Aj]_ddlZddlZddlZddlZddlmZddlmZddlZ ddl Z dZ dZ GddeZGdd eZGd d eZGd d eZGddeZdS)N)chain) itemgetter c(eZdZdZddZdZd dZdS) TextEdgeatDefines a text edge coordinates relative to a left-bottom origin. (PDF coordinate space) Parameters ---------- x : float x-coordinate of the text edge. y0 : float y-coordinate of bottommost point. y1 : float y-coordinate of topmost point. align : string, optional (default: 'left') {'left', 'right', 'middle'} Attributes ---------- intersections: int Number of intersections with horizontal text rows. is_valid: bool A text edge is valid if it intersections with at least TEXTEDGE_REQUIRED_ELEMENTS horizontal text rows. leftcZ||_||_||_||_d|_d|_dS)NrF)xy0y1align intersectionsis_valid)selfr r r rs `C:\Users\Terasoftware\OneDrive\Desktop\faahhh\fyndo\fyndo\venv\Lib\site-packages\camelot/core.py__init__zTextEdge.__init__.s1  c t|jd}t|jd}t|jd}d|d|d|d|jd|jd S)Nz )roundr r r rr)rr r r s r__repr__zTextEdge.__repr__6sl $&!   47A   47A   X1 X X" X X" X XTZ X X X X X r2ctj|j||r_|j|jz|zt |jdzz |_||_|xjdz c_|jt kr d|_dSdSdS)zdUpdates the text edge's x and bottom y coordinates and sets the is_valid attribute. atolTN)npiscloser rr floatTEXTEDGE_REQUIRED_ELEMENTSr)rr r edge_tols r update_coordszTextEdge.update_coords>s :dgr 1 1 1 %(461A5t?QTU?U9V9VVDFDG   ! #  !$>> % $  % %  % %rN)r r)__name__ __module__ __qualname____doc__rrr%rrrrsU0    % % % % % %rrcTeZdZdZd dZedZdZdZdZ dZ d Z d Z d S) TextEdgeszDefines a dict of left, right and middle text edges found on the PDF page. The dict has three keys based on the alignments, and each key's value is a list of camelot.core.TextEdge objects. rc*||_gggd|_dS)Nr rightmiddle)r$ _textedges)rr$s rrzTextEdges.__init__Rs  #%bAArcP|j}|j}|||z dz z}|||d}||S)zYReturns the x coordinate of a text row based on the specified alignment. g@)r r1r0)x0x1)textlinerx_leftx_rightx_middlex_coords r get_x_coordzTextEdges.get_x_coordVs? +Wv-44!XHHu~rct|j|D]%\}}tj|j|dr|cS&dS)zkReturns the index of an existing text edge using the specified x coordinate and alignment. g?rN) enumerater2r r!r )rr:rites rfindzTextEdges.findasTtu566  EArz"$c222  trc|||}|j}|j}t||||}|j||dS)z)Adds a new text edge to the current dict.)rN)r;r r rr2append)rr6rr r r r?s raddz TextEdges.addjs\   Xu - - [ [ aRu - - - %%b)))))rcdD]z}|||}|||}||||G|j||||j|j{dS)z2Updates an existing text edge in the current dict.r/N)r$)r;r@rCr2r%r r$)rr6rr:idxs rupdatezTextEdges.updaters0  E&&x77G))GU++C 5))))&s+99X[4=:   rc|D]N}t|dkr||OdS)zMGenerates the text edges dict based on horizontal text rows. rN)lenget_textstriprF)r textlinestls rgeneratezTextEdges.generate~sV  B2;;==&&(())A-  B  rc`td|jdDtd|jdDtd|jdDd}t|t