vljbzddlZddlZddlmZmZmZddlmZmZm Z m Z ddl m Z ddl mZmZddlmZddlmZmZddlmZdd lmZmZmZmZmZmZmZmZm Z m!Z!m"Z"m#Z#m$Z$ejJe&Z'Gd d Z(Gd d Z)GddZ*GddZ+Gdde*Z,Gdde,Z-Gdde-Z.Gdde-Z/Gdde,Z0Gdde*e+Z1Gdde,e+Z2ed e*!Z3Gd"d#e,ee3Z4Gd$d%e4e3Z5Gd&d'e5e3e+Z6e e2e1fZ7Gd(d)e6e7Z8Gd*d+e8Z9Gd,d-e8Z:Gd.d/e6e8Z;Gd0d1e;Z<Gd2d3e;Z=e e;d4fZ>Gd5d4e6e>Z?Gd6d7e?Z@Gd8d9e?ZAGd:d;e4e,ZBGd<d=eBZCGd>d?eBZDy)@N)IterableIteratorSequence)GenericTypeVarUnioncast) PDFColorSpace) PDFTypeError PDFValueError)PDFFont)ColorPDFGraphicState) PDFStream) INF LTComponentTMatrix PathSegmentPlanePointRectapply_matrix_rectbbox2strfsplit get_bound matrix2struniqc&eZdZddeddfdZddZy) IndexAssignerindexreturnNc||_yNr )selfr s E/var/www/html/pp/venv/lib/python3.12/site-packages/pdfminer/layout.py__init__zIndexAssigner.__init__$s  ct|tr'|j|_|xjdz c_yt|tr|D]}|j |yyN) isinstance LTTextBoxr LTTextGrouprun)r%objxs r&r/zIndexAssigner.run'sM c9 % CI JJ!OJ [ )   *r(r)r0LTItemr!N)__name__ __module__ __qualname__intr'r/r(r&rr#sc$r(rcbeZdZdZ ddedededededzded ed dfd Zdd Zd efd Z y)LAParamsaParameters for layout analysis :param line_overlap: If two characters have more overlap than this they are considered to be on the same line. The overlap is specified relative to the minimum height of both characters. :param char_margin: If two characters are closer together than this margin they are considered part of the same line. The margin is specified relative to the width of the character. :param word_margin: If two characters on the same line are further apart than this margin then they are considered to be two separate words, and an intermediate space will be added for readability. The margin is specified relative to the width of the character. :param line_margin: If two lines are are close together they are considered to be part of the same paragraph. The margin is specified relative to the height of a line. :param boxes_flow: Specifies how much a horizontal and vertical position of a text matters when determining the order of text boxes. The value should be within the range of -1.0 (only horizontal position matters) to +1.0 (only vertical position matters). You can also pass `None` to disable advanced layout analysis, and instead return text based on the position of the bottom left corner of the text box. :param detect_vertical: If vertical text should be considered during layout analysis :param all_texts: If layout analysis should be performed on text in figures. line_overlap char_margin line_margin word_margin boxes_flowNdetect_vertical all_textsr!c||_||_||_||_||_||_||_|jyr#)r;r<r=r>r?r@rA _validate)r%r;r<r=r>r?r@rAs r&r'zLAParams.