o j]_@sddlZddlZddlZddlZddlmZddlmZddlZ ddl Z dZ dZ GdddeZGdd d eZGd d d eZGd d d eZGdddeZdS)N)chain) itemgetter c@s,eZdZdZd ddZddZd dd Zd S) 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. leftcCs(||_||_||_||_d|_d|_dS)NrF)xy0y1align intersectionsis_valid)selfrr r r rR/var/www/html/fyndo/pharma/fyndo/venv/lib/python3.10/site-packages/camelot/core.py__init__.s  zTextEdge.__init__c CsLt|jd}t|jd}t|jd}d|d|d|d|jd|jd S)Nz )roundrr r r r )rrr r rrr__repr__6s   &zTextEdge.__repr__2cCsbtj|j||dr-|j|j|t|jd|_||_|jd7_|jtkr/d|_dSdSdS)zdUpdates the text edge's x and bottom y coordinates and sets the is_valid attribute. atolTN)npiscloser r rfloatTEXTEDGE_REQUIRED_ELEMENTSr )rrr edge_tolrrr update_coords>s   zTextEdge.update_coordsN)rr)__name__ __module__ __qualname____doc__rrr rrrrrs  rc@sVeZdZdZdddZeddZddZd d Zd d Z d dZ ddZ ddZ dS) 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. rcCs||_gggd|_dS)Nrrightmiddle)r _textedges)rrrrrrRszTextEdges.__init__cCs0|j}|j}|||d}|||d}||S)zYReturns the x coordinate of a text row based on the specified alignment. g@)rr)r()x0x1)textliner x_leftx_rightx_middlex_coordrrr get_x_coordVs  zTextEdges.get_x_coordcCs6t|j|D]\}}tj|j|ddr|SqdS)zkReturns the index of an existing text edge using the specified x coordinate and alignment. g?rN) enumerater*rrr)rr1r iterrrfindas zTextEdges.findcCs<|||}|j}|j}t||||d}|j||dS)z)Adds a new text edge to the current dict.)r N)r2r r rr*append)rr-r rr r r5rrraddjs z TextEdges.addcCsZdD](}|||}|||}|dur|||q|j||j||j|jdqdS)z2Updates an existing text edge in the current dict.r'N)r)r2r6r8r*r r r)rr-r r1idxrrrupdaters   zTextEdges.updatecCs,|D]}t|dkr||qdS)zMGenerates the text edges dict based on horizontal text rows. rN)lenget_textstripr:)r textlinestlrrrgenerate~s  zTextEdges.generatecCsjtdd|jdDtdd|jdDtdd|jdDd}t|td 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. cs|] }|jr|jVqdSNr r .0r5rrr  z)TextEdges.get_relevant..rcsrArBrCrDrrrrFrGr(csrArBrCrDrrrrFrGr)r'rkeyr)sumr*maxitemsr)rintersections_sumrelevant_alignrrr get_relevants zTextEdges.get_relevantc Csdd}|jdddi}|D]a}|jrq|s$d||j|j|j|jf<qd}|D]}|j|dkr<|j|dkr<|}nq(|durNd||j|j|j|jf<q|||d t|j|dt|d |jt|d|jf}d||<qd } |D]N} | | j| j7} d}|D]}| j|dkr| j|dkr|}nq|dur||t| j|d t| j|dt|d | j t|d| jf}d||<qv| t t |} i} |D] }d| ||| <q| S) zpReturns a dict of interesting table areas on the PDF page calculated using relevant text edges. cSs@|dt}|dt}|dt}|d|d}||||fS)Nrrr)TABLE_AREA_PADDING)areaaverage_row_heightr+r r,r rrrpads    z&TextEdges.get_table_areas..padcSs|j |jfSrB)r r)r5rrrsz+TextEdges.get_table_areas..rHNrrPrr) sortr rr r popminrKr+r,rr;) rr>relevant_textedgesrU table_areasr5foundrS updated_areasum_textline_heightr?average_textline_heighttable_areas_paddedrrrget_table_areass\   zTextEdges.get_table_areasNr!) r"r#r$r%r staticmethodr2r6r8r:r@rOrarrrrr&Ls      r&c@sFeZdZdZddZddZeddZejddZed d Z d S) 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. cCsn||_||_||_||_||f|_||f|_||f|_||f|_d|_d|_ d|_ d|_ d|_ d|_ d|_dS)NF)r,r x2y2lbltrbrtrr(topbottomhspanvspan_textrr,r rerfrrrrs     z Cell.