we are now testing our prototype using the aspose pdf-lib with realistic data from our customer.
Unfortunately the HTML-blocks that have to rendered in the generated pdf document may have an ugly format as in the example "ugly-html-renamed.zip". Since the Aspose lib seem to do a strong validation of the html, the conversion can crash on some unallowed formats tags or characters.
Now: is there are way to ignore format instruction or character that can't be correctly interpreted or rendered during the conversion, in order that they wouldn't crash, or to check and clean up the block before conversion ?
The problem is we don't have any influence on the quality of the rawdata containing the html-blocks since they can't be copied from a third party tool as word into our database.
so the html tags should only be render if possible
The example shows that the html can contain any strange formatting including inline css
Any idea? Thank for a hint