Date: prev next · Thread: first prev next last
2012 Archives by date, by thread · List index


Hi Robinson,

2012/7/6 Robinson Tryon <bishop.robinson@gmail.com>

On Fri, Jul 6, 2012 at 5:51 AM, Flavio Moringa <
flavio.moringa@caixamagica.pt> wrote:


I know that you can convert documents through the command line, using
LibreOffice headless mode, and that can be something that's useful for
scripting automatic tests... although I know that sometimes the main
problems are "visual" and it's difficult to automatically detect the
problems...


I think that we still need human eyes for the final comparison, however
the rest of the system could be automated a bit more -- e.g. we could put
sample docs in subdirectories named by bug# and add screenshots of the docs
as rendered in MS-Office; add in a script to have LO iterate over the
subdirectories and spit out screenshots of how it renders the original
files, and a little HTML GUI so that you can tab-through 2-ups of the
original rendering vs. LO's rendering, and you've got a decent tool for
testing improvements/regressions.


That's a good ideia, at least it would facilitate the testing, which is
always very helpful... From my initial investigation in document
conversion, the "visual" aspect is alwayst the difficult one because not
all thins are well translated to the XML...



Is there any kind of repository for documents that are candidates for
conversion testing? I mean documents which are known to have conversion
problems, and that are used to test improvements to the filters?


I usually just search bugzilla for "conversion" or "formatting" :-) Even
documents attached to old bugs can be helpful, as they can serve as
regression tests.


I've just replied to Michael Meeks that I'll do just that... try to compile
a list of available documents in bugzilla that have conversion problems,
and test them on the latest beta... And see which problems still exist...


I would like very much to become more involved in improving the conversion
filters, since it seems to be a major problem in LibreOffice adoption, and
everything that can be done to help in that area would certainly boost
LibreOffice adoption specially in the enterprise world.


Yes, fidelity of document rendering is definitely one of the biggest
hurdles I've faced when encouraging people to try LO. Any improvements on
that front will be greatly appreciated!

--R


Thanks

-- 
*Flávio Moringa*
Project Leader



Caixa Mágica Software
Energia Open Source
Rua Soeiro Pereira Gomes, Lote 1 - 4.º B,
Edifício Espanha, 1600-196 Lisboa - Portugal
Tel.: +351 217 921 260 Fax: +351 217 921 261
http://www.caixamagica.pt
https://twitter.com/flaviomoringa
https://www.facebook.com/flaviomoringa<https://www.facebook.com/flavio.moringa>
http://pt.linkedin.com/in/flaviomoringa
http://people.caixamagica.pt/flaviomoringa

Context


Privacy Policy | Impressum (Legal Info) | Copyright information: Unless otherwise specified, all text and images on this website are licensed under the Creative Commons Attribution-Share Alike 3.0 License. This does not include the source code of LibreOffice, which is licensed under the Mozilla Public License (MPLv2). "LibreOffice" and "The Document Foundation" are registered trademarks of their corresponding registered owners or are in actual use as trademarks in one or more countries. Their respective logos and icons are also subject to international copyright laws. Use thereof is explained in our trademark policy.