Re: master d57bb0c: Treat passed strings as raw-text when percent-escaping in epg

[Stefan Monnier]

> The question is not what will or won't work, the question is what is
> better, I think.  And I don't think Stefan explained his rationale for
> preferring UTF-8.

What the code does is:
take a multibyte string with % escapes of bytes which are to be
interpreted as utf-8, so the Elisp does: encode with utf-8, process the %
escapes bytes, and decode back using utf-8.  The encoding+decoding with
the same coding-system should ensure that the non-escapes part of the
string is preserved and the use of utf-8 is chosen because that's what
the %-escapes are defined to use.

So yes, now I'm sure the patch is right.


        Stefan