/j-jSrSSKJr SSKrSSKrSSKrSSKrSSKrSSKrSSK r SSK J r SSK J r SSK Jr SSKJr SSKrSSKrSSKr\R,S:a SS KJr SS KJr O SS KJr SS KJr S S KJr S S KJr S SKJr S SKJ r S SKJ!r! Sr"Sr#/SQr$/SQr%\$\%-r&"SS5r'"SS\'5r("SS5r)"SS\)5r*"SS5r+"SS 5r,S&S'S!jjr-"S"S#\5r."S$S%5r/g)(z6Contains the core functions to parse tables from PDFs.) annotationsN)Iterable)Iterator) itemgetter)Any) ) TypedDict)Unpack)ImageConversionBackend) undo_rotation)build_file_path_in_temp_dir)get_index_closest_point)get_textline_coords leftrightmiddle)topbottomcenterc*\rSrSrSrSrSrSrSrg) TextAlignment1aRepresents a list of textlines sharing an alignment on a coordinate. The alignment can be left/right/middle or top/bottom/center. (PDF coordinate space) Parameters ---------- coord : float coordinate of the initial text edge. Depending on the alignment it could be a vertical or horizontal coordinate. textline : obj the original textline to start the alignment align : str Name of the alignment (e.g. "left", "top", etc) Attributes ---------- coord : float The coordinate aligned averaged out across textlines. It can be along the x or y axis. textlines : array Array of textlines that demonstrate this alignment. align : str Name of the alignment (e.g. "left", "top", etc) c,XlU/UlX0lgN)coord textlinesalignselfr textliner"s I/home/mande/repo/quber/.venv/lib/python3.13/site-packages/camelot/core.py__init__TextAlignment.__init__Ls " cSR[SURSS55RSS5nSURS[ UR5SUS 3$) Nz | c"UR5$r)get_textxs r&(TextAlignment.__repr__..Ss !**,r) z)joinmapr!replacer len)r$ text_insides r&__repr__TextAlignment.__repr__Qsijj &r(: ; '$  tzzl$s4>>/B.CD*m5 2 r)cUR[UR5-U-[[UR5S-5- UlURR U5 g)z%  ! #C    h'r))r"r r!N) __name__ __module__ __qualname____firstlineno____doc__r'r9r>__static_attributes__r)r&rr1s4  (r)rc<^\rSrSrSrU4SjrSrSSjrSrU=r $)TextEdgedaDefines a text edge coordinates relative to a left-bottom origin. (PDF coordinate space) An edge is an alignment bounded over a segment. Parameters ---------- coord : float coordinate of the text edge. Can be x or y. y0 : float y-coordinate of bottommost point. y1 : float y-coordinate of topmost point. align : string, optional (default: 'left') {'left', 'right', 'middle'} Attributes ---------- is_valid: bool A text edge is valid if it intersects with at least TEXTEDGE_REQUIRED_ELEMENTS horizontal text rows. cx>[TU]XU5 URUlURUlSUlg)NF)superr'y0y1is_valid)r$r r%r" __class__s r&r'TextEdge.__init__}s/ %0++++ r)c [URS5n[URS5n[URS5nSUSUSUSURSUR S3 $)Nr1z )roundr rLrMr"rN)r$r.rLrMs r&r9TextEdge.__repr__sd $**a  477A  477A 1#T"T"WTZZL VW X r)c[R"URURUS9(aHURX!5 URUl[ UR 5[ :aSUlggg)zrUpdate text edge coordinates. Update the text edge's x and bottom y coordinates and sets the is_valid attribute. abs_tolTN)mathiscloserLr>r7r!TEXTEDGE_REQUIRED_ELEMENTSrN)r$r.r%edge_tols r& update_coordsTextEdge.update_coordss\ <<h ?  * *8 7kkDG4>>"&@@ $ A @r))rNrLrM2) r@rArBrCrDr'r9r]rE __classcell__rOs@r&rHrHds0  %%r)rHc:\rSrSrSrSr\S5rSrSr Sr g) TextAlignmentsz9Defines a dict of text edges across reference alignments.cB0UlUHn/URU'M gr_text_alignments)r$alignment_namesalignment_names r&r'TextAlignments.__init__s$ "-N46D ! !. 1.r)c[XU5$r)r)r r%r"s r&_create_new_text_alignment)TextAlignments._create_new_text_alignmentsUe44r)c[$r)NotImplementedr$ alignmentr r%s r&_update_alignment TextAlignments._update_alignmentsr)cj[U5nURR5Hup4X#n[XTSS9nSnUcSnOFXFRnUS- Us=:a US-:aO OUR XFXQ5 O X:aUS-nOUnUcMgUR XQU5n URXy5 M g)z1Update an existing text edge in the current dict.cUR$r)r r-s r&r/3TextAlignments._register_textline..sQWWr))fnNrg?r )rrhitemsrr rsrminsert) r$r%coords alignment_idalignment_arrayr idx_closest idx_insert coord_closest new_alignments r&_register_textline!TextAlignments._register_textlines$X.