+
    QV-j¾  ã                   ó  € R t ^ RIHtHt ^ RIHt ^ RIt^ RIt^ RI	H
t
 ^ RIHt ^ RIHt ]P                  P                   P#                  ]4      tR.t] ! R R	]P*                  4      4       t ! R
 R]P.                  4      tR# )a¸  CoNLL format dataset builder.

Reads CoNLL-style files where each line carries one token plus its tag columns,
columns are whitespace-separated, and empty lines mark sentence boundaries.
Each example is one sentence, with each configured column produced as a list
aligned with the tokens list.

Supports CoNLL-2000 (chunking), CoNLL-2003 (NER), CoNLL-U (Universal
Dependencies), and any custom column schema by overriding ``column_names``.
)Ú	dataclassÚfield)ÚOptionalN)ÚKey©Úrequire_storage_cast)Ú
table_castÚtokensc                   ó\   a € ] tR t^t o RtRt]! R R7      tRtRt	Rt
RtRtV 3R ltRtV tR# )	ÚConllConfigaü  BuilderConfig for CoNLL-style files.

Args:
    features: (`Features`, *optional*):
        Cast the data to `features`.
    column_names: (`list[str]`, defaults to `["tokens"]`):
        Names for each whitespace-separated column. The first name is the
        token column; subsequent names cover the tag columns. Common
        schemas:

        - CoNLL-2003 NER: `["tokens", "pos_tags", "chunk_tags", "ner_tags"]`
        - CoNLL-2000 chunking: `["tokens", "pos_tags", "chunk_tags"]`
        - CoNLL-U: `["id", "form", "lemma", "upos", "xpos", "feats",
          "head", "deprel", "deps", "misc"]`
    delimiter: (`str`, *optional*):
        Column delimiter inside each line. If `None` (default), any
        whitespace is used, matching the standard CoNLL convention.
    encoding: (`str`, defaults to `"utf-8"`):
        Encoding to decode the file.
    encoding_errors: (`str`, *optional*):
        Argument to define what to do in case of encoding error. Same as
        the `errors` argument in `open()`.
    skip_docstart: (`bool`, defaults to `True`):
        Skip CoNLL-2003 `-DOCSTART-` document-boundary marker lines.
    comment_prefix: (`str`, *optional*):
        If set, lines starting with this prefix are skipped. CoNLL-U
        comments start with `#`.
Nc                 ó    € \        \        4      # ©N)ÚlistÚDEFAULT_COLUMN_NAMES© ó    Úv/Volumes/fast/ai/experiments/ui-tars-smoke/.venv/lib/python3.14/site-packages/datasets/packaged_modules/conll/conll.pyÚ<lambda>ÚConllConfig.<lambda>=   s
   € ¼DÔAUÔ<Vr   )Údefault_factoryzutf-8Tc                óÒ   <€ V ^8„  d   Qh/ S[ S[P                  ,          ;R&   S[S[,          ;R&   S[ S[,          ;R&   S[;R&   S[ S[,          ;R&   S[;R&   S[ S[,          ;R&   # )é   ÚfeaturesÚcolumn_namesÚ	delimiterÚencodingÚencoding_errorsÚskip_docstartÚcomment_prefix)r   ÚdatasetsÚFeaturesr   ÚstrÚbool)ÚformatÚ__classdict__s   "€r   Ú__annotate__ÚConllConfig.__annotate__   s‡   ø‡ ‚ ñ> ‘x×(Ñ(Õ)Ñ0ñ? ñ@ ‘s•)ÑWñA ñB ™�}Ñ#ñC ñD ÑñE ñF ™c•]Ñ)ñG ñH ÑñI ñJ ™S•MÑ(òK r   r   )Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__r   r   r   r   r   r   r   r   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__©r$   s   @r   r   r      s=   ø‡ € ñð: -1€HÙ#Ñ4VÔW€LØ#€IØ€HØ%)€OØ€MØ$(€N÷K ƒ r   r   c                   óN   a € ] tR t^Et o ]tR tR tV 3R lR ltR t	R t
RtV tR# )	ÚConllc                óX   € \         P                  ! V P                  P                  R 7      # ))r   )r   ÚDatasetInfoÚconfigr   )Úselfs   &r   Ú_infoÚConll._infoH   s   € Ü×#Ò#¨T¯[©[×-AÑ-AÔBÐBr   c                óð  € V P                   P                  '       g#   \        RV P                   P                   24      hRVP                  n        VP                  V P                   P                  4      pVP                  V4      p. pVP                  4        FU  w  rVV Uu. uF  qqP                  V4      NK  	  ppVP                  \        P                  ! VRVRW%,          /R7      4       KW  	  V# u upi )zíThe `data_files` kwarg in load_dataset() can be a str, List[str],
Dict[str,str], or Dict[str,List[str]].

