Support alpha in hexadecimal colors #3052
Conversation
vmcj
force-pushed
the
hex_shorthand_alpha
branch
from
d0b078e to
83e8167
Compare
webapp/src/Utils/Utils.php
Outdated
| @@ -349,6 +352,12 @@ public static function convertToColor(string $hex): ?string | |||
| */ | |||
| public static function parseHexColor(string $hex): array | |||
| { | |||
| // Ignore alpha in hexstrings | |||
There was a problem hiding this comment.
Why silently ignore the alpha channel?
There was a problem hiding this comment.
It might indeed a bit harsh to do this silently. Do we ever think we need the alpha channel? Otherwise I think I'll add a optional boolean for this behaviour (optionally also returning rgba instead of rgb), alternative is to check all calling functions and handle this there.
There was a problem hiding this comment.
It's unlikely that we need it. But I'm either for supporting it properly or not allowing it at all to avoid user confusion.
There was a problem hiding this comment.
The spec allows for sending such a string "#ff00ffaa" so we need to store it. I think my fix already makes that we never pick it up but I'll think about this some more. Maybe just supporting it is easier in that case.
vmcj
force-pushed
the
hex_shorthand_alpha
branch
6 times, most recently
from
892068e to
ebf94a2
Compare
No functional change intended.
vmcj
force-pushed
the
hex_shorthand_alpha
branch
from
a63d9fb to
d1a8c57
Compare
The Alpha is allowed by the CLICS spec but we don't really use those in the UI normally. The only place where we have to fully ignore the alpha is for the luminance. In that case we now by default merge it with a white background most likely this will be the same for either background but this was not tested. As bonus we now also test for the shorthand (#ABC) versuse (#AABBCC). The functions using the strings have moved from the TwigExtension to utils as the they don't really depend on Twig itself. This makes testing easier and can use this in the future for other elements.
vmcj
force-pushed
the
hex_shorthand_alpha
branch
from
5369d0c to
f4e085c
Compare
This is as a basis to detect problematic colors coming from the primary.
I can't think of any place where any consumer would ever want to display the colors with some opacity. So I wonder if it's intentional that the spec links to a document which has this.