o e<@sdZddlmZddlmZddlmZddlmZddl m Z ddl m Z dd l m Z dd lmZmZdd lmZGd d d ZdS)a=Parsing table blocks. * ``lattice table``: explicit borders represented by strokes. * ``stream table`` : borderless table recognized from layout of text blocks. Terms definition: * From appearance aspect, we say ``stroke`` and ``fill``, the former looks like a line, while the later an area. * From semantic aspect, we say ``border`` (cell border) and ``shading`` (cell shading). * An explicit border is determined by a certain stroke, while a stroke may also represent an underline of text. * An explicit shading is determined by a fill, while a fill may also represent a highlight of text. * Border object is introduced to determin borders of stream table. Border instance is a virtual border adaptive in a certain range, then converted to a stroke once finalized, and finally applied to detect table border. ) constants)Element) Collection)Blocks)Shapes)Lines)TableStructure)BorderBorders)Cellc @seZdZdZddZdededefddZdeded efd d Zed e fd dZ ed e de de de fddZ eddZed e de fddZdS)TablesConstructorz7Object parsing ``TableBlock`` for specified ``Layout``.cCs||_|j|_|j|_dSN)_parentblocks_blocksZshapes_shapes)selfparentrPD:\Projects\ConvertPro\env\Lib\site-packages\pdf2docx/table/TablesConstructor.py__init__$s zTablesConstructor.__init__connected_border_tolerancemin_border_clearancemax_border_widthc Cs|jsdSdtfdd}|jjj||d}||}|jj}t}||d}|D] } || j} t| fi| |  } | rG| | | q'|j ||j|dS)acParse table with explicit borders/shadings represented by rectangle shapes. Args: connected_border_tolerance (float): Two borders are intersected if the gap lower than this value. min_border_clearance (float): The minimum allowable clearance of two borders. max_border_width (float): Max border width. N instancescSs`dd}t||}g}|D]}t|dkr|d}n t|ddd}|d}||q|S)z4Delete group when it's contained in a certain group.cSs|j|jp |j|jSr)bboxcontains)abrrr:szJTablesConstructor.lattice_tables..remove_overlap..rcSs |jSr)rZget_area)instancerrrr Ds )key)rgrouplensortedappend)rZfungroupsZ unique_groupsZgroup_instancesr"Z sorted_grouprrrremove_overlap7s   z8TablesConstructor.lattice_tables..remove_overlap)ZdxZdyrr)rlist table_strokesZgroup_by_connectivitytable_fillingsrcontained_in_bboxrr parseto_table_blockZset_lattice_table_blockr(rassign_to_tables) rrrrr*Zgrouped_strokesZfillstablessettingsstrokesZ group_fillstablerrrlattice_tables*s(    z TablesConstructor.lattice_tablesline_separate_thresholdcsjj}jj}j||}jj\}}fdd} t} ||d} |D]} | s.q)tdd| D} tdd| D}t dd| D}t dd| D}| ||\}}| |||f}||||f}t ||}t |}||j}|j|jtjd \}}|s|st | rq)| |||}|sq)|t|fi| |}tjtr|j|jd kr|d d jd urq)|| |q)j| j| d S) zParse table with layout of text/image blocks, and update borders with explicit borders represented by rectangle shapes. Refer to ``lattice_tables`` for arguments description. csP}}jD]}|jj|kr|jj}|jj|kr#|jj}||fSq||fS)afind the vertical boundaries of table in y-range [y0, y1]: - the bottom of block closest to y0 - the top of block closest to y1 ``` +-------------------------+ <- Y0 +--------------+ +--------------+ <- y_lower +------------------------+ <- y0 | table | +------------------------+ <- y1 +-------------------------+ <- y_upper +-------------------------+ +---------------------------+ <- Y1 ``` )rry1y0)r:r9Zy_lowerZy_upperblockY0Y1rrrtop_bottom_boundaries}s   z>TablesConstructor.stream_tables..top_bottom_boundariesr+cSg|]}|jjqSr)rx0.0rectrrr z3TablesConstructor.stream_tables..cSr@r)rr:rBrrrrErFcSr@r)rx1rBrrrrErFcSr@r)rr9rBrrrrErF) thresholdrr!N) rr-r.rZcollect_stream_linesrrrminmaxr _outer_bordersrZ update_bboxr/Zsplit_with_intersectionr FACTOR_A_FEW_is_simple_structure_stream_strokesZsort_in_reading_orderr r0r1 isinstancer Znum_colsZnum_rowsZbg_colorZset_stream_table_blockr(r2)rrrr8r-r.Z tables_linesX0X1r?r3r4Z table_linesrAr:rGr9Z y0_marginZ y1_margin inner_bbox outer_bbox outer_bordersrDexplicit_strokesexplicit_shadings_r5r6rr<r stream_tablesjsJ !      "  zTablesConstructor.stream_tableslinescCs<t|}|dkr dS|dkrt|t|kSdS)zWhether current lines represent a simple table: * only one column -> always flow layout in docx; or * two columns: lines are aligned in each row -> simple paragraph in docx rTrF)r&group_by_columnsZgroup_by_physical_rows group_by_rows)rYnumrrrrMs z&TablesConstructor._is_simple_structurerTrUrVcCsTt}||t||}|||||t}|D] }||q|S)avParsing borders mainly based on content lines contained in cells, and update borders (position and style) with explicit borders represented by rectangle shapes. Args: lines (Lines): lines contained in table cells. outer_borders (tuple): Boundary borders of table, ``(top, bottom, left, right)``. explicit_strokes (Shapes): Showing borders in a stream table; can be empty. explicit_shadings (Shapes): Showing shadings in a stream table; can be empty. Returns: Shapes: Parsed strokes representing table borders. )r extendr _inner_bordersfinalizerr(Z to_stroke)rYrTrUrVbordersZ inner_bordersresborderrrrrNs    z!TablesConstructor._stream_strokescCs|\}}}}|\}}}} td||fdd} td|| fdd} td||fdd} td||fdd} | | | f| | | f| | | f| | | f| | | | fS)aiInitialize outer Border instances according to lower and upper bbox-es. :: +---------------------------------> | | Y0 +------------------------+ + outer bbox | | | | | | y0+----------------+ | | | | | | +<----+ | | | +<--------+ inner bbox | | y1+----------------+ | | | x0 x1 | | Y1 +------------------------+ | X0 X1 v ZHTF) border_range referenceZHBZVLZVR)r Zset_boundary_borders)rRrSrAr:rGr9rPr=rQr>topbottomleftrightrrrrKs   z TablesConstructor._outer_bordersc Cst|}dd|D}t|}|dk}|dkr|D]}d|_qt}d}|\} } } } t|D]} | dkr6| n|}| |dkrA| }n|| jj}|| djj}td||f| | fdd }| ||| }t|}|dkrmq.d}t|D]C}|dkr{| n|}||dkr| }n||jj }||djj }td ||f||f|d }| |t ||||||f}||qsq.|S) aCalculate the surrounding borders of given ``lines``. These borders construct table cells. Two purposes of stream table: * Rebuild layout, e.g. text layout with two columns, and * parsing real borderless table. It's controdictory that the former needn't to deep into row level, just ``1xN`` table convenient for layout recreation; instead, the later should, ``MxN`` table for each cell precisely. So, the principle determining stream tables borders: * Vertical borders contributes the table structure, so ``border.is_reference=False``. * Horizontal borders are for reference when ``n_column=2``, in this case ``border.is_reference=True``. * During deeper recursion, h-borders become outer borders: it turns valuable when count of detected columns >= 2. Args: lines (Lines): Lines in table cells. outer_borders (tuple): Boundary borders of table region. cSsg|] }|jtjdqS))factor)r[rrL)rCZ col_linesrrrrE>sz4TablesConstructor._inner_borders..rFNr!rZVI)Z border_typercr`rdZHI)rZr& is_referencer rangerrGrAr r(r9r:r r^r])rYrTZ cols_linesZ group_linesZcol_numrjrbr`rhZTOPZBOTTOMLEFTRIGHTirgrArGZ rows_linesZrow_numrfjrer:r9Zborders_rrrr^&sP         z TablesConstructor._inner_bordersN)__name__ __module__ __qualname____doc__rfloatr7rX staticmethodrrMtuplerrNrKr^rrrrr !s2 @ h" !r N)rscommonrZcommon.ElementrZcommon.CollectionrZ layout.BlocksrZ shape.ShapesrZ text.Linesrr r r r r rrrrs