Re: Difference when calling guile script from C vs interpreter

[Taylan Kammer]

On 20.10.2020 14:58, Jean Rene Dawin wrote:
> Hi,
>
> when following guile script:
> ______ gp.guile _________
>
> (snip)
> ______________________________
>
> is run from the interpreter, the output shows the following:
>
> (snip)
> Total time: 0.327908556 seconds (0.121140347 seconds in GC)
> #t
>
>
>
> When the same script is loaded from following C program:
> _____ gp.c _______
> #include <libguile.h>
>
> int main(int argc, char **argv)
> {
>     scm_init_guile();
>     scm_c_primitive_load("/home/gp.guile");
>     return 0;
> }
> ___________________
>
> the result looks like this:
>
> (snip)
> Total time: 1.939099187 seconds (0.570765622 seconds in GC)
> #t
>
> Is this difference expected?

When you use the "guile" executable to run a file, it automatically 
compiles it first, then runs the compiled version.  (For this reason I 
wouldn't use the term "interpreter" in this case.)

When using primitive-load, it doesn't auto-compile it (and I think it 
doesn't even look for an already compiled version), instead it directly 
calls the interpreter on the source code.  (This time it really is the 
interpreter.)

It's generally preferable to *extend* Guile with C code, rather than to 
*embed* it in C code.  This means you turn your C code into libraries, 
which are turned into Guile modules, then you write a Guile program that 
uses those modules.  I.e. you call to C code from Guile, not to Guile 
code from C.

If you want to really prefer to embed and not extend, then perhaps you 
can make your Guile files available in the %load-path and use 
scm_primitive_load_path (see documentation).

I don't know if/how the Guile compiler can be called from C.  Maybe 
someone more experienced can help further.


- Taylan