+
    QV-jÎ  ã                   ó
  € ^ RI Ht ^ RIHt ^ RIHt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 R	]	P*                  4      4       t ! R
 R]	P.                  4      tR# )é    )Ú	dataclass)ÚStringIO)ÚLiteralÚOptionalN)ÚKey©Úrequire_storage_cast)Ú
table_castc                   óH   a € ] tR t^t o RtRtRtRtRtRt	Rt
V 3R ltRtV tR# )	Ú
TextConfigaê  BuilderConfig for text files.

Args:
    features: (`Features`, *optional*):
        Cast the data to `features`.
    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.
        This is the same as the `error` argument in `open()`.
    chunksize: (`Features`, *optional*, defaults to "10MB"):
        Chunk size to read the data.
    keep_linebreaks: (`bool`, defaults to False):
        Whether to keep line breaks.
    sample_by (`Literal["line", "paragraph", "document"]`, defaults to "line"):
        Whether to load data per line, praragraph or document.
        By default one row in the dataset = one line.
Nzutf-8FÚlinec                ó¤   <€ V ^8„  d   Qh/ S[ S[P                  ,          ;R&   S[;R&   S[ S[,          ;R&   S[;R&   S[;R&   S[R,          ;R&   # )é   ÚfeaturesÚencodingÚencoding_errorsÚ	chunksizeÚkeep_linebreaksÚ	sample_by)r   Ú	paragraphÚdocument)r   ÚdatasetsÚFeaturesÚstrÚintÚboolr   )ÚformatÚ__classdict__s   "€Út/Volumes/fast/ai/experiments/ui-tars-smoke/.venv/lib/python3.14/site-packages/datasets/packaged_modules/text/text.pyÚ__annotate__ÚTextConfig.__annotate__   sh   ø‡ ‚ ñ* ‘x×(Ñ(Õ)Ñ0ñ+ ñ, Ññ- ñ. ™c•]Ñ)ñ/ ñ0 Ññ1 ñ2 Ñ!ñ3 ñ4 Ð6Õ7Ñ@ò5 ó    © i    )Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__r   r   r   r   r   r   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__©r   s   @r   r   r      s1   ø‡ € ñð& -1€HØ€HØ%)€OØ€IØ!€OØ:@€I÷5 ƒ r"   r   c                   óN   a € ] tR t^-t o ]tR tR tV 3R lR ltR t	R t
RtV tR# )	ÚTextc                óX   € \         P                  ! V P                  P                  R 7      # ))r   )r   ÚDatasetInfoÚconfigr   )Úselfs   &r   Ú_infoÚ
Text._info0   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)r1   Ú
data_filesÚ
ValueErrorÚdownload_configÚextract_on_the_flyÚdownloadÚextractÚitemsÚ
iter_filesÚappendr   ÚSplitGenerator)	r2   Ú
dl_managerÚbase_data_filesÚextracted_data_filesÚsplitsÚ
split_nameÚfilesÚfiler6   s	   &&       r   Ú_split_generatorsÚText._split_generators3   sà   € 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#Ø 1°?ÀLÐRaÕRmÐnô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Text.__annotate__I   s'   ø€ ÷ Cñ C¡B§H¡Hð C±·±ñ Cr"   c                ó"  € V P                   P                  e¿   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# VP                  \        P                  ! R\        P                  ! 4       /4      4      # )Nc              3   óB   "  € T F  p\        V4      '       * x € K  	  R # 5i©Nr   )Ú.0Úfeatures   & r   Ú	<genexpr>Ú#Text._cast_table.<locals>.<genexpr>L   s   é € ÐbÑDa¸Ô+¨GÓ4×4Ñ4ÓDaùs   ‚FTÚtext)
r1   r   Úarrow_schemaÚallÚvaluesÚcastr
   rP   ÚschemaÚstring)r2   rN   r_   s   && r   Ú_cast_tableÚText._cast_tableI   s·   € Ø�;‰;×ÑÒ+Ø—[‘[×)Ñ)×6Ñ6ˆFß‹sÑbÀDÇKÁK×DXÑDX×D_ÑD_ÔDaÓb�s�sŠsÑbÀDÇKÁK×DXÑDX×D_ÑD_ÔDaÓb×bÒbà#Ÿ=™=¨Ó0�ð ˆOô & hÓ7�ØˆOà—=‘=¤§¢¨F´B·I²I³KÐ+@Ó!AÓBÐBr"   c              #  ó&   "  € T R j  x€L