__init__LsD)&&&$." r(c|jZd}t|jttfs t |d|jcxkrdkst |t |yy)Nz@LAParam boxes_flow should be None, or a number between -1 and +1r+)r?r,r7floatr r )r%boxes_flow_err_msgs r&rCzLAParams._validate`sl ?? &R te ="#566-A-#$677.#$677. 'r(c vd|jdd|jdd|jdd|jd S)Nz)r<r=r>rAr%s r&__repr__zLAParams.__repr__jsS%d&6&6s%;<++C01++C01*! - r()?g@rLg?rLFFr!N) r4r5r6__doc__rFboolr'rCstrrKr8r(r&r:r:0s:"   #& %    DL  (8 # r(r:c eZdZdZdeddfdZy)r3z)Interface for things that can be analyzedlaparamsr!Ncy)zPerform the layout analysis.Nr8r%rRs r&analyzezLTItem.analyzevsr()r4r5r6rNr:rUr8r(r&r3r3ss3++T+r(r3c(eZdZdZdefdZdefdZy)LTTextz#Interface for things that have textr!cXd|jjd|jdSN< rI) __class__r4get_textrJs r&rKzLTText.__repr__}s)4>>**+1T]]_,?qAAr(ct)zText contained in this objectNotImplementedErrorrJs r&r]zLTText.get_texts!!r(N)r4r5r6rNrPrKr]r8r(r&rWrWzs-B#B"#"r(rWceZdZdZdeddfdZdefdZdede fdZ dede fd Z dede fd Z dede fd Z deddfd Zde fd Zddde fdZdddefdZdddefdZddde fdZdddefdZdddefdZy) LTComponentzObject with a bounding boxbboxr!NcPtj||j|yr#)r3r'set_bboxr%rcs r&r'zLTComponent.__init__s dr(cbd|jjdt|jdSrY)r\r4rrcrJs r&rKzLTComponent.__repr__s,4>>**+1Xdii-@,ACCr(_ctr#r r%rhs r&__lt__zLTComponent.__lt__r(ctr#rjrks r&__le__zLTComponent.__le__rmr(ctr#rjrks r&__gt__zLTComponent.__gt__rmr(ctr#rjrks r&__ge__zLTComponent.__ge__rmr(c|\}}}}||_||_||_||_||z |_||z |_||_yr#)x0y0x1y1widthheightrc)r%rcrurvrwrxs r&rezLTComponent.set_bboxsIRR"W 2g  r(cB|jdkxs|jdkSNr)ryrzrJs r&is_emptyzLTComponent.is_emptyszzQ2$++"22r(r0ct|tsJtt||j|j kxr|j|j kSr#)r,rbrPtyperurwr%r0s r& is_hoverlapzLTComponent.is_hoverlapD#{+;Sc^;+vv 6TWW%66r(ct|tsJtt||j |ryt t |j|jz t |j|jz Sr| r,rbrPrrminabsrurwrs r& hdistancezLTComponent.hdistancea#{+;Sc^;+   C s477SVV+,c$''CFF2B.CD Dr(ct|tsJtt||j |rLt t |j|jz t |j|jz Syr|rrs r&hoverlapzLTComponent.hoverlapa#{+;Sc^;+   C s477SVV+,c$''CFF2B.CD Dr(ct|tsJtt||j|j kxr|j|j kSr#)r,rbrPrrvrxrs r& is_voverlapzLTComponent.is_voverlaprr(ct|tsJtt||j |ryt t |j|jz t |j|jz Sr| r,rbrPrrrrrvrxrs r& vdistancezLTComponent.vdistancerr(ct|tsJtt||j |rLt t |j|jz t |j|jz Syr|rrs r&voverlapzLTComponent.voverlaprr()r4r5r6rNrr'rPrKobjectrOrlrorqrsrer}rrFrrrrrr8r(r&rbrbs $TdD#D4444Td3$37}77E]EuEMe7}77E]EuEMer(rbceZdZdZ ddedeedededededzd edzd ee dzd e e e fdzd dfd Z d e fdZy)LTCurvezA generic Bezier curve The parameter `original_path` contains the original pathing information from the pdf (e.g. for reconstructing Bezier Curves). `dashing_style` contains the Dashing information if any. N linewidthptsstrokefillevenoddstroking_colornon_stroking_color original_path dashing_styler!c tj|t|||_||_||_||_||_||_||_ ||_ | |_ yr#) rbr'rrrrrrrrrr) r%rrrrrrrrrs r&r'zLTCurve.