DjUUdZddlmZddlZddlZddlmZdZded<dZ d ed <d Z ded <iZ d ed<d|d}dZ d~dZ ddZddZddZddZ d| ddZiddddddddddd d!d"d#d$d%d&d'd(d)d*d+d,d-d.d/d0d1d2d3d4d5d6d7id8d9d:d;dd?d@dAdBdCdDdEdFdGdHdIdJdKdLdMdNdOdPdQdRdSdTdUdVdWdXdYidZd[d\d]d^d_d`dadbdcdddedfdgdhdidjdkdldmdndodpdqdrdsdtdudvdwdxdyZdzed{<y)z&Internal shared utilities for chardet.) annotationsN)Callablei@ intDEFAULT_MAX_BYTESg?floatMINIMUM_THRESHOLDi_DEFAULT_CHUNK_SIZEz3dict[str, Callable[..., codecs.IncrementalDecoder]]_INCREMENTAL_DECODERScP|tk7rtjdt|yy)zDEmit a deprecation warning if *chunk_size* differs from the default.zEchunk_size is not used in this version of chardet and will be ignored stacklevelN)r warningswarnDeprecationWarning) chunk_sizer s L/root/workspace/ytshorts/venv/lib/python3.12/site-packages/chardet/_utils.py_warn_deprecated_chunk_sizer!s%(( S ! )ctj|}| tj|}|t|<|S#t$rYywxYw)aaReturn a fresh incremental decoder for *encoding*, or ``None``. The class lookup is cached; ``None`` means the codec does not exist. Every decode below catches ``UnicodeError`` rather than ``UnicodeDecodeError`` because the utf-16/utf-32 incremental decoders raise a bare ``UnicodeError`` when the stream does not start with a BOM. N)r getcodecsgetincrementaldecoder LookupError)encoding decoder_classs r_incremental_decoderr+sX*--h7M "88BM+8h' ?  s> A  A cft|}|y |j|dy#t$rYywxYw)aReturn ``True`` if *data* decodes cleanly under *encoding*. Equivalent to ``data.decode(encoding, errors="strict")`` except that an incomplete multi-byte sequence at the *end* of *data* is accepted instead of raising. Detection input is nearly always a prefix of a larger whole. Callers pass the first N bytes of a file, and chardet slices further on its own. For a two-byte encoding any of those cuts lands mid-character roughly half the time. A one-shot strict decode cannot tell a truncated tail from corrupt data: it raises either way, so the candidate is discarded and every CJK encoding can disappear from the candidate set over a single dangling lead byte. An incremental decoder with ``final=False`` defers the partial tail instead, while still raising on genuine corruption anywhere before it. :param data: The raw byte data to test. :param encoding: Name of the codec to test *data* against. :returns: ``True`` if *data* decodes without error, ``False`` otherwise. FfinalTrdecode UnicodeErrordatardecoders rdecodes_without_errorr&?sD,#8,Gt5)   $ 00cft|}|y |j|dy#t$rYywxYw)aKReturn ``True`` if *data* decodes under *encoding* with nothing deferred. The strict sibling of :func:`decodes_without_error`: ``final=True`` makes an incomplete multi-byte sequence at the end of *data* an error rather than a deferred tail. This is the question that matters when *data* is the caller's entire input --- ``data.decode(encoding)`` will make exactly this judgment. :param data: The raw byte data to test. :param encoding: Name of the codec to test *data* against. :returns: ``True`` if *data* decodes completely, ``False`` otherwise. FTrr r#s rdecodes_completelyr)_sD#8,Gt4(  r'ct|}|y |j|d}|r|jsy |jddy#t$rYywxYw#t$rYywxYw)aReturn ``True`` if *data* is ASCII text plus an incomplete tail. One decode pass answers both halves of the decode-safety question: the tolerant (``final=False``) decode yields the text before any deferred tail, and flushing the decoder afterwards raises exactly when a deferred tail existed. True means the candidate decoded real ASCII characters and then hit an incomplete multi-byte sequence at the end --- its only *non-ASCII* evidence is the undecodable tail itself. Deliberately False when the tolerant decode yields nothing at all (the entire input is one dangling sequence): that candidate has zero decoded evidence, not ASCII evidence, and a clipped multi-byte fragment is better served by the ranking's own judgment. :param data: The raw byte data to test. :param encoding: Name of the codec to decode *data* with. :returns: ``True`` if *data* decodes to non-empty pure ASCII with an incomplete sequence deferred at the end. FrrT)rr!r"isascii)r$rr%texts rdangling_tail_with_ascii_prefixr-vs~(#8,G~~d%~0 t||~s$'   s"A A AA A&%A&cht|tst|tr|dkr d}t|y)z:Raise ValueError if *max_bytes* is not a positive integer.z$max_bytes must be a positive integerN) isinstanceboolr ValueError) max_bytesmsgs r_validate_max_bytesr5s2)T"*Y*D TU 4oIVrcD|rtjdt|y|S)zEResolve the deprecated *should_rename_legacy* into *prefer_superset*.z?should_rename_legacy is deprecated, use prefer_superset insteadr T)rrr)should_rename_legacyprefer_supersetr s r_resolve_prefer_supersetr9s' M !  rararabicbe belarusianbg bulgarianbrbretoncsczechcywelshdadanishdegermanelgreekenenglisheo esperantoesspanishetestonianfafarsififinnishfrfrenchgairishgdgaelichehebrewhrcroatianhu hungarianid indonesianis icelandicititalianjajapanesekkkazakhkokoreanlt lithuanianlvlatvianmk macedonianmsmalaymtmaltesenldutchno norwegianplpolishpt portugueseroromanianrurussianskslovakslslovenesrserbiansvswedishtgtajikththaitrturkishuk ukrainianund undeterminedururduvi vietnamesezhchinesezdict[str, str]ISO_TO_LANGUAGE))rrr rreturnNone)rstrrz codecs.IncrementalDecoder | None)r$bytesrrrr1)r3rrr)r7r1r8r1r rrr1)__doc__ __future__rrrcollections.abcrr__annotations__rr r rrr&r)r-r5r9rrrrs/," $!3  5"S!NPJO %(@.!HJK  15 CF    3#(3#,3# +3# ( 3#  ' 3#  ' 3# (3# (3# '3# )3# +3# )3# *3# '3# )3#  (!3#" '#3#$ (%3#& ('3#( *)3#* ++3#, ,-3#. +/3#0 )13#2 *33#4 (53#6 (73#8 ,93#: );3#< ,=3#> '?3#@ )A3#B 'C3#D +E3#F (G3#H ,I3#J *K3#L )M3#N (O3#P )Q3#R )S3#T )U3#V 'W3#X &Y3#Z )[3#\ +]3#^ >_3#` &a3#b ,c3#d )e3#3r