__init__c CsFt|j}t|j}t|j}t|j}d|d|d|d|d S)Nz Ws*z-Table.__init__...r)rEcolsr{rr}Wsz"Table.__init__..)rrr) rrowscellsdfshapeaccuracy whitespaceorderpage)rrrrr~rrTs zTable.__init__cCd|jjd|jdS)N.)rr7)rdrowrrrdatais z Table.datacCs&t|jdt|jd|j|jd}|S)zlReturns a parsing report with %accuracy, %whitespace, table number on page and page number. r)rrrr)rrrrr)rreportrrrparsing_reportqs  zTable.parsing_reportcCs2|jD]}|D]}d|_|_|_|_qq|S)zSets all table edges to True.T)rrr(rkrl)rrrrrr set_all_edgess zTable.set_all_edgesrc s|D]fddt|jD}fddt|jD}fddt|jD}|s.q|d}|dgkrp|d}|rV|d} || krUd|j||_|d7}|| ksEqt|j} || krod|j||_|d7}|| ks_q|gkrt|jd}|r|d} || krd|j||_|d7}|| ksqt|j} || krd|j||_|d7}|| ksq|d}|r|d} || krd|j||_d|j||d_|d7}|| ksqt|j} || krd|j||_d|j||d_|d7}|| ksq|D] fddt|jD}fd dt|jD}fd dt|jD}|s-q|d}|dgkru|d}|rY|d} || krXd|j||_|d7}|| ksGqt|j} || krtd|j||_|d7}|| kscq|gkrt|jd}|r|d} || krd|j||_|d7}|| ksqt|j} || krd|j||_|d7}|| ksq|d}|r|d} || krd|j||_d|j|d|_|d7}|| ksqt|j} || kr d|j||_d|j|d|_|d7}|| ksq|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. cs,g|]\}}tjd|ddr|qSrrrrrEr4rt joint_tolvrrr} z#Table.set_edges..c,g|]\}}tjd|ddr|qS)rPrrrrEjrtrrrr}rcrrrrrrEkrtrrrr}rrTrc,g|]\}}tjd|ddr|qSrrrhrrrr}rcs,g|]\}}tjd|ddr|qSrrrrrrr}rcr)rrrrrrrrr}r) r3rrrrr;r(rkrl) rvertical horizontalrr4rrJLKr)rrrr set_edgess                            zTable.set_edgescCstt|jD]}d|j|d_d|j|t|jd_qtt|jD]}d|jd|_d|jt|jd|_q&|S)z Sets table border edges to True.Trr) ranger;rrrrr(rkrl)rr|rzrrr set_border szTable.set_bordercCs|jD]v}|D]q}|j}|j}|j}|j}|jdkrq|jdkrP|s,|r,|r,|r,d|_q|s8|r8|r8|r8d|_q|sD|rD|rD|rDd|_q|sO|rO|rO|rOd|_q|jdkrm|ra|ra|sa|sad|_q|rl|rl|sl|sld|_q|jdvrxd|_d|_qq|S)zSets a cell's hspan or vspan attribute to True depending on whether the cell spans horizontally or vertically. rrPTrry)rrr(rkrlrurmrn)rrrrr(rkrlrrrset_spans<     zTable.set_spancKs0ddddd}|||jj|fi|dS)zWrites Table to a comma-separated values (csv) file. For kwargs, check :meth:`pandas.DataFrame.to_csv`. Parameters ---------- path : str Output filepath. utf-8Fr)encodingindexheaderquotingN)r:rto_csv)rpathkwargskwrrrr6s z Table.to_csvcKs^ddi}|||jjdi|}t|d}||WddS1s(wYdS)zWrites Table to a JSON file. For kwargs, check :meth:`pandas.DataFrame.to_json`. Parameters ---------- path : str Output filepath. orientrecordswNr)r:rto_jsonopenwrite)rrrr