-1-B-B-H-H-J )L(E2+<K J" / < B B 3;<<**'4e#*!,qJ!,J% $ ? ?\!  &&zA9.Kr)rgN) r@rArBrCrDr' staticmethodrmrsrrErFr)r&rdrds)C7 55Br)rdcr^\rSrSrSrSU4SjjrSrSrSrSr Sr S r S r S r S rS rSrSrU=r$) TextEdgeszDefines a dict text edges on the PDF page. The dict contains the 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. c8>[TU][5 Xlgr)rKr'HORIZONTAL_ALIGNMENTSr\)r$r\rOs r&r'TextEdges.__init__s ./ r)c[XU5$r)rHr#s r&rm$TextEdges._create_new_text_alignments//r)cdURXU5nURURU5 g)z(Add a new text edge to the current dict.N)rmrhr=)r$r r%r"tes r&add TextEdges.adds-  , ,Ue D e$++B/r)c<URX#UR5 gr)r]r\rqs r&rsTextEdges._update_alignments?r)cUHAn[UR5R55S:dM0URU5 MC g)z)TextEdges.get_relevant..s',Ib[[!BLL!!,Is66rc3r# UH-nUR(dM[UR5v M/ g7frrrs r&rrs,8B;;"BLL!!877rc3r# UH-nUR(dM[UR5v M/ g7frrrs r&rrs,9B;;"BLL!!9rrrr keyrcUR$r)rNrs r&r/(TextEdges.get_relevant.. sbkkr))sumrhmaxryrlistfilter)r$intersections_sumrelevant_aligns r& get_relevantTextEdges.get_relevants,0,A,A&,I//8 //9 &.446JqMJ1M )4+@+@+P Q  r)cURSS9 URU5nURX15 URU5nUVs0sHoPR XT5S_M nnU$s snf)a Return a dict of interesting table areas on the PDF page. The table areas are calculated using relevant text edges. Parameters ---------- textlines : list List of text line objects that are relevant for determining table areas. relevant_textedges : list List of relevant text edge objects used to identify table areas. Returns ------- dict A dictionary with padded table areas as keys and None as values. c4UR*UR4$r)rLr rs r&r/+TextEdges.get_table_areas.. s/Ar)rN)sort_initialize_table_areas"_extend_table_areas_with_textlines"_calculate_average_textline_height_pad)r$r!relevant_textedges table_areasaverage_textline_heightareapadded_table_areass r&get_table_areasTextEdges.get_table_areas s}& $AB223EF  // G#'"I"I)"TGR GRtIId 4d :{  "!  sA'c0nUHMnU(d2SX#RURURUR4'M<URX#5 MO U$)a  Initialize table areas based on relevant text edges. Parameters ---------- relevant_textedges : list List of relevant text edge objects used to initialize table areas. Returns ------- dict A dictionary of table areas initialized from relevant text edges. N)r rLrM_update_table_areas)r$rrrs r&r!TextEdges._initialize_table_areas-sM $BBF XXruubhh>?((9 % r)cSnUH/nURUS:dMURUS::dM-Un O Uc1SXRURURUR4'gURU5 US[ URUS5[ USUR5[ USUR54nSX'g)z Update table areas by checking for overlaps with new text edges. Parameters ---------- table_areas : dict Current table areas to be updated. te : object The new text edge object to check for overlaps. Returns ------- None Nr rrr1)rMrLr popminr)r$rrfoundr updated_areas r&rTextEdges._update_table_areasDsDuuQBEET!W$4  =>BK255"((BEE: ; OOE "aBEE58$E!Hbhh'E!Hbee$ L )-K %r)c|UHnSnUH/nURUS:dMURUS::dM-Un O UcM?URU5 [URUS5[URUS5[ USUR 5[ USUR54nSX'M g)z Extend table areas based on text lines that overlap vertically. Parameters ---------- table_areas : dict Current table areas to be extended. textlines : list List of text line objects relevant for extending table areas. Returns ------- None Nr rrr1)rLrMrrx0rx1)r$rr!rrrrs r&r,TextEdges._extend_table_areas_with_textlinesfsBE#55DG#a(8 E $  &uQx(uQx(a"%%(a"%%(   -1 )!r)cf[SU55nU(aU[[U55- $S$)z Calculate the average height of text lines. Parameters ---------- textlines : list List of text line objects. Returns ------- float The average height of the text lines. c3R# UHoRUR- v M g7fr)rMrL)rrs