i-
S r SSKrSSKrSSKJr SSKJr SSKJr S rS r S r
S r S
r SS jr
S S S
S
SSSSS S. rSSS.r " S S\5 rg)zZ
This code is refer from: https://github.com/weizwx/html2docx/blob/master/htmldocx/h2d.py
N)Document)
BeautifulSoup)
HTMLParserc H / SQnU R SR U5 SS9$ )Nz
table > trztable > thead > trztable > tbody > trztable > tfoot > trz, F recursive)selectjoin)
table_souptable_row_selectorss l/var/www/html/banglarbhumi/venv/lib/python3.13/site-packages/paddleocr/ppstructure/recovery/table_process.pyget_table_rowsr s, TYY': ;u MM c 8 U ( a U R SS/SS9$ / $ )NthtdFr )find_all)rows r get_table_columnsr $ s :=3<<t < 6E2Er c [ U 5 nU( a [ US 5 O/ nSnU H- nUR R SS5 nU[ U5 -
nM/ X4$ )Nr colspan )r r attrsgetint)r rowscols col_countcolr s r get_table_dimensionsr! ) s\ * %D *. T!W %2DI))-- 1-S\! ?r c v SR U R Vs/ s H n[ U5 PM sn5 $ s snf )N )r contentsstr)soupis r
get_cell_htmlr( 9 s- 88T]]3]SV]3 443s 6c v U R nUR 5 R U5 S =Ul Ul g N)_element getparentremove_p) paragraphps r delete_paragraphr1 @ s/ AKKM AD1:r c U( a [ R " SSU 5 n U( a [ R " SSU 5 n [ R " SSU 5 n [ R " SSU 5 $ )a Remove white space from a string.
Args:
string(str): The string to remove white space from.
leading(bool, optional): Remove leading new lines when True.
trailing(bool, optional): Remove trailing new lines when False.
Returns:
str: The input string with new line characters removed and white space squashed.
Examples:
Single or multiple new line characters are replaced with space.
>>> remove_whitespace("abc\ndef")
'abc def'
>>> remove_whitespace("abc\n\n\ndef")
'abc def'
New line characters surrounded by white space are replaced with a single space.
>>> remove_whitespace("abc \n \n \n def")
'abc def'
>>> remove_whitespace("abc \n \n \n def")
'abc def'
Leading and trailing new lines are replaced with a single space.
>>> remove_whitespace("\nabc")
' abc'
>>> remove_whitespace(" \n abc")
' abc'
>>> remove_whitespace("abc\n")
'abc '
>>> remove_whitespace("abc \n ")
'abc '
Use ``leading=True`` to remove leading new line characters, including any surrounding
white space:
>>> remove_whitespace("\nabc", leading=True)
'abc'
>>> remove_whitespace(" \n abc", leading=True)
'abc'
Use ``trailing=True`` to remove trailing new line characters, including any surrounding
white space:
>>> remove_whitespace("abc \n ", trailing=True)
'abc'
z
^\s*\n+\s* z
\s*\n+\s*$z\s*\n\s*r# z\s+)resub)stringleadingtrailings r remove_whitespacer9 G sX P
r62
r62 VVKf
-F
66&#v &&r bolditalic underlinestrike superscript subscript) bstrongemr' ussupr5 r Courier)codeprec f ^ \ rS rSrU 4S jrS
S jrS rS rS rS r S r
S
S jr S
r S r
S rU =r$ )
HtmlToDocx c h > [ TU ] 5 SSSSS.U l / SQU l S U l S U l g )NT)fix-htmlimagestablesstylesr )super__init__optionsr
table_styleparagraph_style)self __class__s r rR HtmlToDocx.__init__ s?
