kh<dZddlmZddlmZddlmZddlmZddl m Z ddl m Z dd l m Z dd lmZmZdd lmZGd d Zy)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 eZdZdZdZdededefdZdededefdZed e fd Z ed e d e d e d e fdZ edZed e d e fdZy)TablesConstructorz7Object parsing ``TableBlock`` for specified ``Layout``.cV||_|j|_|j|_yN)_parentblocks_blocksshapes_shapes)selfparents \/var/www/teggl/fontify/venv/lib/python3.12/site-packages/pdf2docx/table/TablesConstructor.py__init__zTablesConstructor.__init__$s }} }} connected_border_tolerancemin_border_clearancemax_border_widthc$|jsydtfd}|jjj||}||}|jj}t }||d}|D]j} |j | j} t| fi|j| j} | sJ| j|j| l|jj||jj|y)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 instancescd}t|j|}g}|D]:}t|dk(r|d}nt|d}|d}|j |<|S)z4Delete group when it's contained in a certain group.c|jj|jxs%|jj|jSr)bboxcontains)abs rzJTablesConstructor.lattice_tables..remove_overlap..:s-qvvqvv6Q!&&//!&&:Qrr c6|jjSr)r#get_area)instances rr'zJTablesConstructor.lattice_tables..remove_overlap..DsX]]-C-C-Er)key)rgrouplensortedappend)r fungroups unique_groupsgroup_instancesr+ sorted_groups rremove_overlapz8TablesConstructor.lattice_tables..remove_overlap7s|RC *005FM#) /'*.q1H$*/E$GL+B/H$$X. /! r)dxdyrr)rlist table_strokesgroup_by_connectivitytable_fillingsrcontained_in_bboxr#r parseto_table_blockset_lattice_table_blockr1rassign_to_tables) rrrrr7grouped_strokesfillstablessettingsstrokes group_fillstables rlattice_tablesz TablesConstructor.lattice_tables*s||V !T !*,,44 " "&@E_ " ` )9 ++$8 0 ' %G11',,?K#77h7==kJYY[E--/ e$ % %%f- %%f-rline_separate_thresholdcjj}jj}jj ||}j j \}}fd} t} ||d} |D]!} | st| D cgc]} | j jc} }t| D cgc]} | j jc} }t| D cgc]} | j jc} }t| D cgc]} | j jc} }| ||\}}||||f}||||f}tj||}t!j#|} |j%| j }|j'| j t(j*\}}|s|stj-| rfj/| |||}|s~|j1t3|fi| j5|j7}t9j t:r0|j<|j>zdk(r|ddj@|jC| jE|$jjG| jjG| ycc} wcc} wcc} wcc} w)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. c}}jD]e}|jj|kr|jj}|jj|kDsL|jj}||fS||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 ``` )rr#y1y0)rPrOy_lowery_upperblockY0Y1rs rtop_bottom_boundariesz>TablesConstructor.stream_tables..top_bottom_boundaries}sw* "2WG ::==2%w::==2%#jjmmGG# # G# #rr:) thresholdr r(N)$rr<r>rcollect_stream_linesrr#rminx0rPmaxx1rOr_outer_bordersr update_bboxr?split_with_intersectionr FACTOR_A_FEW_is_simple_structure_stream_strokessort_in_reading_orderr r@rA isinstancer num_colsnum_rowsbg_colorset_stream_table_blockr1rC)rrrrLr<r> tables_linesX0X1rVrFrG table_linesrectrZrPr\rO y0_margin y1_margin inner_bbox outer_bbox outer_bordersexplicit_strokesexplicit_shadings_rHrJrTrUs` @@r stream_tableszTablesConstructor.stream_tablesjs~ 22 44||88I`a **BB $B$8 0  (' !K{;tdiill;OP__aE$,,-u~~-q0U1Xa[5I5I5Q  ( ( * MM% O' !T %%f- %%f-Q<;;;sK <K )K% K* linesct|j}|dk(ry|dk(r3t|jt|jk(Sy)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 r TrF)r/group_by_columnsgroup_by_physical_rows group_by_rows)rwnums rraz&TablesConstructor._is_simple_structuresR %((*+ 6 !Vu3356E | | Y0 +------------------------+ + outer bbox | | | | | | y0+----------------+ | | | | | | +<----+ | | | +<--------+ inner bbox | | y1+----------------+ | | | x0 x1 | | Y1 +------------------------+ | X0 X1 v HTF) border_range referenceHBVLVR)r set_boundary_borders)rprqrZrPr\rOrjrTrkrUtopbottomleftrights rr]z TablesConstructor._outer_borderss$$BB#BBB8uEB8uEB8uEB8uE   $/##T5M2 !!3-0 ""C=1VT5))rc z|j}|Dcgc]"}|jtj$}}t |}|dk}|dk\r|D] }d|_ t }d} |\} } } } t|D]3}|dk(r| n| }||dz k(r| } nY||jj}||dzjj}td||f| | fd} |j| ||}t |}|dk(rd}t|D]}|dk(r| n|}||dz k(r| }nY||jj}||dzjj}td ||f|| f|}|j|tj!|||||| f}|j#|6|Scc}w) 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 contradictory 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. )factorrFNr(r VI) border_typerrrHI)ryr{rr`r/ is_referencer ranger#r\rZr r1rOrPrrr~)rwrr cols_lines col_lines group_linescol_numrrrrTOPBOTTOMLEFTRIGHTirrZr\ rows_linesrow_numrjrrPrOborders_s rrz TablesConstructor._inner_borders&s .++- _ijR[y..i6L6L.Mj jj/z  A:' ,&+# ,)#0 VT5w- )Aa44UD'!)|UU]''**!_)),,4"$b &M#%u%%QJ*oG!|XF7^ )Tcvgai<&#A++..B#AaC--00B $&("X!%u ".0FNN6*-;;JqMCQWY]_dKefx(- )/- )^Aks'F8N)__name__ __module__ __qualname____doc__rfloatrKrv staticmethodrratuplerrbr]rrrrr!sA% =.+0=.%*=."'=.@e.%*e.!&e.).e.P 5  e56ekD**BWUW%WWrrN)rcommonrcommon.Elementrcommon.Collectionr layout.Blocksr shape.Shapesr text.Linesrr r r r rrrrrs3&$*"!*#]]r