diff options
| author | Serhiy Storchaka <storchaka@gmail.com> | 2013-09-16 20:51:56 (GMT) |
|---|---|---|
| committer | Serhiy Storchaka <storchaka@gmail.com> | 2013-09-16 20:51:56 (GMT) |
| commit | dafea851901fc1de278ad79727d3b44f46ba5a31 (patch) | |
| tree | 5f8d95de4856502e61c78168e7918776b161e9b4 /Lib/lib2to3/pgen2 | |
| parent | 975fce37883899a55bbcdaa6300c5c6ffe9d3db2 (diff) | |
| download | cpython-dafea851901fc1de278ad79727d3b44f46ba5a31.zip cpython-dafea851901fc1de278ad79727d3b44f46ba5a31.tar.gz cpython-dafea851901fc1de278ad79727d3b44f46ba5a31.tar.bz2 | |
Issue #18873: The tokenize module, IDLE, 2to3, and the findnocoding.py script
now detect Python source code encoding only in comment lines.
Diffstat (limited to 'Lib/lib2to3/pgen2')
| -rw-r--r-- | Lib/lib2to3/pgen2/tokenize.py | 9 |
1 files changed, 4 insertions, 5 deletions
diff --git a/Lib/lib2to3/pgen2/tokenize.py b/Lib/lib2to3/pgen2/tokenize.py index 31e2969..83656fc 100644 --- a/Lib/lib2to3/pgen2/tokenize.py +++ b/Lib/lib2to3/pgen2/tokenize.py @@ -236,7 +236,7 @@ class Untokenizer: startline = False toks_append(tokval) -cookie_re = re.compile("coding[:=]\s*([-\w.]+)") +cookie_re = re.compile(r'^[ \t\f]*#.*coding[:=][ \t]*([-\w.]+)', re.ASCII) def _get_normal_name(orig_enc): """Imitates get_normal_name in tokenizer.c.""" @@ -281,11 +281,10 @@ def detect_encoding(readline): line_string = line.decode('ascii') except UnicodeDecodeError: return None - - matches = cookie_re.findall(line_string) - if not matches: + match = cookie_re.match(line_string) + if not match: return None - encoding = _get_normal_name(matches[0]) + encoding = _get_normal_name(match.group(1)) try: codec = lookup(encoding) except LookupError: |