- User Since
- Jan 9 2020, 11:14 AM (58 w, 6 d)
- dividing the line width by U.pixelsize as suggested
- change use of fontid for function I missed
Tue, Feb 23
revert to GPU_PRIM_LINES
Mon, Feb 22
@Julian Eisel (Severin) Let me know if the lines have thickness for you now
- Changed the way fontid is aquired as suggested
- Changed the shader for drawing lines from GPU_PRIM_LINES to GPU_SHADER_2D_UNIFORM_COLOR
- Changed RNA_def_float_percentage to RNA_def_float_factor
Sat, Feb 20
update patch to use "rctf" for "UI_draw_roundbox_3ub_alpha"
Jan 25 2021
Jan 21 2021
- boxes behind percentages use UI_draw_roundbox_3ub_alpha now
Jan 20 2021
Jan 19 2021
Jan 18 2021
- rebased onto latest master
- moved static indicators to the bottom
- percentage indicator always follows cursor
- ticks every 10% (changed logic to int for that part)
- that also fixed the issue with 0% sometimes turning into -0%
- Ctl changed to Ctrl
- multiplied padding by U.dpi_fac to be consistent. I think it does the same?
- using wmEvent.mval now
Jan 15 2021
Jan 14 2021
Dec 22 2020
in that case it looks good to me
Dec 13 2020
the same check is also done in nla_eval_domain_action()
Any reason why we can't use the same function there?
Nothing to add from my side. nla_eval_domain() indeed does this check already so it seems to be safe to get rid of
Dec 8 2020
Dec 2 2020
Nov 28 2020
Nov 26 2020
Nov 22 2020
Nov 20 2020
Nov 19 2020
Nov 18 2020
Nov 17 2020
Nov 16 2020
I tweaked the weighted moving means algorithm to respect x-position D9477
However I still think the smoothing algorithm could be better
Nov 15 2020
I implemented a solution to this in D9477
rebase and improve algorithm by taking keyframe x position into account
Nov 13 2020
Sorry I missed that. I didn't know asserts would cause a crash in debug builds.
In that case I will close that issue then.
I think it makes no sense to keep it as a known issue if it never is a problem in final builds.
Nov 12 2020
This can be seen as resolved.
The issue with RNA goes deeper and would need a Design task, but it hasn't been an issue since.
Except that there is a regression that enum properties crash blender T82669
Nov 10 2020
Nov 9 2020
- Moved declarations into ED_util.h
Nov 8 2020
Nov 6 2020
I just realized ED_util.h already exists in editors/include.
thanks for the feedback. I didn't know about LISTBASE_FOREACH. Looking over the code there are quite a lot of places where this can be used. I'll note it down to be refactored in one go at the end
Nov 5 2020
Nov 4 2020
building works now but there seems to be an error still?
This still won't build for me.
Nov 3 2020
@Sybren A. Stüvel (sybren) I agree that having declarations in two headers is less than ideal.
But I am not sure how to proceed on this. Both ED_screen.h and ED_space_api.h have some drawing functions declared. But there is probably a good reason for them to be there.