1. When saving the very large spreadsheet at https://mega.co.nz/#!5d0j3b6A!chMGwxtsd3qKHPPkWKOrDlryAFtaZrI91dUiuQ3Jva4 to xsls-format, calc consumes about 3.8 GB of memory. When closing the document and working with an empty writer document, memory consumption only decreases to about 3GB. 2. Close the LibreOffice instance which was used for step 1. When opening the xsls-file generated in step 1 with a fresh libreoffice instance, calc consumes about 1.3GB of memory (which is ok). When saving the file again to xsls, memory consumtion grows to 6.8 GB. After closing the document and working a bit with writer, it stays at 5GB. Beside the really huge memory consumption, the memory used for both steps should be freed. Instead, memory consumption only decreases slightly when closing the calc window which contains the large spreadsheet.
Excel-2010 just needs 300mb of RAM after loading+saving the document generated by LibreOffice.
Markus - I'll test this one when I'm at home but I thought it might be a good test case for you (150 megs...pretty hefty file). Might have some cool stuff going on that can be used to track down performance issues.
OK, I'll add it to my list. But mainly the memory consumption after closing Calc. That the current OOXML filters are not very efficient is known and has to wait for something like the orcus implementation.
Sorry for the delay - so I confirmed enough to mark this as NEW and confirm that it's a good test case to improve our filters. IMO there are 2 problems that highlight everything else. 1. Opening the ODS itself takes up nearly 3 gigs - seems excessive for a 150 meg file 2. The memory is never freed up because of point #2, I can't confirm the save to XLS as I only have 4 gigs of RAM and 1 gig of SWAP. Markus - do you have a system with insane amount of RAM (8+ gigs?) to confirm behavior in point 1 & 2 about saving to xlsx - first time only being 1ish gigs but then after it just eating RAM alive?
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.