+)') +blank_re = re.compile(r'^[ \t\f]*(?:[#\r\n]|$)') def get_declaration(line): match = decl_re.match(line) @@ -57,7 +58,8 @@ line1 = infile.readline() line2 = infile.readline() - if get_declaration(line1) or get_declaration(line2): + if (get_declaration(line1) or + blank_re.match(line1) and get_declaration(line2)): # the file does have an encoding declaration, so trust it infile.close() return False