If str or List[str], then the dataset returns only the 'train' split.
If dict, then keys should be from the `datasets.Split` enum.
z=At least one data file must be specified, but got data_files=TÚfiles_iterablesÚ
base_files)ÚnameÚ
gen_kwargs)r4   Ú
data_filesÚ
ValueErrorÚdownload_configÚextract_on_the_flyÚdownloadÚextractÚitemsÚ
iter_filesÚappendr   ÚSplitGenerator)	r5   Ú
dl_managerÚbase_data_filesÚextracted_data_filesÚsplitsÚ
split_nameÚfilesÚfiler9   s	   &&       r   Ú_split_generatorsÚConll._split_generatorsK   sâ   € ð �{‰{×%×%Ð%ÜÐ\Ð]a×]hÑ]h×]sÑ]sÐ\tÐuÓvÐvØ8<ˆ
×"Ñ"Ô5Ø$×-Ñ-¨d¯k©k×.DÑ.DÓEˆØ)×1Ñ1°/ÓBÐØˆØ!5×!;Ñ!;Ö!=ÑˆJÙGLÓMÁu¸t×4Ñ4°TÖ:ÁuˆOÐMØ�M‰MÜ×'Ò'Ø#à)¨?Ø$ oÕ&Að ôöñ ">ð ˆùò Ns   Â!C3c                óN   <€ V ^8„  d   QhRS[ P                  RS[ P                  /# )r   Úpa_tableÚreturn)ÚpaÚTable)r#   r$   s   "€r   r%   ÚConll.__annotate__e   s#   ø€ ÷ ñ ¡B§H¡Hð ±·±ñ r   c                ó®  € V P                   P                  e½   V P                   P                  P                  p\        ;QJ d>    R V P                   P                  P	                  4        4       F  '       d   K   RM3	  RM/! R V P                   P                  P	                  4        4       4      '       d   VP                  V4      pV# \        W4      pV# )Nc              3   óB   "  € T F  p\        V4      '       * x € K  	  R # 5ir   r   )Ú.0Úfeatures   & r   Ú	<genexpr>Ú$Conll._cast_table.<locals>.<genexpr>h   s   é € ÐbÑDa¸Ô+¨GÓ4×4Ñ4ÓDaùs   ‚FT)r4   r   Úarrow_schemaÚallÚvaluesÚcastr   )r5   rQ   Úschemas   && r   Ú_cast_tableÚConll._cast_tablee   s”   € Ø�;‰;×ÑÒ+Ø—[‘[×)Ñ)×6Ñ6ˆFß‹sÑbÀDÇKÁK×DXÑDX×D_ÑD_ÔDaÓb�s�sŠsÑbÀDÇKÁK×DXÑDX×D_ÑD_ÔDaÓb×bÒbØ#Ÿ=™=¨Ó0�ð ˆô & hÓ7�Øˆr   c              #  ó&   "  € T R j  x€L
  R #  L5ir   r   )r5   r:   r9   s   &&&r   Ú_generate_shardsÚConll._generate_shardsn   s   é € Ø×Ôùs   ‚ˆ‰c              #  óP  "  € \        V P                  P                  4      pV'       g   \        R 4      h\	        V4      pV P                  P
                  pV P                  P                  pV P                  P                  p\        V4       EFj  w  r‰V	 EF]  p
