extract

tool to extract text from major document formats (namespace: extract)

supported formats

format	parser	remarks
xls	xls-parser	cells comma delimited
xlsx	opc-parser	cells comma delimited
pdf	pdfium-parser
msg	olecf-parser	subject+body
eml	gmime-parser	subject+body
rtf	rtf-parser
txt	txt-parser
html	tidy-parser
ppt	olecf-parser
pptx	opc-parser
doc	olecf-parser	with control characters
docx	opc-parser	without header/footer
jpg	ocrs-parser	only latin alphabet
gif	ocrs-parser	only latin alphabet
ico	ocrs-parser	only latin alphabet
bmp	ocrs-parser	only latin alphabet
webp	ocrs-parser	only latin alphabet
pnm	ocrs-parser	only latin alphabet
png	ocrs-parser	only latin alphabet

acknowledgements

• libolecf
• tidy-html5
• PDFium
• OPC
• ocrs

usage

instantiate the class passing an extension as parameter.

var $extract : cs.extract.extract
$extract:=cs.extract.extract.new(".docx")

use cs.extract.formats to get the list of supported formats.

$extensions:=cs.extract.formats.new().extensions

there are 2 ways to invoke .getText(); synchronous and asynchronous.

synchronous: pass a single parameter and receive a collection of results in return.

$texts:=$extract.getText(${file: $file})

you can pass a single object or a collection of objects in a single call.

asynchronous: pass a second formula parameter. an empty collection is returned at this point.

the formula should have the following signature:

#DECLARE($worker : 4D.SystemWorker; $params : Object)

var $text : Text
$text:=$worker.response

[!TIP] whatever value you pass in data is returned in context

$extract.getText({file: $file.getContent(); data: $file}; Formula(onResponse))

#DECLARE($worker : 4D.SystemWorker; $params : Object)

var $text : Text
$text:=$worker.response
$file:=$params.context

use this to match input against output.

property	type	description
file	4D.File 4D.Blob Text	input