proposal for passing refinements
[1/2] from: agem:crosswinds at: 26-Feb-2001 9:44
my main interest was to discuss the following syntax:
[ a-func/opt/:bool 123 ]
where opt and its arguments are passed if 'bool is true.
As an inbuild enhancement request for refinements,
without need for [do refined].
It looks similar to the usual path-notation for arrays
(array/:i), saying »parts of this path are calculatet«.
And i think it is readable (for few refinements).
IMHO a inbuild »lightweight« refinement-passing would be nice,
but a proper way is unknown.
could this be it?
My implementation is to play with the syntax and look if it is handy.
And to collect some comments from other programmers :)
Volker
>>>>>>>>>>>>>>>>>> Ursprüngliche Nachricht <<<<<<<<<<<<<<<<<<
Am 25.02.01, 11:23:52, schrieb "Ladislav Mecir" <[lmecir--mbox--vol--cz]> zum
Thema [REBOL] Re: (no subject):
[2/2] from: lmecir::mbox::vol::cz at: 26-Feb-2001 21:16
Hi,
> my main interest was to discuss the following syntax:
> [ a-func/opt/:bool 123 ]
<<quoted lines omitted: 10>>
> And to collect some comments from other programmers :)
> Volker
I think, that there is still a problem with referential transparency. It
doesn't look as a solution I would suggest.
My DO-PATH solution was referentially transparent, something like: [do-path
:a /ref1 if bool [/ref2] /end-path arg1 arg2] is possible too, but there
still seems to be some food for thought.
Regards
Ladislav
Notes
- Quoted lines have been omitted from some messages.
View the message alone to see the lines that have been omitted