CSS first-child/last-child styling inconsistent if taskbar sorting is enabled.

[TheBoctor]

When setting sort = true in a taskbar module, new buttons appearing on the taskbar may use incorrect styling. This seems like a possible race condition, but I am unsure.

For example, if you use CSS selectors to style the first/last taskbar buttons differently than all of those in between, then open a properties dialog in Thunar, the buttons will not use their intended styles based on order (Look at the border/seam between Thunar and the terminal):

This never happens if sort is set to false for the taskbar:

[LBCrion]

Interesting. I don't think this is a race condition, since sfwbar doesn't assign first/ last attribute explicitly. When sorting items, sfwbar removes all items from a grid and then add them back in a correct order. It looks like some ordering info may persist from when items were added originally. I suspect I may need to dig into gtk source to identify the cause for this. Can I check which version of gtk+ you're using? I'm travelling this week, but will try to debug this once I'm back in a few days.

…

On Wed, 28 Aug 2024, 16:54 TheBoctor, ***@***.***> wrote: When setting sort = true in a taskbar module, new buttons appearing on the taskbar may use incorrect styling. This seems like a possible race condition, but I am unsure. For example, if you use CSS selectors to style the first/last taskbar buttons differently than all of those in between, then open a properties dialog in Thunar, the buttons will not use their intended styles based on order (Look at the border/seam between Thunar and the terminal): image.png (view on web) <https://github.com/user-attachments/assets/1b9b1ddf-1393-47a5-9e07-557fe9caa7c4> This never happens if sort is set to false for the taskbar: image.png (view on web) <https://github.com/user-attachments/assets/4cbbb5d3-94f5-4109-9539-f025f3ba44e8> — Reply to this email directly, view it on GitHub <#225>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ASHPFFEZETOYYDQIY7AQLL3ZTXJCJAVCNFSM6AAAAABNILIGHSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGQ4TEMJVGI2TOMA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[TheBoctor]

I am using GTK 3.24.43-1, from Arch Linux's repositories. sfwbar is a fresh build of v1.0_beta15.

[LBCrion]

Would you mind posting your config? I haven't been able to replicate this so far, which means I'm probably missing something. Also, which compositor are you using?

…

On Fri, Aug 30, 2024 at 5:42 PM TheBoctor ***@***.***> wrote: I am using GTK 3.24.43-1, from Arch Linux's repositories. sfwbar is a fresh build of v1.0_beta15. — Reply to this email directly, view it on GitHub <#225 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ASHPFFAQTYAYJ3VT4GRBR5LZUCOHZAVCNFSM6AAAAABNILIGHSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRRHE2DCNJZG4> . You are receiving this because you commented.Message ID: ***@***.***>

[TheBoctor]

I was using labwc 0.8.0. An unrelated problem was causing the whole compositor to crash and dump core every few times I un-suspended my laptop after >8 hours of sleep, so I won't be using it again until I can try debugging that.

My configuration is uploaded here. If you haven't yet been able to replicate this with a minimal example, I'm hoping that it isn't a bug and that I've just made a mistake somewhere.

[LBCrion]

I think I see the issue. CSS nodes like 'grid#taskbar > :last-child button#taskbar_active' select the last child in the grid that has a specific name. This may assign the style to children you may not want. What you probably want to use instead is 'grid#taskbar > :last-child button' this will get you the last button the grid regardless of it's name. This should get cleaner with the changes I just committed where taskbar item states will become classes rather than separate names.

…

On Mon, Sep 2, 2024 at 4:30 PM TheBoctor ***@***.***> wrote: I was using labwc 0.8.0. An unrelated problem was causing the whole compositor to crash and dump core every few times I un-suspended my laptop after >8 hours of sleep, so I won't be using it again until I can try debugging that. My configuration is uploaded here <https://gist.github.com/TheBoctor/250c2233c1dcee29c52d5f377ccee774>. If you haven't yet been able to replicate this with a minimal example, I'm hoping that it isn't a bug and that I've just made a mistake somewhere. — Reply to this email directly, view it on GitHub <#225 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ASHPFFGSTMOBLPGAHJASCPTZUSAAZAVCNFSM6AAAAABNILIGHSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGMRUHE4TAMBTGA> . You are receiving this because you commented.Message ID: ***@***.***>