__init__s[ T9S>2"   ,"4**r(cFdjd|jDS)N,c3:K|]}dj|yw)z {:.3f},{:.3f}N)format).0ps r& z"LTCurve.get_pts..sEq...2Es)joinrrJs r&get_ptszLTCurve.get_ptssxxEDHHEEEr(FFFNNNN)r4r5r6rNrFlistrrOrrtuplerr'rPrr8r(r&rrs'++/266:++%[+ +  +  + +"DL+K(4/+VV^,t3+ +.FFr(rc|eZdZdZ ddedededededed edzd edzd ee dzd e e e fdzd dfdZ y)LTLinezOA single straight line. Could be used for separating text or figures. Nrp0p1rrrrrrrr!c Dtj||||g|||||| | yr#rr') r%rrrrrrrrrrs r&r'zLTLine.__init__s4    H        r(r) r4r5r6rNrFrrOrrrrrr'r8r(r&rrs'++/266:              "DL K(4/ VV^,t3   r(rcxeZdZdZ ddededededededzd edzd ee dzd e e e fdzd dfd Z y)LTRectzMA rectangle. Could be used for framing another pictures or figures. Nrrcrrrrrrrr!c f|\} } } } tj||| | f| | f| | f| | fg||||||| yr#r)r%rrcrrrrrrrrurvrwrxs r&r'zLTRect.__init__sW RR  "XBx"bB8 4        r(r) r4r5r6rNrFrrOrrrrrr'r8r(r&rrs'++/266:           "DL K(4/ VV^,t3   r(rc4eZdZdZdedededdfdZdefdZy) LTImagezKAn image object. Embedded images can be in JPEG, Bitmap or JBIG2. namestreamrcr!Ncztj||||_||_|j d|j df|_|j d|_|j dd|_|j d|_t|jts|jg|_yy)N)WWidth)HHeight)IM ImageMask)BPCBitsPerComponentr+)CS ColorSpace) rbr'rrget_anysrcsize imagemaskbits colorspacer,r)r%rrrcs r&r'zLTImage.__init__:sT4(  ~68WX (;<NN#>B  ..)=>$//40#/DO1r(c d|jjd|jdt|jd|j d SNrZ() r[rI)r\r4rrrcrrJs r&rKzLTImage.__repr__EsF''($))B "#1T\\$4A 7 r() r4r5r6rNrPrrr'rKr8r(r&rr4s4 0S 0) 04 0D 0 # r(rc,eZdZdZdeddfdZdefdZy)LTAnnoaActual letter in the text as a Unicode string. Note that, while a LTChar object has actual boundaries, LTAnno objects does not, as these are "virtual" characters, inserted by a layout analyzer according to the relationship between two characters (e.g. a space). textr!Nc||_yr#_text)r%rs r&r'zLTAnno.__init__Ts  r(c|jSr#rrJs r&r]zLTAnno.get_textW zzr()r4r5r6rNrPr'r]r8r(r&rrLs&ST#r(rcreZdZdZdededededededed eeed zefzd e d e d d fdZ d efdZ d efdZ y )LTCharz.Actual letter in the text as a Unicode string.matrixfontfontsizescalingriser textwidthtextdispNncs graphicstater!c tj|||_||_|j|_| |_| |_||z|z|_|jrPt|tsJ|\} } | |dzn| |zdz} d| z |zdz} | | |z|jz| |z| |zf} n,|j|z}d||z|j||z|zf} |j\}}}}}}||z|zdkDxr||zdk|_ t|j| \}}}}||kr||}}||kr||}}tj|||||f|jr|j|_y|j"|_y)NrLgMbP?ir)rWr'rrfontnamerradv is_verticalr,r get_descentuprightrrbrysizerz)r%rrrrrrrrrrvxvyrcdescentabcd_e_frurvrwrxs r&r'zLTChar.__init__^s     (x''1    h. ..HR#%:C2=53HB)x'%/BCdTXX-sX~rDyID&&(83Gw~txx4(1JKD#{{Aq!R1uw*9q1uz ,T[[$?RR 7BR 7BRTBB#34     DI DIr(c d|jjdt|jdt |j d|j d|jd|jd S)NrZr[ matrix=z font=z adv=z text=rI) r\r4rrcrrrrr]rJs r&rKzLTChar.