diff options
| author | Serhiy Storchaka <storchaka@gmail.com> | 2013-09-16 21:00:46 (GMT) |
|---|---|---|
| committer | Serhiy Storchaka <storchaka@gmail.com> | 2013-09-16 21:00:46 (GMT) |
| commit | e787bce79c016cf7b9e743a6230ceb8e97b4a938 (patch) | |
| tree | 2b86121e69e169391770e3adace1e55c28805fc8 /Lib/lib2to3/pgen2 | |
| parent | 74213e4ee941e163ebdfa6b7d3d493ee68f6bb8d (diff) | |
| download | cpython-e787bce79c016cf7b9e743a6230ceb8e97b4a938.zip cpython-e787bce79c016cf7b9e743a6230ceb8e97b4a938.tar.gz cpython-e787bce79c016cf7b9e743a6230ceb8e97b4a938.tar.bz2 | |
Issue #18873: 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 e090aa9..9cf2a69 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: |