 So if you want to hear something intelligent, you can already go to have a coffee We actually didn't know until the last moment what to say we submitted the talk only because we wanted to come here, so We called it update from document liberation project, and let's see what comes out from this thing Since there are one or two people who don't know anything about our project We are going to make all the presentation actually about what it is so I'm going to explain short I'm going to explain our philosophy. So and we have a philosophy We are technical guys, but we have philosophy. So we have a philosophy concerning the ownership of documents because we believe that Your creation your pictures Your bad teenage poetry Shouldn't be locked in a file format that you can't read anymore Because the application doesn't work on Your current operating system I know actually and That the content you be that the content that you created is your content is not Like Picasso would have to give his paintings to the oil paint Trader because he was using his oil paint. So We believe that Only by using and open standards we can achieve long-term accessibility to data that That your works are encoded into But we believe that open standards though they already exist their use Will be promoted gradually and for this century. We think that in the transitory period We should implement in free and open source software some tools that allow to read Those data those documents even though they are not open Yeah, because that's the only Approximation of it if you manage to have some tool that is open source You can rebuild it on your system and you can still access your data Thank you So and from from this philosophy comes our mission our mission is that we Inter we look at it on the file formats We try to understand how the data is encoded and we try to find the useful information in them We are not creating a documentation of the file format We are just extracting your data from it and putting it in some meaningful Format so that you can actually access what is useful for you We do it by implementing free and open source software parsers of this file for maths and Also, we don't Generate these files. We want you to be able to access to them But we want you to force you to use something open for after you access your data because first of all it's extremely difficult to make Generator that actually the original application would accept and Yeah, and we know better what to do with ours per time so and The the other mission PowerPoint is that we Try to promote and Increase the ODF ecosystem So we are part of the open document format ecosystem and we have one of the generators that work on this The generate actually open document Now some few Details the software frame framework is is having four points. One is the infrastructure library That has all the APIs And the the stream abstractions that we used to read the documents. It has the output classes that we use to abstract our output and All the other things they they pluck Themself into the framework that is created by this library We have parser libraries. There is a number of them. Normally when you look at them, you will you will understand what they What they import if you don't understand try to read it from the other side And you will understand maybe and we have three generator libraries If you want to break your tongue who wants to try to pronounce this one It's Lieberman grab So the generator libraries we have this main one is sleep ODF gen it's a library that generates the different audio file formats from the from the From the different interfaces for drawing for spreadsheets for world documents text documents Then we have a this this to Lieberman gavel and leap a pub gen they are libraries that are slightly smaller we started them as a proof of concept but they They somehow stuck because we normally don't throw code so Is for generating the hip-hop gen is for generating different Hip-hop file format so you can have your driving drawing on your on your Kindle or whatever It's called in other parts of the world and the Lieberman go up is Is is is an intent to generate a bivoult file format it was just just intellectual masturbation of one of the guys that That collaborates with us and then the fourth part that is actually the real rivers engineering part We have three introspection tools Olatoy that is Where we actually encode our knowledge of the file format so that you can open the file in there and You can see how how the structure looks then call a pattern As you can see from the name is an hexadecimal editor on steroids And then we got contributed something that it's called MSO damper That can dump all the records of MS office files. So we accepted it because we don't refuse code Okay, now Valek Okay, why me Here in in review for the project and that's the part which is actually not Main contribution from me because I'm backward engineer and the main work was done by forward engineers like Friedrich but What we did they mostly worked on robustness of the libraries to make it work even if you supply some I'm Not up to their format Compliant document so it's not supposed to crash and for that we used something which is American fuzzy lop I don't know what's that but I Assume that's up for skater of the files to check that whatever crap you have on the input you still will provide something reasonable at the output and As a result the libraries are much more robust now you can try to submit something which is Potentially could crash it and it's supposed to survive if it's not you know who to question not me I'm only break things and he builds them up All the toy that's where main toy for me to play with and That's how Friedrich already stated the place where we encode our knowledge about file formats It was initially developed it to just play with for Microsoft publisher format and when I figured out that whenever I do any Introspection of anything new I need to encode half of the things again and again So it was all combined on the under the same roof and it's named it only toy for for the reason that it was not Something which you can easy find on the internet and that's the main Point to find a good name once you have a good name everything else is easy so There are a few improvements in all the toy where new formats Support added to it and whatever was here in some cases improved Again half of these or not more is not my job for past year, but by David Tardon. Some of you were I'm lucky to see him a few years ago on the LGM and Lebsic. Unfortunately, he is not that Open to the world to come to more conferences and we hope that we will be able to bring him more to next LGMs But we cannot commit to that So keynote six now mostly complete before we only were able to deflate the compression of this format and that's where we observe Opposite direction