I've seen this example:
<extern|(lambda (x) `(concat "Hallo " ,x))|Piet>

And I've created this:
<extern|(lambda () `(lambda (x) `(concat "Hallo " ,x)))>
which returns such a lambda, but I've not been able to invoke that macro

<extern|<extern|(lambda () `(lambda (x) `(concat "Hallo " ,x)))>|hi>
fails with: Wrong type to apply:

Is this because extern expects a string and not a list?

I tried this:
<extern|(lambda (x) `(eval x)|<extern|(lambda () `(lambda (x) `(concat
"Hallo
" ,x)))>|jo>

but still failure, so I remove the parameters of the name:
<extern|(lambda (x) `(eval x)|<extern|(lambda () `(lambda () `("Hallo ")))>>
But even that failed.

This one gives no error, only a black ? so it has most hope of being right:
<assign|x|<extern|(lambda () `(lambda () `(concat "Hallo ")))>>
<extern|(lambda (x) `(eval x))|<value|x>>

How should I invoke a lambda returned from scheme?

Sam

On Tue, May 22, 2012 at 8:09 AM, Sam Liddicott <s...@liddicott.com> wrote:

> My fake-page macros are working well, but don't quite do the trick.
>
> If I take the reader through the first part of development of page 1, then
> I have many revisions of a document fake-page page 1. That works fine. Each
> revision has a different effective namespace and it's own section numbering
> and so on.
>
> I then need to take the reader through development of the next part of
> page 1. These are further revisions of document 1 page 1, but showing
> further parts.
>
> Thus the first revision of the next part of page 1 is the SAME revision as
> the previous fake page 1, and will inherit all the labels and numbering and
> so on.
>
> However when I show the next revision of the second part of page 1 it is a
> new revision and inherits nothing! Aggh!
>
> If I made it the same revision then it would inherit not only the last
> revision of the first part, but also the first revision of the second part.
>
> So clearly I need closures, so that in any revision I can extract a
> closure which allows further pages to continue from that point at which the
> closure was made. More than once, without conflict.
>
>
> I'm guessing that I would need a scheme function to create the closure. It
> would need to be a closure that accepted either any number of arguments or
> a tuple, and effectively invoked "compound". Or maybe even just took one
> argument which it eval's and returned, like "identity" but in a closure.
>
> I'm not asking if I'm nuts, but I'm asking how to do this. As I search my
> lisp memory I can think of "curry" and stuff like that but things are
> complicated with the macro layer as well.
>
> I shall post here as I develop the idea but if anyone can "do my homework"
> for me or give me a clue I would be very grateful
>
> Sam
>
_______________________________________________
Texmacs-dev mailing list
Texmacs-dev@gnu.org
https://lists.gnu.org/mailman/listinfo/texmacs-dev

Reply via email to