<html> <head> <base href="https://bugs.documentfoundation.org/"> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - Rewrite old Pocket Word (PWI) file format import filter." href="https://bugs.documentfoundation.org/show_bug.cgi?id=77278#c13">Comment # 13</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - Rewrite old Pocket Word (PWI) file format import filter." href="https://bugs.documentfoundation.org/show_bug.cgi?id=77278">bug 77278</a> from <a class="email" href="mailto:alonso@loria.fr" title="osnola <alonso@loria.fr>"> osnola</a> <pre>The old filter is quite basic: it reads the font names, then uses some heuristics to retrieve paragraphs of text, retrieving: - the main properties of the characters, with the notable exception of the superscript and the subscript… - the properties of the following paragraph: first indent, left/right margin, left/center/.. alignment, a flag to know if it is a bulleted list, I have rewritten a « more robust » version of this code in libwps, but clearly, there are many things that are not recovered (as I can not guess what there means). If you want to try it, I have updated the libwps version compiled with emscripten: <a href="http://libwps.sourceforge.net/convertWPS.html">http://libwps.sourceforge.net/convertWPS.html</a> . To improve this filter, it would be useful to have some Pocket Word files (and their pdfs equivalent) that : - [character properties] use exponents and subscripts, - [paragraph properties] have paragraphs with single/double/double line spacing/..., with a certain spacing before/after the paragraph, lines with fixed height, different types of listings - [general] contains header(s), footer(s), footnote(s), endnotes, comment(s), image(s), table(s)… and simple documents with different page sizes, different margins, some metadata... - …</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the assignee for the bug.</li> </ul> </body> </html>