Standalone document reader using Libreoffice libraries

[Michael Stahl]

On 18.06.2015 22:33, M. Amin Farajian wrote:
> Hi all,
> 
> I am working on a toolkit which does some text analysis on the given
> text documents. This toolkit primarily was supposed to work with XML
> files. But since the input files in the real applications are mostly
> *.doc/*.docx/*.odt/*.ppt/*.pptx/*.odp/*.pdf/ets, I need to write a
> library for reading these file formats and convert their contents into
> the desired XML format.  I was looking for such a library and learned
> that LibreOffice does have such a functionality.

i think your best option is to use LibreOffice's --convert-to command
line option (or equivalent wrappers e.g. unoconv / LibreOfficeKit) to
convert all sorts of formats to ODF, which is relatively easy to handle
with XML tools (Flat-ODF is even easier than the regular zip-based ODF,
but gives you huge base64 encoded images).

if you use Java or DotNet you can use libraries from Apache ODF Toolkit
to easily manipulate ODF files, or use an XML DOM, or XSLT ...

> I searched for the part of the code in LibreOffice which is responsible
> of reading the given files (in different formats), but couldn't find it.
> Could you please point me to this part of the code in the LibreOffice
> project?

it's distributed all around the code base basically.  the only filters
that are easily separable from the LO code are the Document Liberation
ones and those already produce ODF as output.