ó
    O³™j¿  ã                   óZ  • S r SSKJr  SSKJrJrJr  \\\4   r\	r
 \\\	   \\	   4   r \\
\\
\
4   \\
   4   r \\\\\4   \\   4   r \\
\4   r \\\4   r  " S S\5      r " S S\5      r " S S	\5      rS
SKJrJrJrJrJrJrJrJrJrJrJrJrJ r J!r!  S
SK"J#r#J$r$J%r%J&r&J'r'  g)u  Tokenizers â€” fast, batteries-included tokenization library.

Free-threaded Python (3.14t) note:
    Wheels built against free-threaded CPython declare ``Py_MOD_GIL_NOT_USED``
    and use ``RwLock``-guarded interior mutability so component setters are
    safe to call from multiple threads. Compound mutations
    (``tokenizer.post_processor.special_tokens = â€¦``) are still not atomic â€”
    use a Python lock if you need the read-then-write to be serialized.
    See ``docs/free-threading-audit.md`` for the full analysis.
é    )ÚEnum)ÚListÚTupleÚUnionc                   ó   • \ rS rSrSrSrSrg)ÚOffsetReferentialéG   ÚoriginalÚ
normalized© N)Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__ÚORIGINALÚ
NORMALIZEDÚ__static_attributes__r   ó    Ú¡C:\Users\johnn\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.13_qbz5n2kfra8p0\LocalCache\local-packages\Python313\site-packages\tokenizers/__init__.pyr   r   G   s   † Ø€HØƒJr   r   c                   ó   • \ rS rSrSrSrSrg)Ú
OffsetTypeéL   ÚbyteÚcharr   N)r   r   r   r   ÚBYTEÚCHARr   r   r   r   r   r   L   s   † Ø€DØƒDr   r   c                   ó(   • \ rS rSrSrSrSrSrSrSr	g)	ÚSplitDelimiterBehavioréQ   ÚremovedÚisolatedÚmerged_with_previousÚmerged_with_nextÚ
contiguousr   N)
r   r   r   r   ÚREMOVEDÚISOLATEDÚMERGED_WITH_PREVIOUSÚMERGED_WITH_NEXTÚ
CONTIGUOUSr   r   r   r   r   r   Q   s   † Ø€GØ€HØ1ÐØ)ÐØƒJr   r   é   )Ú
AddedTokenÚEncodingÚNormalizedStringÚPreTokenizedStringÚRegexÚTokenÚ	TokenizerÚdecodersÚmodelsÚnormalizersÚpre_tokenizersÚ
processorsÚtrainersÚ__version__)ÚBertWordPieceTokenizerÚByteLevelBPETokenizerÚCharBPETokenizerÚSentencePieceBPETokenizerÚSentencePieceUnigramTokenizerN)(Ú__doc__Úenumr   Útypingr   r   r   ÚintÚOffsetsÚstrÚTextInputSequenceÚPreTokenizedInputSequenceÚTextEncodeInputÚPreTokenizedEncodeInputÚInputSequenceÚEncodeInputr   r   r   Ú
tokenizersr+   r,   r-   r.   r/   r0   r1   r2   r3   r4   r5   r6   r7   r8   Úimplementationsr9   r:   r;   r<   r=   r   r   r   Ú<module>rL      s  ðñ	õ ß %Ñ %ð ��S�‰/€àÐ Ø 5à! $ s¡)¨U°3©ZÐ"7Ñ8Ð ðð ØØ	Ð
Ð.Ð
.Ñ/ØÐ	Ñðñ€ð
ð  ØØ	Ð
#Ð%>Ð
>Ñ?ØÐ	"Ñ#ð%ñÐ ð
ð Ð'Ð)BÐBÑC€ðð �OÐ%<Ð<Ñ=€ðô˜ô ô
�ô ô
˜Tô ÷÷ ÷ ÷ ÷ ö r   