__repr__sr''((499*=)>? -./MM$%88*MMO&a  ) r(c|jSr#rrJs r&r]zLTChar.get_textrr()r4r5r6rNrr rFrPrr rr'rKr]r8r(r&rr[s8+$+$+$ +$  +$  +$+$+$% e 344+$+$&+$ +$Z # #r(rLTItemT)boundcteZdZdZdeddfdZdeefdZde fdZ deddfd Z d e eddfd Z d eddfd Zy) LTContainerz(Object that can be extended and analyzedrcr!Nc>tj||g|_yr#)rbr'_objsrfs r&r'zLTContainer.__init__sT4($& r(c,t|jSr#)iterrrJs r&__iter__zLTContainer.__iter__sDJJr(c,t|jSr#)lenrrJs r&__len__zLTContainer.__len__s4::r(r0c:|jj|yr#)rappendrs r&addzLTContainer.adds #r(objsc4|D]}|j|yr#)r)r%rr0s r&extendzLTContainer.extends C HHSM r(rRcH|jD]}|j|yr#)rrUr%rRr0s r&rUzLTContainer.analyzes!:: "C KK ! "r()r4r5r6rNrr'rrrr7rrrrr:rUr8r(r&rrss2'T'd' (7+ w48G,""T"r(rc$eZdZddZdeddfdZy)LTExpandableContainerr!Nc`tj|ttt t fyr#)rr'rrJs r&r'zLTExpandableContainer.__init__s#TSD3$sd#;.s. -0JsFr!Nc0t|||_yr#)superr'r>)r%r>r\s r&r'zLTTextLine.__init__s &r(cd|jjdt|jd|j dSrY)r\r4rrcr]rJs r&rKzLTTextLine.__repr__s:4>>**+1Xdii-@,A4==?BUUVWWr(rRc|jD]}|j|tj|t dy)N )rrUrrrrs r&rUzLTTextLine.analyzes3:: "C KK ! "fTl+r(planeratioctr#r_)r%rrs r&find_neighborszLTTextLine.find_neighborss "!r(c`t|xs|jjSr#)rr}r]isspace)r%r\s r&r}zLTTextLine.is_emptys%w!>T]]_%<%<%>>r()r4r5r6rNrFr'rPrKr:rUrrrrrOr} __classcell__r\s@r&rrsx 'E'd'X#X,,T, "\""" l  "?$??r(rceZdZdeddfdZdeddffd Zdeedede e fd Z dd ed ede fd Z dd ed ede fd Z dd ed ede fdZdd ed ede fdZxZS)LTTextLineHorizontalr>r!NcHtj||t|_yr#)rr'r_x1r%r>s r&r'zLTTextLineHorizontal.__init__D+.$r(r0cNt|trt|jrh|jt|j|j z}|j |j|z krtj|td|j|_t|%|yNr[) r,rr>r ryrzr$rurrrrwrr%r0marginr\s r&rzLTTextLineHorizontal.addsq c6 "t'7'7%%CIIszz(BBFxx#&&6/)fSk266  Cr(rrc||jz}|j|j|j|z |j|j |zf}|Dcgc]`}t |trN|j||r;|j||s&|j||s|j||r|bc}Scc}w)aBFinds neighboring LTTextLineHorizontals in the plane. Returns a list of other LTTestLineHorizontals in the plane which are close to self. "Close" can be controlled by ratio. The returned objects will be the same height as self, and also either left-, right-, or centrally-aligned.  tolerance) rzfindrurvrwrxr,r"_is_same_height_as_is_left_aligned_with_is_right_aligned_with_is_centrally_aligned_withr%rrrrr0s r&rz#LTTextLineHorizontal.find_neighborss DKK zz477DGGaK$''A+FG  3 45++C1+=..sa.@223!2D66sa6H    A%B?otherr-cLt|j|jz |kS)zr!NcHtj||t |_yr#)rr'r_y0r%s r&r'zLTTextLineVertical.__init__2r&r(r0cNt|trt|jrh|jt|j|j z}|j |z|jkrtj|td|j|_t|%|yr() r,rr>r ryrzrxrArrrrvrr)s r&rzLTTextLineVertical.add8sq c6 "t'7'7%%CIIszz(BBFvv)fSk266  Cr(rrc||jz}|j|j|z |j|j|z|j f}|Dcgc]`}t |trN|j||r;|j||s&|j||s|j||r|bc}Scc}w)a>Finds neighboring LTTextLineVerticals in the plane. Returns a list of other LTTextLineVerticals in the plane which are close to self. "Close" can be controlled by ratio. The returned objects will be the same width as self, and also either upper-, lower-, or centrally-aligned. r,) ryr.rurvrwrxr,r?