Am So., 12. Mai 2019 um 20:42 Uhr schrieb Shiro Kawai <xxxxxx@gmail.com>:
>
> One issue of S-expression metadata is that S-expression reader ismore involved than a simple finite automata.
> If you look for a special character sequence without fully parsing S-expression, you can just use a magic comment
> or #!identifier token just as well.
>
> #!identifier token, as Marc brought up, is nice because we already have it.  The drawback I see is that
>
> - Probably we want something like #!coding=<value>, rather than just #!<identifier>.  E.g. #!coding=utf-8 instead of #!utf-8.
>   The latter can eat up #!-namespace quickly.

I think that standardizing how #!-directives can receive parameters is
also of independent interest:

For example, Scheme may support floating point numbers of different
precisions. One may want to tell the Scheme reader with which
precision floating point constants in the datum representations are
read. For that purpose, something like #!precision=25 looks like the
way to go.

Of course, one has to talk about the precise syntax.

Marc

[...]