 Thank you very much. Is it okay now? Can you hear me? Oh, so My name is Palas Vargo and I'm hacking on the office since 2018 I'm working at Aldotropia since June most of my work is related to cork Chart but I also have some work in writer and impress as well and also I worked a lot of on XML interoperability and new features, of course I will show you new implementations features and backfixes around charts and Autofilter issues in cork, so let's start them This feature was the movable chart data point labels earlier It wasn't possible to move these objects and shapes so it was a shape with predefined six positions like top middle right Left and etc. But there were some other positions like best fit in case of pie chart By the way, it is possible to move these objects in Microsoft Office The expectation was to make the data labors movable Also in chart in the charts, so let's move. Let's see how it looks like Yeah, so as you can see now These data labors are moved away from their predefined six position position It is possible to move them with mouse or arrow keys as well To make it work I had to add some new objects to the data label shapes objects and also I had to manage the position and size of the shapes depending on which chart type we have and a lot of other things as well like Import export filters I had to create new objects For leader lines and had to manage overlaps and etc. I will show you examples for that Obviously moving the database Working in case of or chart types or also the important export almost Let's see another trivial chart type On this slide you can see how it looked like in case of pie chart before any Implementation it is a simple for the outside position and it wasn't possible to move the labors yet There were also many overlaps and unreadable labors The picture was taken after an ODS export Also the pie chart has different kind of labors positions as I mentioned earlier So it was a it was harder to handle the movements from these positions But I will talk about later. Let's see how it looks like now This pie chart had a labors with an outside positions and From the labors were moved from that positions To that exact position where they are now on the picture and The image was made after an ODS export so also the ODS important exports are working well Also, there are no overlaps there which is partly due to a new algorithm that we use on the labors as well You can see leader lines there But let's talk about those later. Let's see another example for the OXML In this slide, there is a picture about how it looked like After an OXML import before any implementation In this case the chart labors moved outside from the chart area on the top and the left side as well By the way, the data labors above should be inside the red rectangle And also on the left side as well in the red rectangle and would be connected to leader lines But it didn't work yet. Let's see how it looks like after a correct import and implementation It's a picture how it looked like after the OXML import fix and the data labors movement implementation These labels were moved to those exact position in Excel where they are on the picture Now they are in the same position in caulk as well By the way, it is also possible to move these labels anywhere inside the chart The calculation is based on simple coordinate geometry theorem and we calculate the move position from fix zero position, which is the top left side of the chart and the previous position of the data label Also as I mentioned the hardest part was making it work in case of pie chart because there is a position named Best fit, which is a very different Algorithm than the other ones. Let's see what's happening there Two things are important here As I mentioned, we calculate the position from the zero position and the left top left Corner of the chart position Second thing is the best fit position position, which is also what didn't work well You can see on the on the left side, there are overlaps The leader lines make no sense Which was a half developed Implementation from earlier on the right side, you can see the results How it looks like now after rewriting some parts of the best fit algorithm How it works First we check that the label is fit Inside the pie slice. If yes, we keep it in if not we move it outside while We're watching the other labels the pie chart position and Side of the chart. Also, we are avoiding to not being any overlapped with the other labels Let's see another Example on the left side. There is another example how it look like after an OXM report. You can see Leader little leader lines as well And the picture on the right side was made after improving the best fit algorithm the leader lines the label movement and the field earth as well. I think it looks better now And the total labels that overlapping each other in case of pie chart as well Let's talk a little bit about the leader lines, which was also other improvements These lines connected to the data labels with the correct data points It is also a new object and it can be turned on and off These lines only visible if the distance between the data labels and points are large enough other hands we don't create them and don't rendering them It's not a very interesting slide, but it's Show some related bugs which had been fixed by improving this area There was a lot different kinds of develop started from similar developments to reviving more complex algorithms but Let's move on to another implementation on this Slide there was a custom shapes Inside the chart objects. It is possible to draw different kind of shapes into the charts like on the top of the slide, but the problem was that We didn't import and export them to OXML The image was created before the implementation of course and there should be Shapes inside the charts each other inside the chart. Let's see how it looks like now After the fix and implementation all kind of Shapes can be import and then exported inside the chart as well all kind of file format. There