Re: Question: How does color-lighten-name work?

[Eli Zaretskii]

> From: Robert Pluim <rpluim@gmail.com>
> Cc: Pedro Andres Aranda Gutierrez <paaguti@gmail.com>,  emacs-devel@gnu.org
> Date: Fri, 28 Oct 2022 10:09:48 +0200
> 
> >>>>> On Fri, 28 Oct 2022 10:09:34 +0300, Eli Zaretskii <eliz@gnu.org> said:
> 
>     >> From: Pedro Andres Aranda Gutierrez <paaguti@gmail.com>
>     >> Date: Fri, 28 Oct 2022 08:23:16 +0200
>     >> 
>     >> (require 'color)
>     >> (let ((newbg "#102030"))
>     >> (message "newbg: %s" newbg)
>     >> (setq newbg (color-lighten-name newbg 10))
>     >> (message "newbg: %s" newbg))
>     >> 
>     >> evaluated in the *scratch* buffer. will print the following:
>     >> 
>     >> newbg: #102030
>     >> newbg: #11ab23563501
>     >> 
>     >> Shouldn't the second colour spec have only 3 bytes?
> 
>     Eli> No, it uses the default format of 4 digits per component.
> 
>     Eli> We could perhaps add an additional optional argument to specify how
>     Eli> many digits per component to produce.
> 
> By the principle of least surprise, perhaps default to the same number
> of digits as the input?

That'd be a backwards-incompatible change.  Too late for that, I
think.  So if we introduce that, it will have to be via the additional
argument.  (FWIW, I see no surprise in the current behavior; I was
actually surprised that it was surprising.  You do understand that the
#RGB notations are equivalent no matter how many bits per component we
show, yes?)