json_stringfrrrrEs    "z Table.to_jsoncKsNd|jd|jdd}||t|}|jj|fi||dS)zWrites Table to an Excel file. For kwargs, check :meth:`pandas.DataFrame.to_excel`. Parameters ---------- path : str Output filepath. page--table-r sheet_namerN)rrr:pd ExcelWriterrto_excelsave)rrrrwriterrrrrVs    zTable.to_excelcKP|jjdi|}t|ddd}||WddS1s!wYdS)zWrites Table to an HTML file. For kwargs, check :meth:`pandas.DataFrame.to_html`. Parameters ---------- path : str Output filepath. rrrNr)rto_htmlrr)rrr html_stringrrrrrj  "z Table.to_htmlcKr)zWrites Table to a Markdown file. For kwargs, check :meth:`pandas.DataFrame.to_markdown`. Parameters ---------- path : str Output filepath. rrrNr)r to_markdownrr)rrr md_stringrrrrryrzTable.to_markdowncKs\ddd}||t|}d|jd|j}|jj||fi|||dS)zWrites Table to sqlite database. For kwargs, check :meth:`pandas.DataFrame.to_sql`. Parameters ---------- path : str Output filepath. replaceF) if_existsrrrN) r:sqlite3connectrrrto_sqlcommitclose)rrrrconn table_namerrr to_sqlites   zTable.to_sqliteN)r)r"r#r$r%rrrrvrrrrrrrrrrrrrrrrrx7s(      rxc@sdeZdZdZddZddZddZdd Zed d Z e d d Z dddZ ddZ dddZdS) TableListzDefines a list of camelot.core.Table objects. Each table can be accessed using its index. Attributes ---------- n : int Number of tables in the list. cCs ||_dSrB_tables)rtablesrrrr zTableList.__init__cCr)Nrz n=r)rr"nrqrrrrrzTableList.__repr__cCs t|jSrB)r;rrqrrr__len__rzTableList.__len__cCs |j|SrBr)rr9rrr __getitem__rzTableList.__getitem__cCst|d|S)Nto_)getattr)tablerrrr _format_funcszTableList._format_funccCst|SrB)r;rqrrrrsz TableList.nNc Ksl|d}|d}|d}|jD]!}|d|jd|j|}tj||}|||} | |qdS)Ndirnamerootext-page-r)getrrrosrrsr) rrrrrrrfilenamefilepath to_formatrrr _write_files      zTableList._write_filec Ks|d}|d}|d}|d}tjtj||d}tj|ddd.}|jD]!}|d |jd |j |} tj|| } | | tj | q-WddS1sZwYdS) Nrrrr.ziprT allowZip64rr) rrrrsrzipfileZipFilerrrrbasename) rrrrrrzipnamezrrrrrr _compress_dirs     "zTableList._compress_dircsvFcCstj|}tj|}tj|\}}|rt}||||d}|dvr=|jdd|i||r;|jdi|d Sd S|dkrtj ||} t | } |j D]} d| j d| j} | jj| | ddqP| |rtj tj||d } tj| d d d }|| tj| Wd d S1swYd Sd S|dkrtj ||} |j D]} | | q|rtj tj||d } tj| d d d }|| tj| Wd d S1swYd Sd Sd S)aCExports the list of tables to specified file format. Parameters ---------- path : str Output filepath. f : str File format. Can be csv, excel, html, json, markdown or sqlite. compress : bool Whether or not to add files to a ZIP archive. )rrrr)rhtmljsonmarkdownrexcelrrrrrrTrNsqliter)rrrrsplitexttempfilemkdtemprrrsrrrrrrrrrrrr)rrrcompressrrrrrrrrrrrrrrexportsF   "  "zTableList.exportrB)rF)r"r#r$r%rrrrrbrrvrrrrrrrrrs      r)rrrr itertoolsroperatorrnumpyrpandasrrrRobjectrr&rcrxrrrrrs$  7Pg