d d }|j|S) zReturns the list of relevant text edges (all share the same alignment) based on which list intersects horizontal text rows the most. c32K|]}|j |jVdSNrr.0r?s r z)TextEdges.get_relevant..s@%'bk rr c32K|]}|j |jVdSrPrQrRs rrTz)TextEdges.get_relevant..s@%'r{ rr0c32K|]}|j |jVdSrPrQrRs rrTz)TextEdges.get_relevant..s@%'  rr1r/rkeyr)sumr2maxitemsr)rintersections_sumrelevant_aligns r get_relevantzTextEdges.get_relevants +/?6+B+/?7+C+/?8+D   .4466JqMMJJJ1M~..rcd}|di}|D]}|jr|sd||j|j|j|jf<)d}|D](}|j|dkr|j|dkr|}n)|d||j|j|j|jf<v|||dt |j|dt|d|jt|d|jf}d||<d} |D]} | | j| jz z } d}|D](}| j|dkr| j|dkr|}n)|||t | j|dt | j|dt|d| j t|d| jf}d||<| tt|z } i} |D]}d| ||| <| S) zpReturns a dict of interesting table areas on the PDF page calculated using relevant text edges. c|dtz }|dtz }|dtz}|d|dzz}||||fS)Nrrr)TABLE_AREA_PADDING)areaaverage_row_heightr4r r5r s rpadz&TextEdges.get_table_areas..padsRa--Ba--Ba--Ba-11BB# #rc |j |jfSrP)r r )r?s rz+TextEdges.get_table_areas..s~rrWNrrarr) sortrr r r popminrZr4r5r"rH) rrKrelevant_textedgesrf table_areasr?foundrd updated_areasum_textline_heightrLaverage_textline_heighttable_areas_paddeds rget_table_areaszTextEdges.get_table_areass}  $ $ $ $=$=>>> $ 9 9B{ 9"9>BKrubdBE :;; E +""5DG+"a0@"$(E!E 9BF RT25"$$>??#...!!HuQx00a"$//a"%00 ( 59 L1  1 1B 2525= 0 E#  5DG#a(8 EE 1&&&uQx((uQx((a"%((a"%((   -1 L)"5c)nn8M8M"M  J JDEI ss4)@AA B B!!rNr&) r'r(r)r*r staticmethodr;r@rCrFrMr^rsr+rrr-r-Ls BBBB\***      ///.G"G"G"G"G"rr-cjeZdZdZdZdZedZejdZedZ dS)Cella\Defines a cell in a table with coordinates relative to a left-bottom origin. (PDF coordinate space) Parameters ---------- x1 : float x-coordinate of left-bottom point. y1 : float y-coordinate of left-bottom point. x2 : float x-coordinate of right-top point. y2 : float y-coordinate of right-top point. Attributes ---------- lb : tuple Tuple representing left-bottom coordinates. lt : tuple Tuple representing left-top coordinates. rb : tuple Tuple representing right-bottom coordinates. rt : tuple Tuple representing right-top coordinates. left : bool Whether or not cell is bounded on the left. right : bool Whether or not cell is bounded on the right. top : bool Whether or not cell is bounded on the top. bottom : bool Whether or not cell is bounded on the bottom. hspan : bool Whether or not cell spans horizontally. vspan : bool Whether or not cell spans vertically. text : string Text assigned to cell. c||_||_||_||_||f|_||f|_||f|_||f|_d|_d|_ d|_ d|_ d|_ d|_ d|_dS)NF)r5r x2y2lbltrbrtr r0topbottomhspanvspan_textrr5r ryrzs rrz Cell.__init__sr(r(r(r(      rc t|j}t|j}t|j}t|j}d|d|d|d|d S)Nz >rN) r'r(r)r*rrpropertyrsetterrr+rrrvrvs''R"999X [..