kh-HdZddlmZmZddlmZddlmZddlmZddl m Z m Z m Z dd l mZdd l mZm Z dd lmZdd lmZGd deZy)aIText block objects based on PDF raw dict extracted with ``PyMuPDF``. Data structure based on this `link `_:: { # raw dict # -------------------------------- 'type': 0, 'bbox': (x0,y0,x1,y1), 'lines': [ lines ] # introduced dict # -------------------------------- 'before_space': bs, 'after_space': as, 'line_space': ls, 'alignment': 0, 'left_space': 10.0, 'right_space': 0.0, 'tab_stops': [15.4, 35.0] } )PtInches)WD_ALIGN_PARAGRAPH)Lines) ImageSpan)RectType TextAlignment lower_round)Block)rgb_component_from_namer ) constants)docxc eZdZdZddeffd ZedZedZedZ edZ edZ ed Z fd Z d Zfd Zd Zdedededededef dZdZdZdZdededededef dZxZS) TextBlockz Text block.rawc|xsi}d|vr|jdt| |t|j |j dg|_|jy)Nbbox)parentlines)popsuper__init__rrestoregetrset_text_block)selfr __class__s S/var/www/teggl/fontify/venv/lib/python3.12/site-packages/pdf2docx/text/TextBlock.pyrzTextBlock.__init__)s^iR S=#''&/ $'//0DE  ct|jDcgc]}|j}}dj|Scc}w)zJText content in block. Note image is counted as a placeholder ````.)rtextjoinrline lines_texts r r$zTextBlock.text7s3-1JJ7Ddii7 7wwz""85ct|jDcgc]}|j}}dj|Scc}w)z5Raw text content in block without considering images.r#)rraw_textr%r&s r r+zTextBlock.raw_text=s315 ;dmm; ;wwz""z-TextBlock.white_space_only..Fs@T4((@)allrrs r r/zTextBlock.white_space_onlyCs@TZZ@@@r!c.|jjS)zsAll lines contained in text block must have same text direction. Otherwise, set normal direction. )rtext_directionr4s r r6zTextBlock.text_directionHs zz(((r!c|jrdnd|jj}t|}|dk(ry|jdz|jz }fd}t t ||}||z |dz z S)z4Average distance between adjacent two physical rows.rrNrc,tfd|DS)Nc3tK|]/}t|jdz|jz 1yw)rN)absr)r0r'idxs r r1z>TextBlock.average_row_gap....[s1*`TX3tyyQ/? #/N+O*`s58max)rowr;s r z+TextBlock.average_row_gap..[ss*`\_*`'`r!)is_horizontal_textrgroup_by_physical_rowslenrsummap)rrowsnum block_heightf_max_row_heightsum_row_heightr;s @r average_row_gapzTextBlock.average_row_gapOs**azz002$i 6$yyQ' #6 `S!1489^+A66r!cHt|jjS)zCount of physical rows.)rBrrAr4s r row_countzTextBlock.row_count`s4::44677r!czt|}|jd|jji|S)Nr)rstoreupdater)rresrs r rNzTextBlock.storefs8gmo TZZ%%'   r!ct|tttfr#|D]}|jj |y|jj |y)zAdd line or lines to TextBlock.N) isinstancerlisttuplerappend)r line_or_linesr's r addz TextBlock.addnsH meT5%9 :% ( !!$' ( JJ  m ,r!ctd}t| ||d|jD]O}td}|j|||jD] }td}|j||"Qy) ztPlot block/line/span area for debug purpose. Args: page (fitz.Page): pdf page. bluez [3.0 3.0] 0)strokedashesred)rZr#)colorN)rrplotrspans)rpagerYr'r\spancrs r r^zTextBlock.plotws'v.  