https://bugs.documentfoundation.org/show_bug.cgi?id=165624
V Stuart Foote <[email protected]> changed: What |Removed |Added ---------------------------------------------------------------------------- Blocks| |125077 CC| |heiko.tietze@documentfounda | |tion.org, [email protected], | |[email protected] --- Comment #3 from V Stuart Foote <[email protected]> --- Reminds me we'd done similar for bug 72125 [1] when populating the expert configuration dialog and bug 109158 populating the autocorrect dialog. But that caused complaints about formatting with listings over the 100 count bug 99071 [2]. Admittedly that was for the UI rather than document canvas, performance hit came from doing pixel perfect calculations of the text strings, height and width. The best performance was with capped count of entries. But worked around by generalizing the width/height calcs, or incrementally exposing the list as needed. For the calc sheet rows we either need pixel perfect rendering of height and width, especially when a printer is involved. But for performance of opening and calculations while working, the user should have the option to ignore row heights string widths. And that is where we arrived for bug 124098 and its expert config value of "RecalcOptimalRowHeightMode" [3] which gives the users control, but not well known and the default is set to 'Recalc always' for "least surprise", though for performance it maybe should be 'Recalc never'? But that might be to extream for folks used to pixel perfect cells on their sheets. [1] https://cgit.freedesktop.org/libreoffice/core/commit/?id=62ea355b2679073b8ee326df5793231996136da9 reverted for bug 99071 with [2] https://gerrit.libreoffice.org/c/core/+/39951 [3] https://gerrit.libreoffice.org/c/core/+/163421 Referenced Bugs: https://bugs.documentfoundation.org/show_bug.cgi?id=125077 [Bug 125077] [META] regressions introduced by row height recalculation on document load -- You are receiving this mail because: You are the assignee for the bug.
