diff options
| author | Brett Cannon <bcannon@gmail.com> | 2008-10-17 03:38:50 (GMT) |
|---|---|---|
| committer | Brett Cannon <bcannon@gmail.com> | 2008-10-17 03:38:50 (GMT) |
| commit | da780432378e6298463889557ab43e0c156758cd (patch) | |
| tree | dc622a9b62874851f90abc45524d3d2653cab9ba /Parser/tokenizer.c | |
| parent | 9e9dcd6d4225faa6a8b19120f009e0253d16ab92 (diff) | |
| download | cpython-da780432378e6298463889557ab43e0c156758cd.zip cpython-da780432378e6298463889557ab43e0c156758cd.tar.gz cpython-da780432378e6298463889557ab43e0c156758cd.tar.bz2 | |
Latin-1 source code was not being properly decoded when passed through
compile(). This was due to left-over special-casing before UTF-8 became the
default source encoding.
Closes issue #3574. Thanks to Victor Stinner for help with the patch.
Diffstat (limited to 'Parser/tokenizer.c')
| -rw-r--r-- | Parser/tokenizer.c | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/Parser/tokenizer.c b/Parser/tokenizer.c index 4edf6d0..ce8129d 100644 --- a/Parser/tokenizer.c +++ b/Parser/tokenizer.c @@ -135,6 +135,7 @@ tok_new(void) tok->decoding_state = STATE_INIT; tok->decoding_erred = 0; tok->read_coding_spec = 0; + tok->enc = NULL; tok->encoding = NULL; tok->cont_line = 0; #ifndef PGEN @@ -274,8 +275,7 @@ check_coding_spec(const char* line, Py_ssize_t size, struct tok_state *tok, tok->read_coding_spec = 1; if (tok->encoding == NULL) { assert(tok->decoding_state == STATE_RAW); - if (strcmp(cs, "utf-8") == 0 || - strcmp(cs, "iso-8859-1") == 0) { + if (strcmp(cs, "utf-8") == 0) { tok->encoding = cs; } else { r = set_readline(tok, cs); |