T$} =JJ )D)%0C IId3I '  )+B/ $a ( )  )r!cd}|D]r}|jtjs |jr/|jj |jsU|j j|sqd}t|S)zParse text format with style represented by rectangles. Args: shapes (Shapes): Shapes representing potential styles applied on blocks. FT) equal_to_typer HYPERLINK is_determinedr intersectsrparse_text_format)rshapesflagshapes r rhzTextBlock.parse_text_formatsu  E&&x'9'9:u?R?RT\99'' 3Xzz++E2  r!line_separate_thresholdline_break_width_ratioline_break_free_space_ratiolines_left_aligned_thresholdlines_right_aligned_thresholdlines_center_aligned_thresholdc\|jrdnd\}} } |j||| | f|||||_|jtjk(r0tj |_|j j||j} | dk(r%|jtj k(rd|_ nZ| dk(r%|jtjk(rd|_ n0| dk(r+|jtjk(rd|_ d|_ |j j|||y)a Set horizontal spacing based on lines layout and page bbox. * The general spacing is determined by paragraph alignment and indentation. * The detailed spacing of block lines is determined by tab stops. Multiple alignment modes may exist in block (due to improper organized lines from ``PyMuPDF``), e.g. some lines align left, and others right. In this case, **LEFT** alignment is set, and use ``TAB`` to position each line. )rrg?)rgrrN)r@_parse_alignment alignmentr NONELEFTrparse_tab_stoprL right_spaceRIGHT left_spaceCENTERparse_line_break) rrrlrmrnrorprqidx0idx1frLs r parse_horizontal_spacingz"TextBlock.parse_horizontal_spacings$(,'>'> L dA..t 4O # ( ) * , >>]// /*//DN JJ % %&= > NN a.sOT:dI3NDOsNrrrc:td|jDS)Nc34K|]}|jywr.) line_heightrs r r1zJTextBlock.parse_relative_line_spacing....s.WDt/?/?.Wr2)r=r_)r's r r?z7TextBlock.parse_relative_line_spacing..s3.WDJJ.W+Wr!c,tfd|DS)Nc3.K|] }|ywr.)r0r'fun_max_line_heights r r1zJTextBlock.parse_relative_line_spacing....s,W4-@-F,Wr<)r>rs r r?z7TextBlock.parse_relative_line_spacing..s,WSV,W)Wr!c3.K|] }|ywr.r)r0r>fun_max_row_heights r r1z8TextBlock.parse_relative_line_spacing..sF#05Fr) rrSr_rDEFAULT_LINE_SPACING line_spacer@rrArCrBr=) rr'r;rGrEstandard_heightrrrs @@r parse_relative_line_spacingz%TextBlock.parse_relative_line_spacingsJJ DOTZZOO"+"@"@  **ayyQ' #6 zz002WWFFF!/1  t9Q;SY5S5S%T $r!c|jrdnd}|jdj}||dz||z }|j|dz|j|z }|j}|dkDr ||z |dz z }n|}||_|xj ||z z c_|j dkr*|xj|j |z z c_d|_yy)a0Calculate exact line spacing, e.g. `spacing = Pt(12)`. The layout of pdf text block: line-space-line-space-line, excepting space before first line, i.e. space-line-space-line, when creating paragraph in docx. So, an average line height is ``space+line``. Then, the height of first line can be adjusted by updating paragraph before-spacing. .. note:: Compared with the relative spacing mode, it has a more precise layout, but less flexible editing ability, especially changing the font size. rrrN)r@rrrLr before_space)rr;rfirst_line_heightrGcountrs r parse_exact_line_spacingz"TextBlock.parse_exact_line_spacings**azz!