r&r?TextEdges._calculate_average_textline_height..s!D)B%%"%%-)s%'r)rr<r7)r$r!sum_textline_heights r&r,TextEdges._calculate_average_textline_heights1"!D)!DD>G"U3y>%::NQNr)chUS[- nUS[- nUS[-nUSUS--nX4XV4$)z Pad a given area by a constant value. Parameters ---------- area : tuple The area to be padded defined as (x0, y0, x1, y1). average_row_height : float The average height of rows to use for padding. Returns ------- tuple The padded area. rr r1r)TABLE_AREA_PADDING)r$raverage_row_heightrrLrrMs r&rTextEdges._padsP !W) ) !W) ) !W) ) !W)A- -r))r\r_)r@rArBrCrDr'rmrrsrrrrrrrrrErarbs@r&rrsN!00 @,  <"@. -D1BO"  r)rc\rSrSrSrSrSr\S5r\RS5r\S Sj5r \S Sj5r \S 5r S r g ) Cellia^Defines a cell in a table. With coordinates relative to a left-bottom origin. (PDF coordinate space) Parameters ---------- x1 : float x-coordinate of left-bottom point. y1 : float y-coordinate of left-bottom point. x2 : float x-coordinate of right-top point. y2 : float y-coordinate of right-top point. Attributes ---------- lb : tuple Tuple representing left-bottom coordinates. lt : tuple Tuple representing left-top coordinates. rb : tuple Tuple representing right-bottom coordinates. rt : tuple Tuple representing right-top coordinates. left : bool Whether or not cell is bounded on the left. right : bool Whether or not cell is bounded on the right. top : bool Whether or not cell is bounded on the top. bottom : bool Whether or not cell is bounded on the bottom. text : string Text assigned to cell. cXlX lX0lX@lX4UlX4UlX24UlX44UlSUlSUl SUl SUl SUl g)NFr3) rrMx2y2lbltrbrtrrrr_textr$rrMrrs r&r' Cell.__init__s]((((    r)c [UR5n[UR5n[UR5n[UR5nSUSUSUSUS3 $)Nz >r)) rrrrrrrrrrrrMrN)returnbool)r@rArBrCrDr'r9propertyrsetterrrrrErFr)r&rrsz%N 9 [[..////??r)rc\rSrSrSrSrSr\S5r\S5r Sr Sr \S 5r \S S j5r \S 5rS rS rS!SjrSrSrSrSrSrSrS"SjrSrSrSrSrSrSrSrSr Sr!g)#Tableia|Defines a table with coordinates relative to a left-bottom origin. (PDF coordinate space) Parameters ---------- cols : list List of tuples representing column x-coordinates in increasing order. rows : list List of tuples representing row y-coordinates in decreasing order. Attributes ---------- df : :class:`pandas.DataFrame` shape : tuple Shape of the table. accuracy : float Accuracy with which text was assigned to the cell. whitespace : float Percentage of whitespace in the table. filename : str Path of the original PDF order : int Table number on PDF page. page : int PDF page number. cXlX lUVVs/sH-o1Vs/sHn[USUSUSUS5PM snPM/ snnUl[R "5UlSUlSUlSUl SUl SUl SUl SUl SUlSUlSUlSUlSUlSUlSUlSUlSUlSUlSUlSUlgs snfs snnf)Nrr )rrr3F)colsrowsrcellspd DataFramedfshapeaccuracy whitespacefilenameorderpagerotationflavorpdf_size_bboxparse parse_details_image _image_path_rows_np_cache_cols_np_cache_neg_y_tops_list_col_widths_list_rows_disjoint)r$rrrcs r&r'Table.__init__&s  LPQDqTBTtAaD!A$!ad3TBDQ ,,.          ! #" !%!% $OCQs C4$C/ C4/C4cH^URcURm[R"T[RS9R SS5UlTVs/sHoS*PM snUl[U4Sj[[T5S- 555Ul URcoURn[R"U[RS9R SS5Ul UVs/sHn[USUS- 5PM snUlggs snfs snf)a+Populate the search caches used by ``get_table_index``. Idempotent: callers may invoke this every call (it short-circuits on the first attribute that is already populated). The NumPy arrays are part of the private interface and kept available so future call sites can do their own bulk processing. N)dtyper1rc3L># UHnTUSTUS-S:v M g7f)r rNrF)rirs r&r-Table._build_search_caches..es.&6JQ d1q5k!n,6Js!$r )r rnpasarrayfloat64reshaperallranger7rrrabsr)r$rrrrs @r&_build_search_cachesTable._build_search_cachesRs    &99D"$**T"D"L"LRQR"SD 59$9DqdUD$9D ! #&&6;CIM6J&#D     &99D"$**T"D"L"LRQR"SD ?C$Dt!S1!