[.??X???rrvceZdZdZdZdZdZedZedZ dZ dd Z d Z d Z d Zd ZdZdZdZdZdS)TableaDefines a table with coordinates relative to a left-bottom origin. (PDF coordinate space) Parameters ---------- cols : list List of tuples representing column x-coordinates in increasing order. rows : list List of tuples representing row y-coordinates in decreasing order. Attributes ---------- df : :class:`pandas.DataFrame` shape : tuple Shape of the table. accuracy : float Accuracy with which text was assigned to the cell. whitespace : float Percentage of whitespace in the table. order : int Table number on PDF page. page : int PDF page number. c|_||_fd|D|_d|_d|_d|_d|_d|_d|_dS)Nc.g|]fdDS)c dg|],}t|dd|dd-Srr)rv)rScrs r z-Table.__init__...Ws7BBBtAaD!A$!ad33BBBrr+)rSrcolss @rrz"Table.__init__..Ws/QQQqBBBBTBBBQQQr)rrr) rrowscellsdfshapeaccuracy whitespaceorderpage)rrrs ` rrzTable.__init__TsZ  QQQQDQQQ     rc2d|jjd|jdS)N*@@4:@@@@rcn|j|jkr|j|jkrdS|j|jkrdSdS)NT)rr)rothers r__lt__z Table.__lt__bsI 9 " zEK' t 9uz ! 4  rc\g}|jD]!}|d|D"|S)z1Returns two-dimensional list of strings in table.c@g|]}|jSr+)rrJ)rScells rrzTable.data..ns$888Ddioo''888r)rrB)rdrows rdataz Table.dataisC : : :C HH88C888 9 9 9 9rctt|jdt|jd|j|jd}|S)zlReturns a parsing report with %accuracy, %whitespace, table number on page and page number. r)rrrr)rrrrr)rreports rparsing_reportzTable.parsing_reportqs?dmQ//33ZI     rc\|jD]#}|D]}dx|_x|_x|_|_$|S)zSets all table edges to True.T)rr r0rr)rrrs r set_all_edgeszTable.set_all_edgessO: G GC G GBFF FDJFDKK G rrc |D]3  fdt|jD} fdt|jD} fdt|jD}|si|d}|dgkrz|d}|r2|d} || kr#d|j||_|dz }|| k#t |j} || kr#d|j||_|dz }|| k#|gkrt |jdz }|r3|d} || kr#d|j||_|dz }|| k#Dt |j} || kr#d|j||_|dz }|| k#|d}|rN|d} || kr>d|j||_d|j||dz _|dz }|| k>t |j} || kr>d|j||_d|j||dz _|dz }|| k>5|D]3 fdt|jD} fdt|jD} fd t|jD}|si|d}|dgkrz|d}|r2|d} || kr#d|j||_|dz }|| k#t |j} || kr#d|j||_|dz }|| k#|gkrt |jdz }|r3|d} || kr#d|j||_|dz }|| k#Dt |j} || kr#d|j||_|dz }|| k#|d}|rN|d} || kr>d|j||_d|j|dz |_|dz }|| k>t |j} || kr>d|j||_d|j|dz |_|dz }|| k>5|S) aNSets a cell's edges to True depending on whether the cell's coordinates overlap with the line's coordinates within a tolerance. Parameters ---------- vertical : list List of detected vertical lines. horizontal : list List of detected horizontal lines. c`g|]*\}}tjd|d(|+Srrr r!)rSr>r joint_tolvs rrz#Table.set_edges..LAq:adAaDy999rc`g|]*\}}tjd|d(|+S)rarrr)rSjrrrs rrz#Table.set_edges..rrc`g|]*\}}tjd|d(|+Srrrr)rSkrrrs rrz#Table.set_edges..rrrTrc`g|]*\}}tjd|d(|+Srr)rSr>rhrs rrz#Table.set_edges..rrc`g|]*\}}tjd|d(|+Srr)rSrrrrs rrz#Table.set_edges..rrc`g|]*\}}tjd|d(|+S)rrrr)rSrrrrs rrz#Table.set_edges..rr) r=rrrr