$
# r c 2 / / S.U l U( a Xl O[ 5 U l U R S U l U R U l SU l U R S U l U R S U l S U l SU l
S U l SU l g )N)spanlistrM TrN rP Fr )
tagsdocr rS bsdocumentinclude_tablesinclude_imagesinclude_stylesr/ skipskip_taginstances_to_skip)rV r_ s r set_initial_attrs HtmlToDocx.set_initial_attrs s
H zDH,,z*
" "ll8 4 "ll8 4
!" r c H UR U l UR U l g)z1Copy settings from another instance of HtmlToDocxN)rT rU )rV others r copy_settings_from HtmlToDocx.copy_settings_from s ,, $ 4 4 r c / nSnU H< nU( a US- nM UR U5 [ UR S5 5 nM> U$ )z
Returns array containing only the highest level tables
Operates on the assumption that bs4 returns child elements immediately after
the parent element in `find_all`. If this changes in the future, this method will need to be updated
:return:
r r table)appendlenr )rV tables_soup
new_tablesnestrm s r ignore_nested_tables HtmlToDocx.ignore_nested_tables sP
E e $u~~g./D ! r c [ U S5 ( d SU l g U R U R R S5 5 U l SU l g )Nr& Frm r )hasattrr` rs r& r rO table_no)rV s r
get_tablesHtmlToDocx.get_tables sB tV$$"'D // 0B0B70KL
r c U R ( a1 [ ( a& [ US5 U l [ U R 5 nU R ( a U R 5 U R
U5 g )N html.parser)r^ r r& r% r` rx feed)rV htmls r run_processHtmlToDocx.run_process sF 77}}%dM:DItyy>D OO $r c [ U[ R R 5 ( d& [ S[ R R - 5 eUR
S nUR S:X a [ U5 U R U5 U R U5 U R R
( d U R R S5 g g )Nz Second argument needs to be a %sr r3 )
isinstancedocxrm _Cell
ValueError
paragraphstextr1 rf r~ r]
add_paragraph)rV r} cellunwanted_paragraphs r add_html_to_cellHtmlToDocx.add_html_to_cell s $
0 011 ?$**BRBR RS S !__Q / " "b ( / 0 t$ xx"" HH " "2 & #r c U( a XR l g U R ( a U R U R l g g ! [ a n[ SU R S35 UeS nAff = f)NzUnable to apply style .)r/ stylerU KeyErrorr )rV r es r apply_paragraph_style HtmlToDocx.apply_paragraph_style sf V',$%%'+';';$ & V 5d6J6J5K1 MNTU U Vs A ,A
A1A,,A1c [ US5 n[ U5 u pEUR [ U5 U5 nUR S Ul [ UR 5 n[ UR 5 nSn [ U5 GHP u p[ U 5 n Sn
U GH1 n[ UR R SS5 5 n[ UR R SS5 5 n[ U5 nUR S:X a SU- nX: d X: a Mz UR X5 nUR S :w a( U
S-
n
UR X5 nUR S :w a M( UR U U- S-
X- S-
5 nUU:w a UR# U5 [% 5 nUR' U 5 UR) U=( d S
U5 X-
n
GM4 U S-
n GMS g ) z
To handle nested tables, we will parse tables manually as follows:
Get table soup
Create docx table
Iterate over soup and fill docx table with new instances of this parser
Tell HTMLParser to ignore any tags until the corresponding closing table tag
r{ z
Table Gridr r r rowspanr z %sr3 r# N)r r! add_tablero rP r r columns enumerater r r r r( namer r mergerJ rj r )rV r} r] r r cols_lenrm num_rowsnum_colscell_rowindexr r cell_colr r r cell_html docx_cell
cell_to_merge child_parsers r handle_tableHtmlToDocx.handle_table s #4 7
-j9
c$i2jj . uzz?u}}%#D/JE$S)DH ciimmIq 9: ciimmIq 9: )# . 88t# +i 7I'8+? !JJx : nn* MH %
8 >I nn* !&
w&*H,>,B!
- OOM2 )| //5 --i.>3 J#7 8
MH? *r c U R ( a g SU R ; a
[ USS5 nU R ( d/ U R R 5 U l U R
5 U R R S5 nU( a U R US U5 g U R R U5 U l
U R S nU H0 nSU; d M U R US 5 nU R U5 M2 U R Hf nU[ ; a* [ U n[ U R R US5 U[ ; d MC [ U nXR R l Mh g )NrH TahrefrZ r )rc r\ r9 r/ r] r r r handle_linkadd_runrunparse_dict_stringadd_styles_to_run font_stylessetattrfont
font_namesr ) rV datalinkspansrZ r tag
font_style font_names r handle_dataHtmlToDocx.handle_data! s( 99 !$T46D~~ !XX 3 3 5DN & & (
yy}}S! T&\4 0 ~~--d3DHIIf%E d? 2 24= AE**51 yy+%!,S!1JDHHMM:t<*$ *3I)2HHMM& !r )r^ r] r_ ra rb r` re rS r/ rU r rc rd r& rw r
rT rO r\ r* )__name__
__module__ __qualname____firstlineno__rR rf rj rs rx r~ r r r r __static_attributes__
__classcell__)rW s @r rJ rJ s< $"#&5
" 'V0d$3 $3r rJ )FF)__doc__r4 r r bs4r html.parserr r r r! r( r1 r9 r r rJ r r r s}
"NF
5 2'l
w3 w3r