%5t$DD ! '%:%Es D2DcTURcUR5 UR$)zReturn rows as an ``(n_rows, 2)`` float64 NumPy array. Lazily constructed on first access and cached. Each row is ``(y_top, y_bot)`` and rows preserve their original descending ``y_top`` order. )r r"rs r&_rows_npTable._rows_npms(    &  % % '"""r)cTURcUR5 UR$)zReturn cols as an ``(n_cols, 2)`` float64 NumPy array. Lazily constructed on first access and cached. Each column is ``(x_left, x_right)``. )rr"rs r&_cols_npTable._cols_npys(    &  % % '"""r)cPSURRSURS3$)zXReturn a string representation of the class . Returns ------- [type] [description] >**+74::,a@@r)cURUR:XaURUR:agURUR:agg)zReturn True if the two pages are less than the current page . Parameters ---------- other : [type] [description] Returns ------- [type] [description] TN)rr)r$others r&__lt__ Table.__lt__s? 99 "zzEKK' 99uzz ! "r)c/nURH<nURUVs/sHo3RR5PM sn5 M> U$s snf)z1Returns two-dimensional list of strings in table.)rr=rr)r$drowcells r&data Table.datasE ::C HHC8CDiioo'C8 99s#A cZ[SURS- SURS- - -5$)uA unified per-table quality score in ``[0.0, 1.0]``. Computed from the existing per-flavor signals as ``(accuracy / 100) * (1 - whitespace / 100)``. The intent is a single number suitable for production filtering and automated validation — ``confidence >= 0.8`` works as a reasonable first-cut threshold; tune for the source PDFs. Components and their meaning are identical across flavors: - ``accuracy`` (0-100): how well the detected cells line up with the parser's structural hints (line joints for lattice, text alignments for stream/network/hybrid). Higher is better. - ``whitespace`` (0-100): percentage of cells that are empty after stripping. Lower is better (a perfectly populated table is 0; a mostly-empty one trends toward 100). - ``confidence`` (0-1): the composite. ``accuracy=90``, ``whitespace=10`` → ``confidence≈0.81``; either signal going to its worst value pulls ``confidence`` to 0. See #659. Y@g?)rrrrs r& confidenceTable.confidences-03.359P3PQRRr)c[URS5[URS5URUR[UR S5S.$)uPer-table parsing report. Standard keys across all flavors: ================= ============================================== ``page`` 1-based page number the table was found on. ``order`` 1-based rank within that page (left-to-right / top-to-bottom). ``accuracy`` Float in ``[0, 100]``. See ``confidence`` for component-by-component definitions. ``whitespace`` Float in ``[0, 100]``. ``confidence`` Float in ``[0, 1]``. Unified quality score — combines accuracy and whitespace. ================= ============================================== See #659. r1r)rrrrr:)rTrrrrr:rs r&parsing_reportTable.parsing_reportsF(dmmQ/3ZZII3   r)cURcURcp[[RR UR 5S5Ul[SS9nURUR URURS9 [[R"UR5UR5UlUR$)zCompute pdf image and cache it.z.pngT) use_fallback)r)r r rospathbasenamerr convertrrcv2imreadr)r$backends r& get_pdf_imageTable.get_pdf_images ;; '#>GG$$T]]3V$ 1dC t/?/?diiP' 43C3C(DdmmTDK{{r)cvURH(nUHnS=Ul=Ul=UlUlM! M* U$)zSet all table edges to True.T)rrrrr)r$r3r4s r& set_all_edgesTable.set_all_edgess=::CBFF FDJFDK r)cJURX5 URX#5 U$)aqSet the edges of the joint. Set 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. joint_tol : int, optional Tolerance for determining proximity, by default 2 )_set_vertical_edges_set_horizontal_edges)r$vertical horizontal joint_tols r& set_edgesTable.set_edgess%   5 "":9 r)cn[U5H&upE[R"X%SUS9(dM$Us $ g)NrrW) enumeraterYrZ)r$r{r rRrrs r&_find_close_pointTable._find_close_points2f%DA||EQ4;;&r)c4UHnURURUSU5nUcM(URURUSU5nUc[UR5nURURUSU5nUR XEU5 M g)Nrr r)rWrr7r_update_vertical_edges)r$rPrRvstartendrs r&rNTable._set_vertical_edges sA**499adIFE}((AaD)DC{$))n&&tyy!A$ BA  ' 'A 6r)cUcI[UR5S- nUS:a*[X5HnSURUUlM ggUS:Xa*[X5HnSURUSlM g[X5H4nSURUUlSURUUS- lM6 gNr rT)r7rr rrrr$r\r]indexjs r&rZTable._update_vertical_edgess = NQ&Ezu*A15DJJqM%(.