diff options
Diffstat (limited to 'src/parser.c')
| -rw-r--r-- | src/parser.c | 20 |
1 files changed, 19 insertions, 1 deletions
diff --git a/src/parser.c b/src/parser.c index 575eb73..1ebbaab 100644 --- a/src/parser.c +++ b/src/parser.c @@ -217,9 +217,27 @@ hubbub_error hubbub_parser_parse_chunk(hubbub_parser *parser, perror = parserutils_inputstream_append(parser->stream, data, len); if (perror != PARSERUTILS_OK) - return !HUBBUB_OK; + return hubbub_error_from_parserutils_error(perror); error = hubbub_tokeniser_run(parser->tok); + if (error == HUBBUB_BADENCODING) { + /* Ok, we autodetected an encoding that we don't actually + * support. We've not actually processed any data at this + * point so fall back to Windows-1252 and hope for the best + */ + perror = parserutils_inputstream_change_charset(parser->stream, + "Windows-1252", HUBBUB_CHARSET_TENTATIVE); + /* Under no circumstances should we get here if we've managed + * to process data. If there is a way, I want to know about it + */ + assert(perror != PARSERUTILS_INVALID); + if (perror != PARSERUTILS_OK) + return hubbub_error_from_parserutils_error(perror); + + /* Retry the tokenisation */ + error = hubbub_tokeniser_run(parser->tok); + } + if (error != HUBBUB_OK) return error; |