_is_same_width_as_is_lower_aligned_with_is_upper_aligned_withr2r3s r&rz!LTTextLineVertical.find_neighbors@s DJJ zz477Q;1dggFG  3 23**3!*<//q/A223!2D66sa6H    r4r5r-cLt|j|jz |kS)z8Whether the lower edge of `other` is within `tolerance`.)rrvr7s r&rEz)LTTextLineVertical._is_lower_aligned_with\r8r(cLt|j|jz |kS)z8Whether the upper edge of `other` is within `tolerance`.)rrxr7s r&rFz)LTTextLineVertical._is_upper_aligned_with`r8r(ct|j|jzdz |j|jzdz z |kS)z=Whether the vertical center of `other` is within `tolerance`.r;)rrvrxr7s r&r2z-LTTextLineVertical._is_centrally_aligned_withdr<r(cLt|j|jz |kSr#)rryr7s r&rDz$LTTextLineVertical._is_same_width_asls5;;+, 99r(r2)r4r5r6rFr'rbrrrrrrrOrErFr2rDrr s@r&r?r?1sEd {t \"   j   84K4E4RV44K4E4RV4UUU  U:{:u::r(r?c0eZdZdZddZdefdZdefdZy)r-zRepresents a group of text chunks in a rectangular area. Note that this box is created by geometric analysis and does not necessarily represents a logical boundary of the text. It contains a list of LTTextLine objects. r!Nc<tj|d|_y)NrE)r r'r rJs r&r'zLTTextBox.__init__xs  & r(c d|jjd|jdt|jd|j d Sr)r\r4r rrcr]rJs r&rKzLTTextBox.__repr__|sH''($**R "#1T]]_$7q : r(ctr#r_rJs r&get_writing_modezLTTextBox.get_writing_modes!!r(rM)r4r5r6rNr'rPrKrOr8r(r&r-r-ps$ # "#"r(r-c4eZdZdeddffd ZdefdZxZS)LTTextBoxHorizontalrRr!Nc^t|||jjdy)Nc|j Sr#)rxr0s r&z-LTTextBoxHorizontal.analyze.. r(keyrrUrsortr%rRr\s r&rUzLTTextBoxHorizontal.analyze" ! /0r(cy)Nzlr-tbr8rJs r&rOz$LTTextBoxHorizontal.get_writing_moder(r4r5r6r:rUrPrOrr s@r&rQrQ"11T1#r(rQc4eZdZdeddffd ZdefdZxZS)LTTextBoxVerticalrRr!Nc^t|||jjdy)Nc|j Sr#)rwrTs r&rUz+LTTextBoxVertical.analyze..rVr(rWrYr[s r&rUzLTTextBoxVertical.analyzer\r(cy)Nztb-rlr8rJs r&rOz"LTTextBoxVertical.get_writing_moder^r(r_r s@r&rbrbr`r(rbr.c.eZdZdeeddffd ZxZS)r.rr!NcDt||j|yr#)rr'r)r%rr\s r&r'zLTTextGroup.__init__s  Dr()r4r5r6rTextGroupElementr'rr s@r&r.r.s!X&67Dr(c(eZdZdeddffd ZxZS)LTTextGroupLRTBrRr!Nct|||jJ|j|jj fdy)Nchdz |jzdz|j|jzzz Sr*)rurvrxr0r?s r&rUz)LTTextGroupLRTB.analyze..s1Q^svv5:~#&&366/23r(rWrrUr?rrZr%rRr?r\s @r&rUzLTTextGroupLRTB.analyzesF !""...((  3  r(r4r5r6r:rUrr s@r&rjrj  T  r(rjc(eZdZdeddffd ZxZS)LTTextGroupTBRLrRr!Nct|||jJ|j|jj fdy)Ncjdz |j|jzzdz |jzz Sr*)rurwrxrms r&rUz)LTTextGroupTBRL.analyze..s4a*n-#&&A:~'(r(rWrnros @r&rUzLTTextGroupTBRL.analyzesF !""...((  (  r(rpr s@r&rsrsrqr(rsceZdZdeddfdZdedeedee fdZ dedee dee fd Z ded e e deefd Zdeddfd Zy) LTLayoutContainerrcr!Nc>tj||d|_yr#)rr'groupsrfs r&r'zLTLayoutContainer.