NFN: strings in expressions

[basilkohler]

Possible syntax: 'str

[tschudin]

On Wed, 29 Oct 2014, Basil Kohler wrote:

Possible syntax: 'str

Isn't this how it is already implemented? At least I made it that way
for "constant identifiers".

When it comes to strings, I think that spaces as part of the string have
to be supported, wherefore we need to go for either ".." or '..'

[basilkohler]

We have not really looked into it yet. I have quickly tested it with a "call 2 /foo/bar 'foobar" and the abstract machine did not handle it correctly in that case. So I would say we keep it open until we have tested it some more.

Ofcourse you are absolutely right with the spaces and we will adopt "..." or '...'.

[tschudin]

so far, identifier constants were used internally and I have to figure
out how exporting values from the result stack for doing the upcall to
the outside world works.

On Wed, 29 Oct 2014, Basil Kohler wrote:

We have not really looked into it yet. I have quickly tested it with a "call 2
/foo/bar 'foobar" and the abstract machine did not handle it correctly in that
case. So I would say we keep it open until we have tested it some more.

Ofcourse you are absolutely right with the spaces and we will adopt "..." or
'...'.

—
Reply to this email directly or view it on
GitHub.[ADxFAkwKTgYeTHvQJltQd_s4004miQ39ks5nIQiugaJpZM4C0XH1.gif]

[blacksheeep]

Should now work with 'string... ' for call expression.
Maybe we also want to overload add etc. for strings?

[basilkohler]

'foo bar' works and is also supported by nfn-scala