. p\        V4       Uu. uF  p. NK  	  pp\        W P                  P                  V P                  P                  R7      ;_uu_ 4       pV EF>  pVP                  R4      pVP                  4       '       g=   V^ ,          '       d,   VP                  V4       \        V4       Uu. uF  p. NK  	  ppKg  V'       d   VP!                  R4      '       d   Kˆ  Ve   VP!                  V4      '       d   K¥  Ve   VP#                  V4      MVP#                  4       p\	        V4      V8  d#   VR.V\	        V4      ,
          ,          ,           pM\	        V4      V8”  d   VRV p\        V4       F  w  ppVV,          P                  V4       K   	  EKA  	  V^ ,          '       d   VP                  V4       RRR4       V'       g   EKÖ  \        V4       UUu. uF0  p\$        P&                  ! V Uu. uF  pVV,          NK  	  up4      NK2  	  ppp\$        P(                  P+                  VVR7      p\-        V^ 4      V P/                  V4      3x € EK`  	  EKm  	  R# u upi u upi   + '       g   i     L¸; iu upi u uppi 5i)z1ConllConfig.column_names must be a non-empty list)r   Úerrorsz
z
-DOCSTART-NÚ )Únames)r   r4   r   r>   Úlenr   r   r   Ú	enumerateÚrangeÚopenr   r   ÚrstripÚstriprE   Ú
startswithÚsplitrS   ÚarrayrT   Úfrom_arraysr   ra   )r5   r:   r9   r   Únum_colsr   r   r   Ú	shard_idxÚfiles_iterablerM   Ú	sentencesÚ_ÚcurrentÚfÚraw_lineÚlineÚpartsÚiÚvalÚcol_idxÚsentenceÚarraysrQ   s   &&&                     r   Ú_generate_tablesÚConll._generate_tablesq   sw  é € Ü˜DŸK™K×4Ñ4Ó5ˆßÜÐPÓQÐQÜ�|Ó$ˆØ—K‘K×)Ñ)ˆ	ØŸ™×1Ñ1ˆØŸ™×3Ñ3ˆä)2°?×)CÑ%ˆIÜ&�Ø35�	Ü8=¸h¼Ó+H¹°1«B¹�Ð+HÜ˜$¯©×)=Ñ)=ÀdÇkÁk×FaÑFa×bÕbÐfgÜ$%˜à'Ÿ™¨vÓ6˜Ø#Ÿz™zŸ|š|à& qŸzœzØ )× 0Ñ 0°Ô 9Ü7<¸X´Ó*G±°!«2± Ð*GÙ$ß(¨T¯_©_¸\×-JÒ-JÙ$Ø)Ò5¸$¿/¹/È.×:YÒ:YÙ$Ø9BÒ9N §
¡
¨9Ô 5ÐTX×T^ÑT^ÓT`˜ä˜u›:¨Ô0Ø$)¨R¨D°H¼sÀ5»zÕ4IÕ,JÕ$J™EÜ  ›Z¨(Ô2Ø$)¨)¨8Ð$4˜EÜ&/°Ö&6™F˜A˜sØ# A�J×-Ñ-¨cÖ2ô '7ñ' %&ð, ˜q—z”zØ!×(Ñ(¨Ô1÷1 c÷4 ’9ÜejÐksÔetÔuÑetÐZaœbŸhšhÉ)Ó'TÉ)¸h¨°×(9Ð(9É)Ñ'TÖUÑet�FÑuÜ!Ÿx™x×3Ñ3°FÀ,Ð3ÓO�HÜ˜i¨Ó+¨T×-=Ñ-=¸hÓ-GÐGÕGôA 'ó *Dùò ,Iùò +H÷ c×büò6 (UùÓuùs†   ‚B&L&Â(K>Â3?L&Ã2<L	Ä/L	ÅLÅL	Å$L	Å;L	ÆB.L	ÉL	ÉL&É+L&É9L 
ÊLÊ$L 
Ê/AL&ÌL	ÌLÌL&ÌL 
Ì L&r   N)r'   r(   r)   r*   r   ÚBUILDER_CONFIG_CLASSr6   rN   ra   rd   rƒ   r-   r.   r/   s   @r   r1   r1   E   s1   ø‡ € Ø&ÐòCò÷4ð ò÷*Hð *Hr   r1   )r+   Údataclassesr   r   Útypingr   ÚpyarrowrS   r   Údatasets.builderr   Údatasets.features.featuresr   Údatasets.tabler   ÚutilsÚloggingÚ
get_loggerr'   Úloggerr   ÚBuilderConfigr   ÚArrowBasedBuilderr1   r   r   r   Ú<module>r’      s   ðñ	÷ )Ý ã ã Ý  Ý ;Ý %ð 
�‰×	Ñ	×	*Ñ	*¨8Ó	4€ð !�zÐ ð ô$)�(×(Ñ(ó $)ó ð$)ôNVHˆH×&Ñ&ö VHr   