__init__sT4(04 r(rRrc#Kd}d}|D]}||j|xrt|j|j|jz|j |kxr?|j |t |j|j|jzk}|jxr|j|xrt|j|j|jz|j|kxr?|j|t |j|j|jzk}|rt|ts|r"t|tr|j!|n||d}n|r:|s8t|j"}|j!||j!|nh|r:|s8t|j"}|j!||j!|n,t|j"}|j!||d}|}"|*t|j"}|J|j!||ywr#)rrrzr;rrr ryr<r@rrrr,r"r?rr>)r%rRrobj0lineobj1halignvaligns r& group_objectszLTLayoutContainer.group_objectss" "&? D$$T*IDKK58M8MMmmD)*It,$**djj1H4H4HHI .,,K((.KDJJ 3h6K6KKmmD)*Kt,$++t{{3h6J6JJK z$0DEz$0BCHHTN%JDF-h.B.BCDHHTNHHTNF/0D0DEDHHTNHHTN/0D0DEDHHTNJDD? @ <'(<(<=D# ## HHTN sIIlinesc#DKt|j}|j|i}|D]}|j||j}|g}|D]8}|j |||vs|j|j |:t|tr t} n t} t|D]} | j| | || <t} |D]7}||vr||} | | vr| j| | jr4| 9yw)z$Group neighboring lines to textboxesN)rrcrrr=rpopr,r"rQrbrrsetr}) r%rRrrboxesr| neighborsmembersr}boxr0dones r&group_textlinesz!LTLayoutContainer.group_textlines s $)#3 U-/ !D++E83G3GHIfG! 4t$5=NN599T?3 4$ 45!4!6')G} !  c  ! !u D5 +Cd{ HHSM<<>  sA'D *B.D D rc tttf}t|jdt dt dt fd}d|d|dt|ffd }g}tt|D]Y}||}t|dzt|D]7} || } |jd||| t|t| || f9[tj|j|t} t|dkDr%tj|\} } }}}}|| vr|| vr| s&|||rtj |d | ||||f[t#|t$t&fst#|t$t&frt'||g}n t)||g}j+|j+|| j-||gD]7}tj |d|||t|t|||f9j/|t|dkDr%Dcgc]}t1t|c}Scc}w) axGroup textboxes hierarchically. Get pair-wise distances, via dist func defined below, and then merge from the closest textbox pair. Once obj1 and obj2 are merged / grouped, the resulting group is considered as a new object, and its distances to other objects & groups are added to the process queue. For performance reason, pair-wise distances and object pair info are maintained in a heap of (idx, dist, id(obj1), id(obj2), obj1, obj2) tuples. It ensures quick access to the smallest element. Note that since comparison operators, e.g., __lt__, are disabled for LTComponent, id(obj) has to appear before obj in element tuples. :param laparams: LAParams object. :param boxes: All textbox objects to be grouped. :return: a list that has only one element, the final top level group. r}obj2r!ct|j|j}t|j|j}t|j|j}t|j |j }||z ||z z|j |jzz |j |jzz S)aA distance function between two TextBoxes. Consider the bounding rectangle for obj1 and obj2. Return its area less the areas of obj1 and obj2, shown as 'www' below. This value may be negative. +------+..........+ (x1, y1) | obj1 |wwwwwwwwww: +------+www+------+ :wwwwwwwwww| obj2 | (x0, y0) +..........