[Bf-committers] Luminance patches

[Kévin Dietrich]

Le 2015-02-18 00:15, Campbell Barton a écrit : 

>> The color wheels (HSV) come to mind. If the developer uses the rgb_to_bw function, they just broke their entire color chain. Now the developer shouldn't use the rgb_to_bw function, but how should they know that?
> 
> just include a comment above `rgb_to_bw`, only to use when handling
> colors outside Blender's/OCIO's color-pipeline.

(To reply to this above mail and recent comments made in D1082: ) 

I do admit that color managing the UI is a bit of an overkill, but Troy
does have a point: nothing would prevent someone to screw things up if
there is a stray function in BLI, even if there is some comment
explaining what's up (let's be paranoid for a second ;) ). IMO, if
there's no real way to get this kind of changes in master unless we keep
a function for UI use then this puppy should be in the UI's own
(private) API (some `UI_get_luminance()`), as I feel BLI is more for a
general/generic use. Then this function should have weights that makes
sense, by which I mean: so far no one can truly explain where this
`rgb_to_bw()` comes from; it feels very arbitrary (as in why not using
just the red channel, or the blue one?) and should be changed. With our
patches (D1082 by Troy, and D517 by myself), we can rationally explain
those weights, and the tendency should be kept.