was also Issues around the ODS import and export, but that's also works now Let's see another feature It's really interesting one. This feature is That now we can use named Ranges for data ranges in charts in the picture You can see that it wasn't possible to give any named ranges or data ranges and the text box of the range was read Also the chart wasn't appeared by the way It is possible in Microsoft text and because of that after an OXML import the charts are disappeared and Transformed sometimes transforming into images Sometimes it's appeared Let's see how it looks like now As you can see it is possible now to use named ranges and database ranges in charts Also the ODF and OXML filters are working So there won't be any data losses Of course, you can only use valid named ranges or Data base ranges otherwise the text box of the data range will be read just like before Also, as I mentioned you can use database ranges as well Let's move on to another area That's the auto filter area It is a very popular and widespread around the users The first part of these improvements was a date filtering The most problem was related to the OXML import and export for example There was missing datum filters and the results of the Filtering in other columns was also missing as you can see on the picture It shouldn't be empty. Of course Also after the export all the filter disappeared Let's see on the left side You can see that after OXML import the data range and the auto filter arrows are completely disappeared So the header rows cannot make visible again Only if you create a new filter on that range or just make the header rows visible again On the right side you can see the results after the Filters fix Filtering works well in the correct and the correct values are checked and unchecked in case of all date and time filters They are imported and exported correctly the OXML as well. Let's see another feature It is the formatted values in the filter list as you can see before any implementation not the formatted filter elements, but But the unformatted one appeared in the auto filter list So it was less user friendly and it was Pretty slow with a large amount of data for example, if you had a column with formatted Guarantee or rounded values the default unformatted numbers appeared in the filter list Also some number values have been compared As a string to each other in the auto filter and because of that it was much slower Then if we would compare them as a number and filtering by that So let's see how it looks like now As you can see there are Formatted filter values in the C column and also in the auto filter list as well In this case, it is just an example for currency format, but it should work with all kinds of cell format but even if they are formatted values the auto filter engine comparing them as a formatted numbers and Not as a string of course, which makes it faster especially with a lot of data The hardest part was here to handle the duplicated values which are came from the rounded values because if we rounded the values very possible that we Get some get some same values but that's also We are also handling those as well. That's all so the next one This is also a new feature in the auto filter area With this you can see the hidden elements in and their status in the filter list What are hidden by other filters of different columns for Somebody it can be useful especially if once to see which elements were checked or unchecked before They are disappeared by Another filters especially in case of large filters and lots of columns. It can be very useful Also, it cannot be changed which is the checkboxes. She's very important. Otherwise the functionality of the auto filter Would have changed So you cannot check or uncheck the elements and the inactive elements are at the end of the list So it won't bother using the auto filter By the way in the future, maybe This feature will be able to turn on or off, but that's not implemented yet it's also Slide some related bugs Which had been fixed by improving the auto filter area. It was a lot different kind of Develops develops Also, I couldn't present a lot of other things like the issue when the counting Counting the the number of auto filter record doesn't work in the status bar and there was wrong results on the button of the On the sidebar That's also working as well. Let's see some future plans Tell allotropia. I'm also Working on running the office in the assembly Now we have a good results as well and the car can be run in a Browser that still have a lot of other issue there which waiting for more work But it's running. We will talk about it with Torsten for a later talk and other Features in the web assembly but let's talk about later and Time for questions If there anyone I was too fast, but I Like talking fast, so I hope everyone everything was clear But if there's no question, thank you very much Have you already any idea what you will be doing next in Calc? I don't Know the moment, but probably it's working so probably I will continue my work around the auto filter and Chat issues there are still remaining some also at the chart issues for example the the best fit algorithm it's not working perfectly in case of Exporting to the OSM out because as I mentioned the the excel and Cog use different algorithm to count in the best position of the Data labels and because of that we cannot Export it then to the same position of course and the most the hardest part That moving the data labels from the best fit positions because the original position is different from in the key in the application because of that we cannot know We cannot import and we can import perfectly, but we cannot export perfectly to the same position So it will never be the same in the two application, but at least It works at the hotel imports perfectly at the moment So your question probably keep working on the child and auto filter and many others side Of the clock of course any other question if now is thank you very much