+aZ5&(, 1 a %'5&,0 1 e$)15 1 eai(.'r)c4UHnURURUSU5nUcM(URURUSU5nUc[UR5nURURUSU5nUR XEU5 M g)Nrr1r )rWrr7r_update_horizontal_edges)r$rQrRhr\r]rs r&rOTable._set_horizontal_edges&sA**499adIFE}((AaD)DC{$))n&&tyy!A$ BA  ) )%a 8r)cUcI[UR5S- nUS:a*[X5HnSURUUlM ggUS:Xa*[X5HnSURSUlM g[X5H4nSURUUlSURUS- UlM6 gr`)r7rr rrrras r&rfTable._update_horizontal_edges3s = NQ&Ezu*A26DJJu%a(/+aZ5&'+ 1 a $'5&+/ 5!!$(26 519%a(/'r)c^[UR5n[UR5mUS:XdTS:XaU$[UR5U:wd$[ U4SjUR55(a [ S5e[ U5H4nSURUSlSURUTS- lM6 [ T5H4nSURSUl SURUS- Ul M6 U$)z Sets table border edges to True.rc3@># UHn[U5T:gv M g7frr7)rr3num_colss r&r#Table.set_border..Ks. ,6SCH JszQInconsistent cells structure: cells should match the dimensions of rows and cols.Tr ) r7rrrany ValueErrorr rrrr)r$num_rows row_index col_indexrns @r& set_borderTable.set_borderAstyy>tyy> q=HMK tzz?h &#. ,0JJ. + + c  xI,0DJJy !! $ )8nSnUH)nUR U5nUcMU"5(dM'SnM+ U(aM= U$ U$)a Copies over text in empty spanning cells. Parameters ---------- copy_text : list of str, optional (default: None) Select one or more of the following strings: {'h', 'v'} to specify the direction in which text should be copied over when a cell spans multiple rows or columns. Returns ------- camelot.core.Table The updated table with copied text in spanning cells. Notes ----- Iterates the directional copy passes until the table is stable. A single pass-per-direction misses cells spanned in *both* directions (a 2D span): the source cell from which the 2D-spanned cell would copy hasn't itself been filled yet, so the empty string propagates through. Repeating the chosen passes until no cell changes converges in O(spans) iterations and fixes the symptom reported in #349. )rgr[r r1FT)_copy_horizontal_text_copy_vertical_textr rr7rget)r$ copy_textpasses_changed directionpass_fns r&copy_spanning_textTable.copy_spanning_text^s4  K118P8PQs3tzz?A.23AG&  **Y/&799"G'7 4 r)cSn[[UR55Hn[[URU55HnURUUnURR 5S:XdM5UR (dMHUR (aM[URUUS- RnXTR:wdMXTlSnM M U$)uCopy text rightwards into empty horizontally-spanned cells. Fills empty cells that span horizontally with the text from the left adjacent cell. Returns ``True`` when at least one cell's text was changed — used by :meth:`copy_spanning_text` to decide whether another pass is needed. Fr3r T)r r7rrrrrr$r~rrcr4sources r&rxTable._copy_horizontal_textss4::'A3tzz!}-.zz!}Q'99??$*tzzz$)))!ZZ]1q5166F*$* "& /(r)cSn[[UR55Hn[[URU55HnURUUnURR 5S:XdM5UR (dMHUR (aM[URUS- URnXTR:wdMXTlSnM M U$)uCopy text downwards into empty vertically-spanned cells. Fills empty cells that span vertically with the text from the top adjacent cell. Returns ``True`` when at least one cell's text was changed — used by :meth:`copy_spanning_text` to decide whether another pass is needed. Fr3r T)r r7rrrrrrs r&ryTable._copy_vertical_textss4::'A3tzz!}-.zz!}Q'99??