Review Request 123809: Use list of hard coded colors instead of generating it

[Philipp A.]

> On Mai 16, 2015, 3:44 nachm., Philipp A. wrote:
> > i’m also against this. i’ve dived a bit into color theory, and if the algorithm to choose the set of colors isn’t named, there’s not much to argue for it.
> > 
> > i think we want a qualitative color palette, and shhould choose a way to generate easily distinguishable colors.
> > 
> > * http://stanford.edu/~mwaskom/software/seaborn/tutorial/color_palettes.html
> > * http://colorbrewer2.org/?type=qualitative&scheme=Paired&n=12
> > * http://datavisualization.ch/inside/how-we-created-color-scales/
> 
> Sergey Kalinichev wrote:
>     >i’m also against this. i’ve dived a bit into color theory, and if the algorithm to choose the set of colors isn’t named, there’s not much to argue for it.
>     
>     See my previous comment.
>     
>     >i think we want a qualitative color palette, and should choose a way to generate easily distinguishable colors.
>     
>     I took a quick look at the links that you provided, but couldn't find a set of 30-35 distinguishable colors or an algorithm to generate it. Seems like all those links about generating 10-15 colors which is hardly enough.
> 
> Philipp A. wrote:
>     then how about [this](http://tools.medialab.sciences-po.fr/iwanthue/)?

just play with the sliders and keep in mind:

> If you produce too many colors, you will still be able to use the first colors, since they will be distinct. 

so we can create 100 colors and they will simply become less distinguishable the more variables there are, but stay distinguishable.

we’ll want to leave out very low & very high brightness (“white” and “black”), as well as very low chroma (gray):

so maybe (all hues) chroma ? [0.2, 3] and lightness ? [0.1, 1.4]

for 100 colors, you’ll also need the force vector setting.

also we can eventually do this algorithmically in kdevelop, not on the website.


- Philipp


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/123809/#review80465
-----------------------------------------------------------


On Mai 16, 2015, 8:07 vorm., Sergey Kalinichev wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/123809/
> -----------------------------------------------------------
> 
> (Updated Mai 16, 2015, 8:07 vorm.)
> 
> 
> Review request for KDevelop.
> 
> 
> Repository: kdevplatform
> 
> 
> Description
> -------
> 
> This fixes 2 issues:
> 
> *Currently rainbow highlighting only works for the first 10 declarations in Function and Other contexts. All other declarations get highlighted with the default color.
> *The current interpolation algorithm is very poor, it can generate only about 10 different colors, all other look alike.
>     
> So this patch adds ~35 distinguishable colors, which should be more than enough for most cases.
> 
> 
> Diffs
> -----
> 
>   language/highlighting/colorcache.cpp e2729f2 
> 
> Diff: https://git.reviewboard.kde.org/r/123809/diff/
> 
> 
> Testing
> -------
> 
> 
> File Attachments
> ----------------
> 
> Before
>   https://git.reviewboard.kde.org/media/uploaded/files/2015/05/16/2fb289f0-69d8-4817-bf0d-d1691c98d3c1__test_highl_b.png
> After
>   https://git.reviewboard.kde.org/media/uploaded/files/2015/05/16/7bcdb744-e47f-435b-a02e-d0ad35b4a59e__test_highl_a.png
> After (dark)
>   https://git.reviewboard.kde.org/media/uploaded/files/2015/05/16/b67be6b5-b4bf-4d74-8ed9-206e897daa14__test_highl_adark.png
> 
> 
> Thanks,
> 
> Sergey Kalinichev
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20150517/38bc658a/attachment.html>