LibreOffice at FOSDEM
|Dates||February 2-3, 2013|
|Location||ULB Campus Solbosch,
FOSDEM is the premier open source developers conference, and the most significant developers conference for LibreOffice after the annual LibreOffice Conference.
The LibreOffice project has been granted a booth table on both Saturday and Sunday. Please see here for where we are.
Details for the stand planning are on this page Marketing/Events/Fosdem2013/Stand.
Michael Meeks gave a talk about re-factoring the LibreOffice code-base, you can see hybrid PDF slides for that or watch .
- Come and hear about the growth and success of LibreOffice and how you can get involved in this exciting project at the cutting edge of Free Software. Hear from many of the core developers, work out how best to get your most annoying problems fixed, and find how best to get plugged into the team. Co-ordinate with your co-developers, get caught up with the latest developments all over the project, meet friends you've hacked with on-line, all this and more. If you're just a user and want to go deeper, to help improve things we'll have something for you too.
LibreOffice Devroom Talks
|Title||Recent changes in localization toolchain|
|Abstract||What happened on l10n front recently? A lot. New Pootle version, PO files directly used in the build (no more SDF files), automatic checks for common translation errors build-time, and more. I will present how these developments save time for everyone.|
|Abstract||With UNO service information changed from active to passive, XML-based representation a while ago, UNO type information is the remaining topic that needs some love, clean-up, and improvement. From a curiously verbose UNOIDL syntax, to the awkward binary .rdb format, to the fact that type information is duplicated in different formats for binary UNO and Java UNO, all the way to the still unsolved great challenge of becoming incompatible compatibly, etc. etc. -- you get the idea.|
|Title||How to debug Writer file format issues?|
|Abstract||Did you see a Writer feature that is not imported or exported correctly to ODT/DOC/DOCX/RTF? In most cases the problem can be fixed with little effort. We won't give you fish, but we'll teach you how to fish.|
|Title||MinGW-w64 & Wine: Developing LibreOffice for Windows...without Windows|
|Speaker||Fridrich Strba on the behalf of Jan Holesovsky|
|Abstract||If you are a Linux developer, but need to check your LibreOffice code on Windows from time to time, this is the session for you! Cross-compiling LibreOffice for Windows using MinGW is trivial, and you can even directly run the result on your Linux machine without having to upload the results to a virtual host, or to a Windows machine. You can use your familiar environment during development, and also the build times are shorter than on real Windows.
This How-To session will help you to set up the environment for cross-compilation, and will show few tricks you should learn.
|Title||The Upgrade of the LibreOffice Template Repository|
|Abstract||The LibreOffice template repository is currently created using Plone and an add-on named Plone Software Center. Many contributors to the templates repository find it to complicate to upload and publish a template on the site, because they had to create a project and then a release for every new template version. Only afterwards they could upload their template onto the site. It's currently not possible to translate the project pages (and sub pages) or e.g. the categories of the projects. Thus I started to work on a new add-on for a LibreOffice template center that has a more flat hierarchy and make it possible to translate the projects (and sub pages) as well as the categories of the projects into different languages. The presentation will show the current state of this Upgrade of the LibreOffice templates site.|
|Title||Finding and fixing performance problems in Calc|
|Biography||Libreoffice calc hacker and mathematics student|
|Abstract||This talk will show how to find performance problems in Calc and how to identify the right place to fix it. The main focus of the talk will be around calc but the main ideas can also be used in other parts of Libreoffice. The presentation will show how to use gdb, callgrind, kcachegrind and a bit of understanding of the Calc core makes fixing performance problems easy.|
|Title||General easy hacks for new hackers|
|Abstract||Come and hear about how to get involved in coding on the project with no prior experience. Starting with some easy first hacks, and then how to build up to bigger and more interesting things.|
|Title||Simple introduction to LibreOffice QA|
|Biography||Wonderful Dutch LibreOffice champion.|
|Abstract||Come and hear about how to get involved in QA: keeping the developers honest, and improving the quality of the product through better bug reports, bisecting out regressions, and making friends with both users and developers.|
|Title||Give me my drawing back! - Dragging your Visio, Publisher and CorelDraw files to
|Abstract||Since the Google Summer of Code, in collaboration with re-lab, LibreOffice community started to open to the broader free-software ecosystem several proprietary file-formats. This talk will present the libraries for parsing Visio (libvisio), CorelDraw (libcdr) or MS Publisher (libmspub) documents that are currently used by several free-software projects. The talk will focus on interesting details of reverse-engineering, and explain our method of "incremental" reverse-engineering which allowed us to open the file-formats of *all* CorelDraw versions. The same method is currently used to open the file-formats of all remaining versions of MS Visio. The talk will also present some of the introspection tools we develop and use. Those tools are free-software and their use goes beyond the reverse-engineering work.|