$*tzzz$(((!ZZA.q166F*$* "& /(r)c nSSSSS.nURU5 URR"U40UD6 g)zWrite Table(s) to a comma-separated values (csv) file. For kwargs, check :meth:`pandas.DataFrame.to_csv`. Parameters ---------- path : str Output filepath. utf-8Fr )encodingrbheaderquotingN)updaterto_csv)r$rBkwargskws r&r Table.to_csvs3"EUq Q & t"r"r)c URSS5nSS0nURU5 URR"S0UD6n[ X5nUR U5 SSS5 g!,(df  g=f)aWrite Table(s) to a JSON file. For kwargs, check :meth:`pandas.DataFrame.to_json`. The optional ``mode`` kwarg (``"w"`` to overwrite, ``"a"`` to append) is consumed by the file-open call; the rest are forwarded to ``DataFrame.to_json`` (#317). Parameters ---------- path : str Output filepath. mode : str, optional (default: 'w') File open mode. Pass ``"a"`` to append to an existing file rather than overwrite it. modeworientrecordsNrF)rrrto_jsonopenwrite)r$rBrrr json_stringfs r&r Table.to_jsonsa"zz&#& " &ggoo++ $  GGK   s A** A8c &URSS5nSSS.nSURSUR3nURU5 [R "XS9nUR R"U4SU0UD6 S S S 5 g !,(df  g =f) uWrite Table(s) to an Excel file. For kwargs, check :meth:`pandas.DataFrame.to_excel`. The optional ``mode`` kwarg is forwarded to :class:`pandas.ExcelWriter` (``"w"`` to overwrite, ``"a"`` to append a new sheet to an existing workbook) — see #317. Parameters ---------- path : str Output filepath. mode : str, optional (default: 'w') ExcelWriter open mode. Use ``"a"`` to add a new sheet to an existing workbook (requires openpyxl). rrF)rbrpage--table-)r sheet_nameN)rrrrr ExcelWriterrto_excel)r$rBrrrrwriters r&rTable.to_excels~"zz&#& .TYYKwtzzl;  &^^D , GG  V A Ab A- , ,s  B Bc URSS5nURR"S0UD6n[XSS9nUR U5 SSS5 g!,(df  g=f)aTWrite Table(s) to an HTML file. For kwargs, check :meth:`pandas.DataFrame.to_html`. The optional ``mode`` kwarg is consumed by the file-open call (#317). Parameters ---------- path : str Output filepath. mode : str, optional (default: 'w') File open mode. Pass ``"a"`` to append to an existing file rather than overwrite it. rrrrNrF)rrto_htmlrr)r$rBrr html_stringrs r&r Table.to_htmlsM zz&#&ggoo// $w /1 GGK 0 / / A A"c URSS5nURR"S0UD6n[XSS9nUR U5 SSS5 g!,(df  g=f)uWrite Table(s) to a Markdown file. For kwargs, check :meth:`pandas.DataFrame.to_markdown`. The optional ``mode`` kwarg is consumed by the file-open call — passing ``mode="a"`` appends every successive call to the same file rather than overwriting (#317). Parameters ---------- path : str Output filepath. mode : str, optional (default: 'w') File open mode. Pass ``"a"`` to append to an existing file rather than overwrite it. rrrrNrF)rr to_markdownrr)r$rBrr md_stringrs r&rTable.to_markdown sO"zz&#&GG''1&1 $w /1 GGI 0 / /rc SSS.nURU5 [R"U5nSURSUR3nUR R "XT40UD6 UR5 UR5 g)zWrite Table(s) to sqlite database. For kwargs, check :meth:`pandas.DataFrame.to_sql`. Parameters ---------- path : str Output filepath. r6F) if_existsrbrrN) rsqlite3connectrrrto_sqlcommitclose)r$rBrrconn table_names r& to_sqliteTable.to_sqlite"si%u 5 &t$TYYKwtzzl;  z.2.  r))rrrr r rrr rrrrrrrrr r rrrrr)rr<)r1r)"r@rArBrCrDr'r"rr%r(r9r/r5r:r=rHrKrSrWrNrZrOrfrurrxryrrrrrrrErFr)r&rrs>*$XE6 # ###A&SS2  4 ( 7 6 9 7:)V&& #!0BB!*,r)rc@[U5S:Xa[R"US5$[R"US5nUR/nUSSGH}n[R"U5nURGbURGbURSnURSnXg- nUR V V s/sH upX-X-4PM sn n UlUR Hn U Hn U =RU- slU =RU- slU RU R4U l U RU R4U l U RU R4U l U RU R4U lM M URSURSU-URSURSU-4UlUR UR -UlUR UR -Ul[URSURS5URS[!URSURS5URS4UlURn U(a-U R"(dU R$SSR'SS9n UR)U 5 GM [*R,"USS9UlUVs/sHn[/US S5cMUR0PM! nnUVs/sHn[/US S5cMUR2PM! nnU(a[5U5[U5- UlU(a[5U5[U5- UlU$s sn n fs snfs snf) uVertically concatenate a run of contiguous Tables (#628 POC). All tables in ``run`` must share the same column count — the caller (:meth:`TableList.stack_contiguous`) is responsible for that invariant. A single-element run is returned via a deep copy so the output never aliases the input. Geometry: * The first table's ``rows`` / ``cells`` / ``_bbox`` stay in place. Each subsequent table's y-coordinates are shifted so its top sits at the previous table's bottom — the stacked result reads as one