+------+ )rrurvr rwrxryrz)r}rrurvrwrxs r&distz/LTLayoutContainer.group_textboxes..distGsTWWdgg&BTWWdgg&BTWWdgg&BTWWdgg&BbR"W%**t{{*+**t{{*+ r(cft|j|j}t|j|j}t|j|j}t|j |j }t j||||f}|j||fS)z8Check if there's any other object between obj1 and obj2.) rrurvr rwrxrr. difference)r}rrurvrwrxrrs r&isanyz0LTLayoutContainer.group_textboxes..isany]sTWWdgg&BTWWdgg&BTWWdgg&BTWWdgg&Buzz2r2r"234D??D$<0 0r(r+FrT)rr-r.rrcrbrFrrangerridheapqheapifyrheappopheappushr,rbrsrjremoveupdaterr )r%rRrElementTrrdistsibox1jbox2r skip_isanyrid1id2r}rgroupr5grs @r&group_textboxesz!LTLayoutContainer.group_textboxes.s5,K/0!&tyy!1 { + % , 1 1 1S] 1IKs5z" XA8D1q5#e*- XQx eT$%5r$xD4QUVW X X  e Uu%j1n49MM%4H 1ZCdD4co!eD$&7NN54CdD*IJd%6$HIZ&8N*9$)FE+T4L9E T" T" S#J'"ENNUE 2BuIr%y%QVW  % /%j1n2/44[!$444s-Ictd|\}}|D]}|j||syt|j||}td|\}}|D]}|j|t|j ||}|j M|D]}|j|dt dttttffd} |j| ng|j|||_ t} |jD]$} | j|| j| &|jdttt ||zttt |z|_y)Nc"t|tSr#)r,rrTs r&rUz+LTLayoutContainer.analyze..s :c63Jr(c"|jSr#)r}rTs r&rUz+LTLayoutContainer.analyze..s #,,.r(rr!ct|trd|j |j fSd|j |jfS)Nrr+)r,rbrwrvrurs r&getkeyz)LTLayoutContainer.analyze..getkeys=c#45w00w//r(rWc|jSr#r$rs r&rUz+LTLayoutContainer.analyze..s 399r()rrUrrrr?r-rr7rFrZrryrr/r rbr) r%rRtextobjs otherobjsr0 textlinesempties textboxestextboxrassignerrs r&rUzLTLayoutContainer.analyzesq!''JD Q9 "C KK ! " ++Hh?@ %&@)L) "C KK ! "--h BC    &$ *) * 0I 0%UE0A*B 0 NNvN &..xCDK$H $ h' U# $ NN4N 5 k"I . 4 $g. / r()r4r5r6rr'r:rrbrrrr-rrrr.rrUr8r(r&rwrws5T5d5 KK{#K *  KZ   #  )  D[5[5 "[5 k  [5z# # T# r(rwcDeZdZdZdedededdfdZdefdZd e ddfd Z y) LTFigurezRepresents an area used by PDF Form objects. PDF Forms can be used to present figures or pictures by embedding yet another PDF document within a page. Note that LTFigure objects can appear recursively. rrcrr!Nc||_||_|\}}}}||||z||zf}t||}tj ||yr#)rrrrwr') r%rrcrr1ywhrects r&r'zLTFigure.__init__sO   Aq!1a!eQU# .""4.r(c d|jjd|jdt|jdt |j d S)NrZrrrrI)r\r4rrrcrrrJs r&rKzLTFigure.__repr__sO''($))B "#$ -.a 1 r(rRcJ|jsytj||yr#)rArwrUrTs r&rUzLTFigure.analyzes!! !!$1r() r4r5r6rNrPrrr'rKr:rUr8r(r&rrsG/S//f// # 22T2r(rc 6eZdZdZd dedededdfdZdefdZ y) LTPagezRepresents an entire page. Like any other LTLayoutContainer, an LTPage can be iterated to obtain child objects like LTTextBox, LTFigure, LTImage, LTRect, LTCurve and LTLine. pageidrcrotater!NcLtj||||_||_yr#)rwr'rr)r%rrcrs r&r'zLTPage.__init__s ""4.  r(c d|jjd|jdt|jd|j d S)NrZrrz rotate=rI)r\r4rrrcrrJs r&rKzLTPage.__repr__sI''($++ "#$kk_A ' r(r2) r4r5r6rNr7rrFr'rPrKr8r(r&rrs4 s$d  # r(r)Erloggingcollections.abcrrrtypingrrrr pdfminer.pdfcolorr pdfminer.pdfexceptionsr r pdfminer.pdffontr pdfminer.pdfinterprrpdfminer.pdftypesrpdfminer.utilsrrrrrrrrrrrrr getLoggerr4loggerrr:r3rWrbrrrrrrrrrr TextLineElementrr"r?r-rQrbrhr.rjrsrwrrr8r(r&rs 88,>$5'   8 $  @ @ F++""F&FR!Fk!FH W B W B k 0 VV :[&:z )6 *"+ww/"2 K0 $ +G4f '?1?<<<:<<~<:<:~" +",) M12/"23  k    k  u K0u p2 2:  r(