Created attachment 137400 [details] Screenshot with the bug SUMMARY Some symbols are not rendered properly in Elisa (see the screenshot). STEPS TO REPRODUCE 1. Open Elisa 2. Select track and click 'View Details' button 3. Click 'Modify' button 4. Select a tag type and try to type something in the new field 5. Observe the bug OBSERVED RESULT Glyphs are not rendered properly, i.e. they have their top cut. EXPECTED RESULT Glyphs are rendered properly SOFTWARE/OS VERSIONS Linux/KDE Plasma: Arch GNU/Linux KDE Plasma Version: 5.21.3 KDE Frameworks Version: 5.80.0 Qt Version: 5.15.2
*** This bug has been marked as a duplicate of bug 414266 ***
I suspect this isn't a duplicate of bug 414266, where the entire rendered font texture has duplicated or missing rows/columns of pixels when shown on-screen. Looking at the attachment, I think each glyph of the font (PT Sans?) has its top row (or few) of pixels cut off, meaning it's a text layout/shaping/rendering bug rather than a texture-drawing bug. But I'm not 100% confident. To verify, can you type something like F-a._#@///\\\ to make sure there are no missing rows of pixels extending across the whole text area? The first few glyphs have various letter heights, and the slashes/backslashes will make missing rows of pixels obvious. If you want to go overkill, try debugging Elisa in RenderDoc, and look at the underlying textures to make sure the glyphs have the top missing (meaning it's not bug 414266).
Created attachment 137410 [details] Screenshot 2
(In reply to nyanpasu64 from comment #2) > PT Sans? Yes. > To verify, can you type something like F-a._#@///\\\ to make sure there are > no missing rows of pixels extending across the whole text area? The first > few glyphs have various letter heights, and the slashes/backslashes will > make missing rows of pixels obvious. Done, see screenshot 2. Looks like you are right. Note that the bottom row is also missing. But if you take a closer look, you will notice some pixels got dark-grey and not disappeared completely. > If you want to go overkill, try debugging Elisa in RenderDoc, and look at > the underlying textures to make sure the glyphs have the top missing > (meaning it's not bug 414266). Sorry, I'm not really familiar with KDE stack.
I don't think it's bug 414266 then. I don't know if normal users like me are allowed to change status, or only KDE staff, but it seems I'm allowed to pick any status except for CONFIRMED (due to the "canconfirm" permission), so I'll do it. > Note that the bottom row is also missing. I don't know if the bottom row is actually missing, or if the underscore _ vanished because it's only 1 character tall. I don't know if the period was cut off at all or not, since it looks normal. You could try comparing . and :, and see if . has missing pixels compared to the lower dot of :. > But if you take a closer look, you will notice some pixels got dark-grey and not disappeared completely. Are you referring to the underscore _? If so, it's probably slightly more than 1 pixel tall. > Sorry, I'm not really familiar with KDE stack. RenderDoc is an OpenGL profiler unrelated to KDE, but I understand if you don't know how to use it.
> To verify, can you type something like F-a._#@///\\\ to make sure there are no > missing rows of pixels extending across the whole text area? Works for me with more or less the same versionf of the software that you have. I'm using 200% scale and 11pt Noto Sans font though. If it's not exactly Bug 414266, it's probably something similarly deep in the stack. Either way, it's not a bug in Elisa itself. :)
*** Bug 435538 has been marked as a duplicate of this bug. ***
*** Bug 436080 has been marked as a duplicate of this bug. ***
Should be fixed by the fixes for upstream bugs https://bugreports.qt.io/browse/QTBUG-96112 and https://bugreports.qt.io/browse/QTBUG-83626, which is being backported to the KDE patch collection.