§
    9²¸j  ã                   ó    — d Z ddlZddlmZmZ ddlmZmZmZ  e¦   «         r
ddl	Z	ddl	m
Z
  ej        e¦  «        Z G d„ de¦  «        ZdgZdS )	z'
Feature extractor class for MarkupLM.
é    Né   )ÚBatchFeatureÚFeatureExtractionMixin)Úis_bs4_availableÚloggingÚrequires_backends)ÚBeautifulSoupc                   ó@   ‡ — e Zd ZdZˆ fd„Zd„ Zd„ Zd„ Zdefd„Z	ˆ xZ
S )ÚMarkupLMFeatureExtractorao  
    Constructs a MarkupLM feature extractor. This can be used to get a list of nodes and corresponding xpaths from HTML
    strings.

    This feature extractor inherits from [`~feature_extraction_utils.PreTrainedFeatureExtractor`] which contains most
    of the main methods. Users should refer to this superclass for more information regarding those methods.

    c                 ó\   •— t          | dg¦  «          t          ¦   «         j        di |¤Ž d S )NÚbs4© )r   ÚsuperÚ__init__)ÚselfÚkwargsÚ	__class__s     €úˆ/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/transformers/models/markuplm/feature_extraction_markuplm.pyr   z!MarkupLMFeatureExtractor.__init__*   s8   ø€ Ý˜$  Ñ(Ô(Ð(Ø�‰ŒÔÐ"Ð"˜6Ð"Ð"Ð"Ð"Ð"ó    c           
      óª  ‡— g }g }|j         r|n|j        Š‰j        D ]‹}|                     ‰j         d¬¦  «        }|                     ‰j         ¦  «         |                     t          |¦  «        dk    rdn(t          ˆfd„t          |d¦  «        D ¦   «         ¦  «        ¦  «         |ŠŒŒ|                     ¦   «          |                     ¦   «          ||fS )NF)Ú	recursiveé   r   c              3   ó*   •K  — | ]\  }}|‰u ¯	|V — Œd S )Nr   )Ú.0ÚiÚsÚchilds      €r   ú	<genexpr>z6MarkupLMFeatureExtractor.xpath_soup.<locals>.<genexpr>6   s3   øè è € Ð1eÐ1e¹¸¸1ÐZ[Ð_dÐZdÐZd°!ÐZdÐZdÐZdÐZdÐ1eÐ1er   )	ÚnameÚparentÚparentsÚfind_allÚappendÚlenÚnextÚ	enumerateÚreverse)r   ÚelementÚ
xpath_tagsÚxpath_subscriptsr    Úsiblingsr   s         @r   Ú
xpath_soupz#MarkupLMFeatureExtractor.xpath_soup.   sî   ø€ Øˆ
ØÐØ"œ<Ð;��¨W¬^ˆØ”mð 	ð 	ˆFØ—’ u¤z¸U�ÑCÔCˆHØ×Ò˜eœjÑ)Ô)Ð)Ø×#Ò#Ý˜‘]”] aÒ'Ð'��­TÐ1eÐ1eÐ1eÐ1eÅ	È(ÐTUÑ@VÔ@VÐ1eÑ1eÔ1eÑ-eÔ-eñô ð ð ˆEˆEØ×ÒÑÔÐØ× Ò Ñ"Ô"Ð"ØÐ+Ð+Ð+r   c                 ó’  — t          |d¦  «        }g }g }g }|j        D ]Ç}t          |t          j        j        ¦  «        r¦t          |j        ¦  «        t          j        j        urŒGt          j
        |¦  «                             ¦   «         }|sŒp|                     |¦  «         |                      |¦  «        \  }}	|                     |¦  «         |                     |	¦  «         ŒÈt          |¦  «        t          |¦  «        k    rt          d¦  «        ‚t          |¦  «        t          |¦  «        k    rt          d¦  «        ‚|||fS )Nzhtml.parserz3Number of doc strings and xtags does not correspondz3Number of doc strings and xsubs does not correspond)r	   ÚdescendantsÚ
isinstancer   r(   ÚNavigableStringÚtyper    ÚTagÚhtmlÚunescapeÚstripr#   r,   r$   Ú
ValueError)
r   Úhtml_stringÚ	html_codeÚall_doc_stringsÚstring2xtag_seqÚstring2xsubs_seqr(   Útext_in_this_tagr)   r*   s
             r   Úget_three_from_singlez.MarkupLMFeatureExtractor.get_three_from_single=   sF  € Ý! +¨}Ñ=Ô=ˆ	àˆØˆØÐà Ô,ð 	:ð 	:ˆGÝ˜'¥3¤;Ô#>Ñ?Ô?ð :Ý˜œÑ'Ô'­s¬{¬Ð>Ð>Øå#'¤=°Ñ#9Ô#9×#?Ò#?Ñ#AÔ#AÐ Ø'ð Øà×&Ò&Ð'7Ñ8Ô8Ð8à/3¯ª¸wÑ/GÔ/GÑ,�
Ð,Ø×&Ò& zÑ2Ô2Ð2Ø ×'Ò'Ð(8Ñ9Ô9Ð9øåˆÑÔ¥3 Ñ#7Ô#7Ò7Ð7ÝÐRÑSÔSÐSÝˆÑÔ¥3Ð'7Ñ#8Ô#8Ò8Ð8ÝÐRÑSÔSÐSà Ð1AÐAÐAr   c                 ód   — d}t          ||¦  «        D ]\  }}|d|› �z  }|dk    r	|d|› d�z  }Œ|S )NÚ ú/r   ú[ú])Úzip)r   r)   r*   ÚxpathÚtagnameÚsubss         r   Úconstruct_xpathz(MarkupLMFeatureExtractor.construct_xpathZ   sV   € ØˆÝ  Ð-=Ñ>Ô>ð 	%ð 	%‰MˆG�TØ�]˜�]�]Ñ"ˆEØ�qŠyˆyØ˜˜T˜˜˜Ñ$�øØˆr   Úreturnc                 óÜ  — d}t          |t          ¦  «        rd}nLt          |t          t          f¦  «        r0t	          |¦  «        dk    st          |d         t          ¦  «        rd}|s t          dt          |¦  «        › d�¦  «        ‚t          |t          t          f¦  «        ot          |d         t          ¦  «        }|s|g}g }g }|D ]Š}|                      |¦  «        \  }}}	|                     |¦  «         g }
t          |||	¦  «        D ]1\  }}}|  
                    ||¦  «        }|
                     |¦  «         Œ2|                     |
¦  «         Œ‹||dœ}t          |d¬¦  «        }|S )	a\  
        Main method to prepare for the model one or several HTML strings.