tall page even though source pages differ. ``parsing_report`` is mean-aggregated across the run (accuracy, whitespace, confidence are averaged); ``page`` and ``order`` are taken from the first table. r rNrr1T)drop) ignore_indexrr)r7copydeepcopyrrrrrMrrrrrrrrremptyiloc reset_indexr=rconcatgetattrrrr)rundrop_repeated_headeroutout_dfsnxtnxt_copy out_bottomnxt_topdyr0r1r3r4nxt_dfraccswsss r& _vstack_runr6s& 3x1}}}SV$$ --A CvvhG12w==% 99 X^^%?1JnnQ'G%B@H N HRbgrw/ NHM~~DGGrMGGGrMG $ww0DG#ww0DG#ww0DG#ww0DG &q!q!B&q!q!B& HN xx(--/CH HNN2CICIIaL(.."34q!CIIaL(.."34 ! CI  [[_00d0;Fv_bYYwT 2CF !$ P1wq*d'CJAJJD P!$ SA<(F<1<<C S 4y3t9,  SCH, JYOJ Q Ss;N4N N N6NcB\rSrSr%SrS\S'S\S'S\S'S\S'S rg ) _Kwiz.Helper class to define file related arguments.zos.PathLike[Any] | strrBstrdirnamerootextrFN)r@rArBrCrD__annotations__rErFr)r&rrs8  L I Hr)rc\rSrSrSrSSjrSrSrSSjrSSjr SSjr \ S 5r \ SS j5rSSS jjrSSS jjrSSSjjrSSjrSSSjjrSrg ) TableListiaDefines a list of camelot.core.Table objects. Each table can be accessed using its index. Attributes ---------- n : int Number of tables in the list. Examples -------- >>> from camelot.core import TableList >>> tables = TableList([]) >>> tables.n 0 >>> tables c$[U5Ulgr)r_tables)r$tabless r&r'TableList.__init__s%)L r)cPSURRSURS3$)Nr+z n=rS)rOr@nrs r&r9TableList.__repr__s&4>>**+3tvvha88r)c,[UR5$r)r7rrs r&__len__TableList.__len__s4<<  r)c URU$rr)r$idxs r& __getitem__TableList.__getitem__s||C  r)c,[UR5$r)iterrrs r&__iter__TableList.__iter__sDLL!!r)c[U5$r)nextrs r&__next__TableList.__next__s Dzr)c [USU35$)Nto_)r)tablers r& _format_funcTableList._format_funcsuA3i((r)c[U5$)z!The number of tables in the list.rmrs r&r TableList.ns4yr)cURVs/sHSnURSU:dMURSU:dM-URU:dM?URU::dMQUPMU nn[ U5$s snf)uTReturn a new TableList keeping only tables that pass all thresholds. A post-extraction convenience for dropping noise / low-quality tables (single stray cells, mostly-empty regions, …). Parsing is unchanged — everything is still detected; this just selects from the result. Every threshold defaults to a no-op, so calling ``filter()`` with no arguments returns an equivalent list and a legitimate single-row or single-column table is never dropped unless you ask for it. Parameters ---------- min_rows : int, optional (default: 1) Drop tables with fewer than this many rows. min_columns : int, optional (default: 1) Drop tables with fewer than this many columns. min_accuracy : float, optional (default: 0.0) Drop tables whose ``parsing_report`` accuracy (0-100) is below this value. max_whitespace : float, optional (default: 100.0) Drop tables whose ``parsing_report`` whitespace (0-100) is above this value. Returns ------- camelot.core.TableList A new list (the original is left untouched), so calls compose: ``tables.filter(min_rows=2).filter(min_accuracy=90)``. rr )rrrrr)r$min_rows min_columns min_accuracymax_whitespacerkepts r&rTableList.filtersN %{{1~)  A+- ,.    N2 %   sA6A6A6A6#A6cpUS;a[SU<35eUR(d [/5$/n/nURGH5nU(dURU5 MUSn[ UR 5[ UR 5:HnU=(a US:g=(d} UR R(d`UR R(dE[UR RS5[UR RS5:HOSnU(a!US:XdU(aURU5 GMUR[UUS:H=(a U(+S95 U/nGM8 U(a*UR[UUS:H=(a U(+S95 [U5$) uVertically stack tables that look like continuations across pages. Many PDF reports break a single logical table over several pages — a header on every page, a footer on every page, body rows in between. ``read_pdf`` returns one :class:`Table` per page; this helper stitches contiguous ones back together so the resulting :class:`TableList` has one entry per *logical* table instead of one per *physical* page. Parameters ---------- match : str, optional (default: 'column_count') How to decide whether two adjacent tables are continuations of each other. * ``'column_count'`` — same number of columns (the rule from #628's POC; the common case). * ``'first_row'`` — same column count *and* identical text in the first row (catches PDFs that repeat the header on every page). keep_first_header : bool, optional (default: False) When ``match='first_row'``, the matching first row of every continuation table is dropped (so the stacked table has exactly one header row). Set to ``True`` to keep every page's header row in the stacked output. Returns ------- camelot.core.TableList A new TableList with continuation runs collapsed. Tables that don't continue from the previous one (different column count or different first row, depending on ``match``) are passed through unchanged. The originals in ``self`` are not mutated. Notes ----- Originally proposed by @TimothyOfDelphi in #628. Consolidates #8 / #133 / #357 / #531. Limitations: * The stacked :class:`Table`'s ``page`` keeps the *first* stitched table's page number; ``order`` keeps the first table's order. Callers iterating with both shouldn't be surprised by a missing row-of-pages. * ``parsing_report`` is averaged: ``accuracy`` and ``whitespace`` are mean-aggregated across the stitched tables; ``confidence`` is recomputed from the averaged accuracy + whitespace. * Cell geometry (``_bbox``, ``cells``, ``rows``) is preserved via the y-shift trick from #628's POC, so downstream plotting on a single stitched table still works page-locally. Cells from the second-and-later tables are shifted to sit below the first table's bottom. ) column_count first_rowz1match must be 'column_count' or 'first_row', got rrrFr)r) rqrrr=r7rrrrrr) r$matchkeep_first_headerstitchedrrhead same_cols same_headers r&stack_contiguousTableList.stack_contiguoussoz 5 5CE9M ||R= "\\E 5!r7DDII#ejj/9I#$ 77==a)T%((--2B-CC e~5 5!.3{.B.2 11g1"2  OO*/;*>*.-- ""r)Nc USnUSnUSnURH[nUSURSURU3n[RR X75nUR Xa5n U "U5 M] g)Nrrr-page-r)rrrrArBr4r) r$rrrrrrrfilepath to_formats r& _write_fileTableList._write_filehsx#f~Um\\Evejj\ cUKHww||G6H))%3I h  "r)c USnUSnUSnUSn[RR[RRU5U5S-n[R "USSS9nUR HqnUS URS URU3n [RRX95n URU [RRU 55 Ms SSS5 g!,(df  g=f) NrBrrr.ziprT allowZip64rr) rArBr4rzipfileZipFilerrrrrC) r$rrBrrrzipnamezrrrs r& _compress_dirTableList._compress_dirrsf~#f~Um'',,rwwt4d;fD __Wcd ;q"V6%**WU[[M#O77<<:"''"2"28"<=&< ; ;s +BC66 Dc[RRU5n[RRU5n[RR U5upgU(a[ R "5nXXgS.nUS;a/UR"SSU0UD6 U(aUR"S0UD6 g g US:XGa$[RRXE5n [R"U 5n URH8n SU RSU R3n U RR!XS9 M: U R#5 U(a[RR[RRU5U5S-n [$R&"U S S S 9nUR)U [RRU 55 S S S 5 g g US :Xa[RRXE5n URHn U R+U 5 M U(a[RR[RRU5U5S-n [$R&"U S S S 9nUR)U [RRU 55 S S S 5 g g g !,(df  g =f!,(df  g =f)zExport the list of tables to specified file format. Parameters ---------- path : str Output filepath. f : str File format. Can be csv, excel, html, json, markdown or sqlite. compress : bool Whether or not to add files to a ZIP archive. )rBrrr)csvhtmljsonmarkdownrexcelrr)rrrTrNsqliterF)rArBrrCsplitexttempfilemkdtemprrr4rrrrrrrrrrrr)r$rBrcompressrrCrrrrrrrrrs r&exportTableList.export~s''//$'77##D)GG$$X.  &&(G#R 3 3    +q +F +"",V, '\ww||G6H^^H-F$UZZL }E !!&!@& LLN'',,rwwt'+B+Br)r)F)rz list[Table]rrrr)0rD __future__rrrYrArsysr%rcollections.abcrroperatorrtypingrrEnumpyrpandasr version_infor r typing_extensionsbackendsr image_processingrutilsrrrr[rrVERTICAL_ALIGNMENTSALL_ALIGNMENTSrrHrdrrrrrrrFr)r&r:s<"  $$ w  ,(,+.*&41&)<<0(0(f8%}8%v/B/Bd\ \ ~S?S?lmm`Vr ) QBQBr)