rHr0rr) rvertical horizontalrr>rrJLKrrs ` @@r set_edgeszTable.set_edgess: : A%di00A %di00A %di00A  !AQCx% aD !Aa%04 1 a(-Qa%DIAa%04 1 a(-Qa%b  NNQ& !Aa%15 1 a(.Qa%DIAa%15 1 a(.Qa%aD !Aa%04 1 a(-59 1 a!e,2Qa% DIAa%04 1 a(-59 1 a!e,2Qa% : : A%di00A %di00A %di00A  !AQCx% aD !Aa%/3 1 a(,Qa%DIAa%/3 1 a(,Qa%b  NNQ& !Aa%26 1 a(/Qa%DIAa%26 1 a(/Qa%aD !Aa%/3 1 a(,6: 1q5)!,3Qa% DIAa%/3 1 a(,6: 1q5)!,3Qa%  rctt|jD]G}d|j|d_d|j|t|jdz _Htt|jD]G}d|jd|_d|jt|jdz |_H|S)z Sets table border edges to True.Trr) rangerHrrr rr0rr)rrrs r set_borderzTable.set_border ss49~~&& ; ;A$(DJqM!  !6:DJqM#di..1, - 3 3s49~~&& < # # #!!!"( ! ! !   rrcleZdZdZdZdZdZdZedZ e dZ dd Z d Z dd ZdS) TableListzDefines a list of camelot.core.Table objects. Each table can be accessed using its index. Attributes ---------- n : int Number of tables in the list. c||_dSrP_tables)rtabless rrzTableList.__init__s  rc2d|jjd|jdS)Nrz n=r)rr'nrs rrzTableList.__repr__s"84>*88tv8888rc*t|jSrP)rHrrs r__len__zTableList.__len__s4<   rc|j|SrPr)rrEs r __getitem__zTableList.__getitem__s|C  rc(t|d|S)Nto_)getattr)tablers r _format_funczTableList._format_funcsuiAii(((rc t|SrP)rHrs rrz TableList.ns4yyrNc F|d}|d}|d}|jD]Y}|d|jd|j|}tj||}|||} | |ZdS)Ndirnamerootext-page-r)getrrrosrrr) rrrrrrr filenamefilepath to_formats r _write_filezTableList._write_files**Y''zz&!!jj\  EKKejKKKcKKHw||GX66H))%33I Ih       rc v|d}|d}|d}|d}tjtj||dz}t j|dd5}|jD]k}|d |jd |j |} tj|| } | | tj | l ddddS#1swxYwYdS) Nrrrr.ziprT allowZip64rr) rrrrrzipfileZipFilerrrrbasename) rrrrrrzipnamezr rrs r _compress_dirzTableList._compress_dirs\zz&!!**Y''zz&!!jj',,rwt44d;;fD _Wcd ; ; ; >q > >"OO%*OOU[O#OO7<<::"'"2"28"<"<==== > > > > > > > > > > > > > > > > > > >s,A4D..D25D2csvFctj|}tj|}tj|\}}|rt j}||||d}|dvr"|jdd|i||r|jdi|d Sd S|dkr*tj ||} tj | } |j D]1} d| j d| j} | j| | d2| |rtj tj||d z} t%j| d d 5}|| tj| d d d d S#1swxYwYd Sd S|dkrtj ||} |j D]} | | |rtj tj||d z} t%j| d d 5}|| tj| d d d d S#1swxYwYd Sd Sd S)aCExports the list of tables to specified file format. Parameters ---------- path : str Output filepath. f : str File format. Can be csv, excel, html, json, markdown or sqlite. compress : bool Whether or not to add files to a ZIP archive. )rrrr)r%htmljsonmarkdownrexcelrrrrrrTrNsqliter+)rrrr!splitexttempfilemkdtemprr$rrrrrrrrrrr rr)rrrcompressrr!rrrrrr rr"r#s rexportzTableList.exports'//$''7##D))G$$X.. c  )&((G7DMM 3 3 B D  + +q +F + + + -"",,V,,,,, - - '\ Bw||GX66H^H--F S SEUZEE EE !!&Z'!RRRR KKMMM B',,rwt'<' > >+B+B+B+B+B+Brr)rrrr- itertoolsroperatorrnumpyr pandasrr#rcobjectrr-rvrrr+rrr6s  4%4%4%4%4%v4%4%4%nX"X"X"X"X"X"X"X"vM?M?M?M?M?6M?M?M?`bbbbbFbbbJ `B`B`B`B`B`B`B`B`B`Br