}!! QK$s)3yyQ' #6  19&'8857CJ%J$ .;;   q OOt0058 8O #D  !r!c tj|}tt|jdd}tt|j dd}t ||_t ||_|jdk(r%t t|jd|_ nt|jd|_ |j}|jdkr||jz}t ||_t |j|_t |j|_|j$t&j(k(rt*j(|_|j,D]3}|j,j/t |j|z5t1|jt2j4z d}t7||_n |j$t&j8k(rCt*j8|_t1|t2j4z d}t7||_n|j$t&j:k(rzt*j:|_t1|t2j4z d}t7||_t1|jt2j4z d}t7||_nt*j<|_|j>D]}|jA||S)aCreate paragraph for a text block. Refer to ``python-docx`` doc for details on text format: * https://python-docx.readthedocs.io/en/latest/user/text.html * https://python-docx.readthedocs.io/en/latest/api/enum/WdAlignParagraph.html#wdparagraphalignment Args: p (Paragraph): ``python-docx`` paragraph instance. .. note:: The left position of paragraph is set by paragraph indent, rather than ``TAB`` stop. rrrr)!rreset_paragraph_formatr=roundr after_spacer space_before space_afterline_space_typer line_spacingr{first_line_space left_indentry right_indentfirst_line_indentrur rwr tab_stops add_tab_stopr rITPrrzr|JUSTIFYr make_docx) rppfbefore_spacing after_spacingr{posdr's r rzTextBlock.make_docx sH ( ( + U4#4#4a8#>E$"2"2A6< ^,M*    " t!:;BO#DOOQ7BOoo   " $// /JZ.d../!$"7"78 >>=-- --22BL~~ E ))"T__s-B*CD ED,,Y]]:A>A$QiBO ^^]00 0-33BLJy}}4a8A#AYBN ^^]11 1-44BLJy}}4a8A#AYBND,,Y]]:A>A$QiBO.55BL JJ1Dq 11r!text_direction_paramch|\}}} t|j|||z | zdt|||j|z | zd} t| z dz dtdt| d} t||||z |_| |_|j j} | D]|} t| dk(rtdt| D cgc]0} | | j|| | dz j|z | z|k\2}} t|sltjcSfd}t| dk(r|S| Dcgc]}|dj|}}| Dcgc]}|dj|}}t||Dcgc] \}}||zdz }}}t| dk\r |dd|dd}}tt|t|z |k}tt|t|z |k}tt|t|z k}|r(|r&t| dk\rtjn|}ng|rtj }nT|rtj"}nA|r/t| dk\rtj$ntj"}ntj}|tj"k(s|tjk(r0| ddj|| ddj|z |_|Scc} wcc}wcc}wcc}}w) aDetect text alignment mode based on layout of internal lines. It can't decide when only one line, in such case, the alignment mode is determined by externally check. Args: text_direction_param (tuple): ``(x0_index, x1_index, direction_factor)``, e.g. ``(0, 2, 1)`` for horizontal text, while ``(3, 1, -1)`` for vertical text. rg@rctkrtjSdzkrtjStjS)Ng?)r:r r|rwrz)Wd_centerd_leftrqsr external_alignmentz6TextBlock._parse_alignment..external_alignments@8}==$+++46!$)))$***r!rrsN)rrr=r:r{ryrrArBrangeanyr rvzipminrr|rwrzr)rrrrlrorprqr~rrd_rightrEr>idisrrX0X1x0x1X left_aligned right_alignedcenter_alignedrurrrs ` @@@r rtzTextBlock._parse_alignmentts- dA$))D/$t*4a7;$t*TYYt_4a7;&.#-q1vs#w$ T 4:% &""zz002 *C3x{H!&q#c(!35FKK%c!A#hmmD&991<>UU5C53x$)))  * + t9>"4"6615 5ueAhmmD! 5 504 5ueBinnT" 5 5*-b"+ 6hr2r"uck 6 6 t9>BqrFBqHr2SWSW_-/KK SWSW_-/LL SVCF]+-KK M14TA --CUCWI %,,I %**I /24y!| ++ASASI&**I m(( (I}7L7L,L$(GAJOOD$9DGAJOOD#$LPhc).c,1c27c38 c 49 cr!rN)r docx.sharedrrdocx.enum.textrrimage.ImageSpanr common.sharer r r common.Blockr rcommonrrrrr!r rs82$-'AA Appr!