fjdZddlmZddlZddlmZddlmZddlm Z ddl m Z ddl m Z mZmZdd lmZmZdd lmZmZmZdd lmZmZmZdd lmZdd lmZmZGddZ dS)u3UniversalDetector — streaming encoding detection.) annotationsN)Iterable)MappingProxyType)ClassVar)_utils)DEFAULT_MAX_BYTES_resolve_prefer_superset_validate_max_bytes) EncodingEraLanguageFilter)PREFERRED_SUPERSETapply_compat_namesapply_preferred_superset) _NONE_RESULT DetectionDictDetectionResult) run_pipeline)_validate_encodingnormalize_encodingsceZdZUdZejZeeZde d<e j de j e fddddddd d%dZd&dZd'd!Zd(d"Zed)d#Zed'd$ZdS)*UniversalDetectoraStreaming character encoding detector. Implements a feed/close pattern for incremental detection of character encoding from byte streams. Compatible with the chardet 6.x API. All detection is performed by the same pipeline used by :func:`chardet.detect` and :func:`chardet.detect_all`, ensuring consistent results regardless of which API is used. .. note:: This class is **not** thread-safe. Each thread should create its own :class:`UniversalDetector` instance. z$ClassVar[MappingProxyType[str, str]] LEGACY_MAPFTNcp1252zutf-8)prefer_superset compat_namesinclude_encodingsexclude_encodingsno_match_encodingempty_input_encoding lang_filterr should_rename_legacybool encoding_erar max_bytesintrrrIterable[str] | NonerrstrrreturnNonec|tjkrtjdtdt ||}||_||_t|||_ ||_ t|d|_ t|d|_ t| d|_t| d|_t#|_d|_d|_d |_d S) a3Initialize the detector. :param lang_filter: Deprecated -- accepted for backward compatibility but has no effect. A warning is emitted when set to anything other than :attr:`LanguageFilter.ALL`. :param should_rename_legacy: Deprecated alias for *prefer_superset*. :param encoding_era: Restrict candidate encodings to the given era. :param max_bytes: Maximum number of bytes to buffer from :meth:`feed` calls before stopping accumulation. :param prefer_superset: If ``True``, remap ISO subset encodings to their Windows/CP superset equivalents (e.g., ISO-8859-1 -> Windows-1252). :param compat_names: If ``True`` (default), return encoding names compatible with chardet 5.x/6.x. If ``False``, return raw Python codec names. :param include_encodings: If given, restrict detection to only these encodings (names or aliases). :param exclude_encodings: If given, remove these encodings from the candidate set. :param no_match_encoding: Encoding to return when no candidate survives the pipeline. Defaults to ``"cp1252"``. :param empty_input_encoding: Encoding to return for empty input. Defaults to ``"utf-8"``. zMlang_filter is not implemented in this version of chardet and will be ignored) stacklevelrrrrFN)r ALLwarningswarnDeprecationWarningr _prefer_superset _compat_namesr _encoding_era _max_bytesr_include_encodings_exclude_encodingsr_no_match_encoding_empty_input_encoding bytearray_buffer_done_closed_result) selfr r!r#r$rrrrrrs dC:\Users\Terasoftware\OneDrive\Desktop\faahhh\fyndo\fyndo\venv\Lib\site-packages\chardet/detector.py__init__zUniversalDetector.__init__2s L ., ,  M&"      3 /  !0)I&&&)#"5 2# # #6 2# # #5 2# # &8 "8& & "!{{   /3 byte_strbytes | bytearrayc|jrd}t||jrdS|jt |jz }|dkr"|j|d|t |j|jkr d|_dSdS)aFeed a chunk of bytes to the detector. Data is accumulated in an internal buffer. Once *max_bytes* have been buffered, :attr:`done` is set to ``True`` and further data is ignored until :meth:`reset` is called. :param byte_str: The next chunk of bytes to examine. :raises ValueError: If called after :meth:`close` without a :meth:`reset`. z+feed() called after close() without reset()NrT)r< ValueErrorr;r4lenr:extend)r>rBmsg remainings r?feedzUniversalDetector.feedxs < "?CS// ! :  FOc$,&7&77 q= 6 L  ) 4 5 5 5 t|   / DJJJ  rArc |jscd|_t|j}t||j|j|j|j|j|j }|d|_ d|_ |j S)zFinalize detection and return the best result. Runs the full detection pipeline on the buffered data. :returns: A dictionary with keys ``"encoding"``, ``"confidence"``, and ``"language"``. T)r$rrrrr) r<bytesr:rr3r4r5r6r7r8r=r;result)r>dataresultss r?closezUniversalDetector.closesz| DL&&D""/"&"9"&"9"&"9%)%?G#1:DLDJ{rAcVt|_d|_d|_d|_dS)z2Reset the detector to its initial state for reuse.FN)r9r:r;r<r=r>s r?resetzUniversalDetector.resets& {{    rAc|jS)z9Whether detection is complete and no more data is needed.)r;rRs r?donezUniversalDetector.dones zrAc|jG|j}|jrt||jrt ||St jS)z"The current best detection result.)r=to_dictr1rr2rr)r>ds r?rMzUniversalDetector.resultsg <  $$&&A$ ,(+++! &"1%%%H#%%%rA)r r r!r"r#r r$r%rr"rr"rr&rr&rr'rr'r(r))rBrCr(r))r(r)r(r))r(r")__name__ __module__ __qualname____doc__rMINIMUM_THRESHOLDrr r__annotations__r r-r rr@rJrPrSpropertyrUrMrAr?rrs%  08H7G88J '5&8%*$/O* D4!&!2626!)$+D4D4D4D4D4D4L,0X & & &X & & &rAr)!r\ __future__rr.collections.abcrtypesrtypingrchardetrchardet._utilsrr r chardet.enumsr r chardet.equivalencesr rrchardet.pipelinerrrchardet.pipeline.orchestratorrchardet.registryrrrr`rAr?rlss99""""""$$$$$$"""""" 65555555 JIIIIIIIII666666DDDDDDDDa&a&a&a&a&a&a&a&a&a&rA