U ]9Y@s0ddlZddlZddlmZGdddeZdS)N) ProbingStatec@sneZdZdZdddZddZeddZd d Zed d Z d dZ e ddZ e ddZ e ddZdS) CharSetProbergffffff?NcCsd|_||_tt|_dSN)_state lang_filterloggingZ getLogger__name__Zlogger)selfrr 7/usr/lib/python3/dist-packages/chardet/charsetprober.py__init__'szCharSetProber.__init__cCs tj|_dSr)rZ DETECTINGrr r r r reset,szCharSetProber.resetcCsdSrr rr r r charset_name/szCharSetProber.charset_namecCsdSrr )r bufr r r feed3szCharSetProber.feedcCs|jSr)rrr r r state6szCharSetProber.statecCsdS)Ngr rr r r get_confidence:szCharSetProber.get_confidencecCstdd|}|S)Ns([-])+ )resub)rr r r filter_high_byte_only=sz#CharSetProber.filter_high_byte_onlycCs\t}td|}|D]@}||dd|dd}|sL|dkrLd}||q|S)u9 We define three types of bytes: alphabet: english alphabets [a-zA-Z] international: international characters [€-ÿ] marker: everything else [^a-zA-Z€-ÿ] The input buffer can be thought to contain a series of words delimited by markers. This function works to filter all words that contain at least one international character. All contiguous sequences of markers are replaced by a single space ascii character. This filter applies to all scripts which do not use English characters. s%[a-zA-Z]*[-]+[a-zA-Z]*[^a-zA-Z-]?Nr) bytearrayrfindallextendisalpha)rfilteredZwordsZwordZ last_charr r r filter_international_wordsBs  z(CharSetProber.filter_international_wordscCst}d}d}tt|D]n}|||d}|dkr characters. Also retains English alphabet and high byte characters immediately before occurrences of >. This filter can be applied to all scripts which contain both English characters and extended ASCII characters, but is currently only used by ``Latin1Prober``. Frr>s