WindowSwitcher: blur background #2330
Conversation
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
3 times, most recently
from
50d77bd to
8061863
Compare
danirabbit
left a comment
danirabbit
left a comment
There was a problem hiding this comment.
Approve for UX, but still needs code review :)
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
9 times, most recently
from
6f4a0ed to
f942cd9
Compare
vapi/mutter-cogl-14.vapi
Outdated
| public void set_uniform_1f (int uniform_location, float value); | ||
| public void set_uniform_1i (int uniform_location, int value); | ||
| public void set_uniform_float (int uniform_location, int n_components, int count, float value); | ||
| public void set_uniform_float (int uniform_location, int n_components, int count, float *value); |
There was a problem hiding this comment.
Is there a reason this is a pointer? According to the vapi docs if possible you should use the vala provided syntax to make the semantics clearer: https://docs.vala.dev/developer-guides/bindings/writing-a-vapi-manually/07-00-binding-a-c-function-s-parameter-and-return-types/07-07-pointers.html
There was a problem hiding this comment.
It's a pointer to the start of array of float values in C header. I don't know if there's other way to pass it
There was a problem hiding this comment.
Well I guess you could pass it as an array? IIUC the array length would be n_components * count? In that case you'd probably have to do array_length = false in the vapi. On the other hand the deprecated Cogl.Program has a method with the same name and annotates the count as array length.
But i don't know enough about shaders and that stuff to say which it is 😅
Also the vapi seems to be generated by vapigen (at least that's what it says in the header) so ig we should probably put the change in the metadata file? https://docs.vala.dev/developer-guides/bindings/generating-a-vapi-with-gobject-introspection.html#fixing-vapi-generation-with-metadata
There was a problem hiding this comment.
I put it in the vapi file for now. I always assumed our vapi files were written by hand since the meson script in the vapi folder never worked for me. I guess we should ask @tintou about where should this change be
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
from
f942cd9 to
a686740
Compare
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
from
a686740 to
0c16fe6
Compare
leolost2605
left a comment
leolost2605
left a comment
There was a problem hiding this comment.
Gonna be honest no clue how the blur works but structure LGTM. It has a bit of a weird frame in light mode on a light window but ig we can ignore that for now 🤷
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
from
0c16fe6 to
353c813
Compare
|
@leolost2605 I don't see anything weird here. Can you send a screenshot please? |
|
I dont know if that's intended but it looks like the background overflows the frame especially in the corners |
|
@leolost2605 That's a ShadowEffect issue. I'll look into fixing it later |
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
from
353c813 to
27ad901
Compare
|
@lenemter what's blocking here? Would be great to get this in for 8.1 :) |
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
2 times, most recently
from
27ad901 to
9431076
Compare
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
3 times, most recently
from
4049181 to
9f9aafc
Compare
lenemter
force-pushed
the
lenemter/background-blur-effect
branch
from
9f9aafc to
812e6e6
Compare
New clean branch that should be easier to review
Window switcher is used as an example of how this effect can be used because Danielle showed some interest in blurring it: elementary/granite#747 (comment)
Can be rebase merged