|Title||LibreOffice now accesses my documents through CMIS, how to support new server types by reusing that work?|
|Biography||In a previous life I have been providing consulting service on open source Enterprise Content Management tools, started or helped starting several attempts to integrate them into my favourite office suite. I finally found the ultimate way to do it with the CMIS integration in LibreOffice little more than a year ago. In my current life I am hacking LibreOffice Writer for SUSE.|
|Abstract||LibreOffice easily connects to Enterprise Content Management servers through CMIS protocol using the libcmis library. I'll show how this could be easily extended to support other non-standard backends (like SharePoint native protocol or Google Drive) by implementing CMIS-wrappers for them.|
|Title||Widget Layout, tutorial.|
|Abstract||Widget layout has landed in LibreOffice. By reusing the gtk builder format this means that LibreOffice dialogs can now be edited visually with glade with immediate results, instead of the traditional; blind-edit, compile, run and profuse swearing, development cycle of the legacy .src format. Here's a tutorial in how to convert an existing dialog to .ui format. Tip, tricks and common design patterns to help you get involved.|
|Title||Introducing LibreOffice 4.0|
|Biography||Italo Vignoli is a founder and a member of TDF BoD, in charge of marketing communications. He is one of the main spokespersons, and is presenting TDF and LibreOffice at free software events.|
|Abstract||LibreOffice 4.0 is going to be a major announcement for The Document Foundation. The session will introduce the new version of the software using visuals, intended to help the entire community - starting from developers - to position the program in the right way, and present the new features.|
|Title||Language tags - or, what is BCP 47 and why would we want it|
|Biography||Eike is a StarOffice/OpenOffice.org/LibreOffice hacker since the early days at Star Division, later Sun Microsystems, one other company and now Red Hat, Inc.
He is a Calc spreadsheet core engine formula compiler and interpreter hacker and a number formatter stricken i18n transpositionizer.
|Abstract||During the development cycle of LibreOffice 4.0 I changed much code in preparation to understand BCP 47 language tags. Some critical areas weren't changed yet for 4.0 and work is ongoing for 4.1. In this talk I'll give some overview what I did and why I'm undergoing this effort.|
|Title||Debugging calc filter problems|
|Slides||File:Calc filter branding.pdf|
|Biography||Libreoffice calc hacker|
|Abstract||A short introduction to debugging calc import and export filters. The talk will show where to put breakpoints and how to find the right source code file for different filters/features.|
|Title||Hacking Android remote|
|Abstract||This talk will show that hacking the Libreoffice Android remote code is fun (and, actually, how it works).|
|Biography||Open source hacker and perpetual PhD student.|
|Abstract||Interoperability between office suites is difficult but with some reverse engineering and the use of a debugger, LibreOffice support for commonly used file formats is greatly improving. This talk mainly focuses on OOXML (.docx et al) interoperability and gives some suggestions for those who are interested in contributing to this area.|
|Title||New data structure for Writer core? cancelled due to illness|
|Abstract||One major problem that is hard to fix with the current core data structures of Writer is anything that relates to tracking of changes - undo and change-tracking (redlining) itself. It is because the data structures use naive approach to the document - there is one array of 'nodes' (BigPtrArray), and additional data structures that take care of the changes.
There are editors that use different data structure, so called "piece table", that has the advantage that it contains the change tracking information in the data structure itself, so it is trivial to traverse the changes. But, it is unusable for us because it would mean a complete Writer core rewrite.
But - is there a possibility to upgrade/rewrite BigPtrArray so that it would contain the change tracking information in itself in a git-like fashion? ;-) That would allow us to do the changes incrementally - first upgrade the BigPtrArray without affecting the current functionality, and then do the rest.
|Title||How to ensure that eGovernment forms are compatible with Free Software?|
|Biography||FOSS enthusiast, GNU/Linux hacker, helping governments realize the problems they are creating for free software users.|
|Abstract||Governments on a local and European level work to create eGovernment services that are accessible to all citizens, independent from the operating system and applications they use. This means providing forms that are in an open format and that can be used with FOSS tools. Providing forms and documents in proprietary formats makes it difficult for FOSS users to make use of the full potential of the services that are offered online.
The Free Software Client Reference System is a tool used to help local governments ensure that all the services that are provided to citizens are compatible with FOSS. The goal here is to ensure that every citizen does not need to own a proprietary system to access eGovernment services.