idZddlmZddlmZmZmZmZmZddl m Z ddl m Z ddl mZmZddlmZddlmZmZdd lmZgd Zd ej0eef dd Zd d ej0eef ddZy)u?Universal character encoding detector — MIT-licensed rewrite.) annotations)_DEFAULT_CHUNK_SIZEDEFAULT_MAX_BYTESMINIMUM_THRESHOLD_validate_max_bytes_warn_deprecated_chunk_size) __version__)UniversalDetector) EncodingEraLanguageFilter)apply_legacy_rename) DetectionDictDetectionResult) run_pipeline) rrrrr r r r detect detect_allTct|t|t|tr|n t|}t |||}|dj }|r t ||S)aDetect the encoding of the given byte string. Parameters match chardet 6.x for backward compatibility. *chunk_size* is accepted but has no effect. :param byte_str: The byte sequence to detect encoding for. :param should_rename_legacy: If ``True`` (the default), remap legacy encoding names to their modern equivalents. :param encoding_era: Restrict candidate encodings to the given era. :param chunk_size: Deprecated -- accepted for backward compatibility but has no effect. :param max_bytes: Maximum number of bytes to examine from *byte_str*. :returns: A dictionary with keys ``"encoding"``, ``"confidence"``, and ``"language"``.  max_bytesr)rr isinstancebytesrto_dictr )byte_strshould_rename_legacy encoding_era chunk_sizerdataresultsresults `/Users/bowang/.openclaw/workspace/ChatDev/.venv/lib/python3.12/site-packages/chardet/__init__.pyrr!sX, + "!(E28hD4CG QZ   !FF# MFc\t|t|t|tr|n t|}t |||}|Dcgc]}|j } }|s | D cgc]} | dt kDs| } } | r| } |r| D] } t| t| ddScc}wcc} w)axDetect all possible encodings of the given byte string. Parameters match chardet 6.x for backward compatibility. *chunk_size* is accepted but has no effect. When *ignore_threshold* is False (the default), results with confidence <= MINIMUM_THRESHOLD (0.20) are filtered out. If all results are below the threshold, the full unfiltered list is returned as a fallback so the caller always receives at least one result. :param byte_str: The byte sequence to detect encoding for. :param ignore_threshold: If ``True``, return all candidate encodings regardless of confidence score. :param should_rename_legacy: If ``True`` (the default), remap legacy encoding names to their modern equivalents. :param encoding_era: Restrict candidate encodings to the given era. :param chunk_size: Deprecated -- accepted for backward compatibility but has no effect. :param max_bytes: Maximum number of bytes to examine from *byte_str*. :returns: A list of dictionaries, each with keys ``"encoding"``, ``"confidence"``, and ``"language"``, sorted by descending confidence. r confidencec |dS)Nr#)ds r zdetect_all..ksqr!T)keyreverse) rrrrrrrr sorted) rignore_thresholdrrrrrrrdictsr&filtereds r rrAs< + "!(E28hD4CG") *'QQYY['E * $Lu!,:K(KAuL EA  " %6 EE +LsB$%B)7B)N) rbytes | bytearrayrboolrr rintrr1returnr)rr/r+r0rr0rr rr1rr1r2zlist[DetectionDict])__doc__ __future__rchardet._utilsrrrrrchardet._versionr chardet.detectorr chardet.enumsr r chardet.equivalencesr chardet.pipelinerrchardet.pipeline.orchestratorr__all__ALLrrr%r!r r>sE").54;6  "& +)&      D#!% +)& *F*F*F*F *F  *F  *F*Fr!