>>>>> "Eli" == Eli Barzilay <xxxxxx@barzilay.org> writes:

Eli> This is a comment on the #!...!# thing.  I think that more Schemes
Eli> implement a #|...|# multiline comment, which is why the srfi says that
Eli> the suggestion is not to add another comment construct but handle this
Eli> specifically as a preprocessor.  So my comment is that given the fact
Eli> that #|...|# is usually used, then there is no need for the ugly and
Eli> one-line-limited

Eli>   #!/bin/sh
Eli>   string=? ; shell-command

Eli> hack, since you can achieve the same effect with

Eli>   #!/bin/sh
Eli>   #|
Eli>   shell stuff, ends in an exec
Eli>   |#

There must be a misunderstanding: There's no need for this hack anyway
in this SRFI, and it is mentioned nowhere.

Eli> Also, I don't know how reliable this is but I think that Unix systems
Eli> are supposed to have /bin/sh available - the man page for system
Eli> specifies using "/bin/sh -c string" to do its job.

Eli> The last comment is that I think it is more common, and therefore
Eli> would be better accepted, to stick to a command line flag to load some
Eli> library (as in "-l srfi0" instead of "-srfi0") and just some
Eli> specification of the way arguments are made available to the
Eli> program

SRFI 0 and SRFI 7 are *special* SRFIs, and they certainly will not be
"loaded" in the usual sense in most contexts.  Specifically, SRFI 7
contains a loading mechanisms.  This is why this SRFI talks only about
those two.

--
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla