Re: master 64e25cd: More robust NS hex colour string parsing

[Mattias Engdegård]

13 juni 2020 kl. 19.09 skrev Eli Zaretskii <eliz@gnu.org>:

> But the code you want to supplant explicitly does something different:

Yes, which is part of the reason why it's being supplanted.

> Of course, single-digit hex RGB specs are rarely if ever used these
> days, AFAIK, so maybe this isn't a problem in practice.  But still, I
> wonder what could we lose here or break.

This isn't going to be a problem in practice.

> How about (tty-color-values "#12345")? does it contradict yours?

No, it is an error that is (correctly) rejected by the new code in the same way 
that #xyz is currently rejected.

> You are changing the behavior, so get ready to hear bug reports, is
> all I'm saying.

That's fine, I'll be ready.
It's impossible to fix bugs without changing behaviour, so if we want the 
former, we have to allow for the latter, within bounds appropriate for the 
situation.

> Then color-values-from-color-spec, I guess.

Thank you, but a smidgen too many 'color' in there. Would 
color-values-from-spec do?