        Args:
            html_strings (`str`, `list[str]`):
                The HTML string or batch of HTML strings from which to extract nodes and corresponding xpaths.

        Returns:
            [`BatchFeature`]: A [`BatchFeature`] with the following fields:

            - **nodes** -- Nodes.
            - **xpaths** -- Corresponding xpaths.

        Examples:

        ```python
        >>> from transformers import MarkupLMFeatureExtractor

        >>> page_name_1 = "page1.html"
        >>> page_name_2 = "page2.html"
        >>> page_name_3 = "page3.html"

        >>> with open(page_name_1) as f:
        ...     single_html_string = f.read()

        >>> feature_extractor = MarkupLMFeatureExtractor()

        >>> # single example
        >>> encoding = feature_extractor(single_html_string)
        >>> print(encoding.keys())
        >>> # dict_keys(['nodes', 'xpaths'])

        >>> # batched example

        >>> multi_html_strings = []

        >>> with open(page_name_2) as f:
        ...     multi_html_strings.append(f.read())
        >>> with open(page_name_3) as f:
        ...     multi_html_strings.append(f.read())

        >>> encoding = feature_extractor(multi_html_strings)
        >>> print(encoding.keys())
        >>> # dict_keys(['nodes', 'xpaths'])
        ```FTr   zQHTML strings must of type `str`, `list[str]` (batch of examples), but is of type ú.)ÚnodesÚxpathsN)ÚdataÚtensor_type)r/   ÚstrÚlistÚtupler$   r6   r1   r=   r#   rC   rG   r   )r   Úhtml_stringsÚvalid_stringsÚ
is_batchedrK   rL   r7   r9   r:   r;   Úxpath_stringsÚnodeÚtag_listÚsub_listÚxpath_stringrM   Úencoded_inputss                    r   Ú__call__z!MarkupLMFeatureExtractor.__call__b   s±  € ð` ˆõ �l¥CÑ(Ô(ð 	%Ø ˆMˆMÝ˜¥t­U mÑ4Ô4ð 	%Ý�<Ñ Ô  AÒ%Ð%­°LÀ´OÅSÑ)IÔ)IÐ%Ø $�àð 	Ýð8Ý"& |Ñ"4Ô"4ð8ð 8ð 8ñô ð õ
   ­tµU¨mÑ<Ô<ÐcÅ*È\ÐZ[Ì_Õ^aÑBbÔBbˆ
àð 	*Ø(˜>ˆLð ˆØˆØ'ð 	)ð 	)ˆKØAE×A[ÒA[Ð\gÑAhÔAhÑ>ˆO˜_Ð.>Ø�LŠL˜Ñ)Ô)Ð)ØˆMÝ,/°ÀÐRbÑ,cÔ,cð 3ð 3Ñ(��h Ø#×3Ò3°H¸hÑGÔG�Ø×$Ò$ \Ñ2Ô2Ð2Ð2Ø�MŠM˜-Ñ(Ô(Ð(Ð(ð ¨&Ð1Ð1ˆÝ%¨4¸TÐBÑBÔBˆàÐr   )Ú__name__Ú
__module__Ú__qualname__Ú__doc__r   r,   r=   rG   r   r[   Ú__classcell__)r   s   @r   r   r       s–   ø€ € € € € ðð ð#ð #ð #ð #ð #ð,ð ,ð ,ðBð Bð Bð:ð ð ðT¨ð Tð Tð Tð Tð Tð Tð Tð Tr   r   )r_   r3   Úfeature_extraction_utilsr   r   Úutilsr   r   r   r   r	   Ú
get_loggerr\   Úloggerr   Ú__all__r   r   r   ú<module>rf      sÔ   ððð ð €€€à LÐ LÐ LÐ LÐ LÐ LÐ LÐ LØ AÐ AÐ AÐ AÐ AÐ AÐ AÐ AÐ AÐ Að ÐÑÔð "Ø€J€J€JØ!Ð!Ð!Ð!Ð!Ð!ð 
ˆÔ	˜HÑ	%Ô	%€ðVð Vð Vð Vð VÐ5ñ Vô Vð Vðr &Ð
&€€€r   