  R #  L5irU   r#   )r2   r7   r6   s   &&&r   Ú_generate_shardsÚText._generate_shardsV   s   é € Ø×Ôùs   ‚ˆ‰c              #  ól  "  € V P                   P                  e    \        V P                   P                  4      MR.p\        V4       EFF  w  rEV EF9  p\	        W`P                   P
                  V P                   P                  R7      ;_uu_ 4       pV P                   P                  R8X  dü   ^ p VP                  V P                   P                  4      p	V	'       g   EM¢W—P                  4       ,          p	\        V	4      P                  4       p	V P                   P                  '       g    V	 U
u. uF  qªP                  R4      NK  	  p	p
\        P                   P#                  \        P$                  ! V	4      .VR7      p\'        WH4      V P)                  V4      3x € V^,          pKù  V P                   P                  R8X  Ed=   ^ pRp	 VP                  V P                   P                  4      pV'       g   M³Wœ,          p	W—P                  4       ,          p	V	P+                  R4      p	\        P                   P#                  \        P$                  ! V	R R
  Uu. uF  qÝ'       g   K  VNK  	  up4      .VR7      p\'        WH4      V P)                  V4      3x € V^,          pV	R
,          p	Kâ  V	'       dN   \        P                   P#                  \        P$                  ! V	.4      .VR7      pWH3V P)                  V4      3x € M�V P                   P                  R	8X  dg   VP                  4       p\        P                   P#                  \        P$                  ! V.4      .VR7      p\'        V^ 4      V P)                  V4      3x € R R R 4       EK<  	  EKI  	  R # u up
i u upi   + '       g   i     EKa  ; i5i)NrZ   )r   Úerrorsr   Ú
)Únamesr   Ú z

r   éÿÿÿÿ)r1   r   ÚlistÚ	enumerateÚopenr   r   r   Úreadr   Úreadliner   Ú	readlinesr   ÚrstriprP   rQ   Úfrom_arraysÚarrayr   ra   Úsplit)r2   r7   r6   Úpa_table_namesÚ	shard_idxÚfiles_iterablerJ   ÚfÚ	batch_idxÚbatchr   rN   Ú	new_batchÚexamplerZ   s   &&&            r   Ú_generate_tablesÚText._generate_tablesY   sÛ  é € Ø7;·{±{×7KÑ7KÒ7Wœ˜dŸk™k×2Ñ2Ô3Ð^dÐ]eˆÜ)2°?×)CÑ%ˆIÜ&�ä˜$¯©×)=Ñ)=ÀdÇkÁk×FaÑFa×bÕbÐfgØ—{‘{×,Ñ,°Ô6Ø$%˜	Ø"Ø$%§F¡F¨4¯;©;×+@Ñ+@Ó$A˜Eß#(Ù %Ø!§Z¡Z£\Õ1˜Eä$,¨U£O×$=Ñ$=Ó$?˜EØ#'§;¡;×#>×#>Ð#>ÙGLÓ(MÁu¸t¯©°TÖ):Áu Ð(MÜ')§x¡x×';Ñ';¼R¿XºXÀe»_Ð<MÐUcÐ';Ó'd˜Hô #& iÓ";¸T×=MÑ=MÈhÓ=WÐ"WÒWØ%¨�NšIØŸ™×.Ñ.°+Õ=Ø$%˜	Ø "˜Ø"Ø()¯©¨t¯{©{×/DÑ/DÓ(E˜Iß#,Ø %Ø!Õ.˜EØ!§Z¡Z£\Õ1˜EØ$)§K¡K°Ó$7˜EÜ')§x¡x×';Ñ';Ü!#§¢À%ÈÈÁ*Ó*XÁ*°wÓPW¯7¨7Á*Ñ*XÓ!YÐ ZÐbpð (<ó (˜Hô #& iÓ";¸T×=MÑ=MÈhÓ=WÐ"WÒWØ%¨�N˜IØ$)¨"¥IšEß Ü')§x¡x×';Ñ';¼R¿XºXÀuÀgÓ=NÐ<OÐWeÐ';Ó'f˜HØ#,Ð"8¸$×:JÑ:JÈ8Ó:TÐ"TÒTøØŸ™×.Ñ.°*Ô<Ø Ÿv™v›x˜Ü#%§8¡8×#7Ñ#7¼¿ºÀ4À&Ó9IÐ8JÐR`Ð#7Ó#a˜Ü! )¨QÓ/°×1AÑ1AÀ(Ó1KÐKÒK÷W cÒbô 'ó *Dùò )Nùò$ +Y÷; c×bÑbüsZ   ‚BN4ÂAN	ÃAN	Ä.NÅB.N	Ç6A N	É	NÉ$NÉ*AN	Ê2CN	ÎN4Î
N	ÎN1Î*
N4r#   N)r$   r%   r&   r'   r   ÚBUILDER_CONFIG_CLASSr3   rK   ra   rd   r~   r*   r+   r,   s   @r   r.   r.   -   s3   ø‡ € Ø%ÐòCò÷,Cð Cò÷0Lð 0Lr"   r.   )Údataclassesr   Úior   Útypingr   r   ÚpyarrowrP   r   Údatasets.builderr   Údatasets.features.featuresr	   Údatasets.tabler
   ÚutilsÚloggingÚ
get_loggerr$   ÚloggerÚBuilderConfigr   ÚArrowBasedBuilderr.   r#   r"   r   Ú<module>rŽ      ss   ðÝ !Ý ß $ã ã Ý  Ý ;Ý %ð 
�‰×	Ñ	×	*Ñ	*¨8Ó	4€ð ôA�×'Ñ'ó Aó ðAô8\Lˆ8×%Ñ%ö \Lr"   