if you compare coral and apple coral moves toward where Making format more easy readable and apple make it more difficult to read for whatever reason Another two is pages 5 and number 3. It's not yet completed some progress T602 and all this part is probably not very well known outside of a check here, but it was a company who made a lot of the tools some some years ago and now You would like still to be able to open these files There is a bit for support for Yamaha PPI files. It's a sound files from keyboards made by Yamaha and I also was involved in some Attempts to understand fireworks Multilated PNG files so I added some chopping for PNG and also deflating of a fireworks file format But because interest from from the forward engineer who were asking for that was lost it's in the middle of nowhere if anybody wants that we will continue and Also because I'm a backward engineer I had to fix all a toy to support coral draw x8 Which was released like a month and a half ago by coral and if I were a smart head Who would be able to build stuff like forward engineers when it would just work? so Leap CDR and we are officially announced it here can support Latest released version of a coral draw file formats. You do not need to change anything because it's built by smart people not not me And now I think I can pass it back to you And you have just three minutes that five minutes that was for Valek me. I have still my time. No, it's like sunders and So So what's new in LibVizio? So For those who don't know LibVizio, it's library for parsing and importing Vizio diagrams. So We started to support layers not that we would output layers But we in Vizio layers have properties and the properties overwrite the shape properties So we we can now overwrite color and line of lines and text We can support visibility and printability because you can have things in Invisio that are printable but not visible and you have things that are visible but not printable That's very useful especially if you want to hide something from your relative spouses They are we have we have quite fixes for text support We we had some incorrect handling of text that comes from the model from the stencils, which are the shape Masters that you that you use to to compose your diagram. We support bullets bullet lists We we support now the tap stops so that certain files that were not looking very nice Because they were relying heavily on tabs They look much better And we have better mapping of paragraphs and text puns that results from our understanding what actually Vizio wants to say in in certain records we we had some Requests because in Vizio the page are named pages I named we we also named them when we were importing them page one page two page three But they they can have different other names. So now we found where actually these names are included. So we import them We have some basic incomplete partial and Yeah support for hatch feel Because before when we were importing we were considering hatch feel as a solid feel by a color and we were basically hiding everything that was under the hatch and The pictures were looking nice, but we had a buck and then we actually realized that we are hiding things So so we support the hatch feel and we emulate the rounding of line joints because Vizio Has to do differently and then other kids So the rounding of a line is a property of the line So when you when you join two bezier corpse in in a sharp angle It will find some tongue and and they it will draw nice circle instead of that sharp thing and You have to emulate it because it doesn't exist in free and open source software world after for the publisher library We support tables and embedded fonts using Leap How it's called now No There is there is one guy that contributed this this publisher library who actually created a library for support of the of of Mommy, I'm not no it yeah, dr. Ozheimer stole it from my head. So There is a library, I think it's called EOT or some Leap EOT he created and we use it to to support embedded for Fonts and David he promised that he is going to work on publisher to and The support will come next The keynote important import so We extended version support for for different of the I work file formats We can support now pages one to four numbers one to it's not completely finished because this time we had first time that we had an unsuccessful Google summer of code project and We extended to keynote six this this year for the ebook ebook import library We have improved support for fiction book and Sony broadcast ebook file format For live WPS that was originally a library for World Microsoft works documents, but now it became a little bit like a container for everything that we don't know it should be going so We We have now support for some version of lotus symphony and quattro pro and from the word processing We had there is not much graphic actually in this library. So we support Microsoft right and Microsoft word for those That's related to the batting edge poetry because you can maybe have some on your discs in this file format live wow live wow, it's for old Macintosh file formats and We added support for this and much more, but I I didn't get email from the guy So we just what we remembered. So it's a rack time Clarice draw Apple picked Mac draft and And surely this project is an open project. So anybody can participate We you can either contribute a code if you have a pet file format that you want to import And you want to reuse the framework do it we accept any library. We have two possibilities to host it One is in in in a get it somewhere at LibreOffice website Or we have also a search for projects where we can put it So we have enough place and you can even put it in github or wherever we we just include it in Yeah, then if you are People who during the long summer know a long winter evenings like to look into hexadecimal editor You can start to understand file formats and you can contribute the knowledge of the file formats into Olatoi you need to know some little python but It's not really complicated by it on even me. I understand how it is done. So it should be possible to understand for any dummy Or if even that is complicated to you like maybe it would be for my mother And you have an application that you would like to be supported so you can generate for us documents and Smaller and with some little differences documents. They are better It is for us so that we can understand how these things go Yeah, and also We actually like people sending us documents Because we find documents on web and on web We don't know what their licenses are and we can't include them Officially into our regression testing frameworks. So when we have documents where you can give it to us under some But no some some normal license that will not create for us a big legal fees. So It will be useful for us