Hi,
Calling Parrot_init explicitly on a Parrot_Interp returned by Parrot_new is
useless.
(Parrot_new does it itself and Parrot_init doesn't do anything on already
initiated interpreters)
Because it makes the api more complicated i propose to remove it from
the docs.
I attach a patch which removes the description of Parrot_init from
doc/embed.pod, and the call to it from the code examples withing.
I currently work on embeding parrot from perl5 and would like to know
of any previews attemps.
Pawel Murias
Index: embed.pod
===================================================================
--- embed.pod (revision 12189)
+++ embed.pod (working copy)
@@ -17,7 +17,6 @@
argc--; argv++; /* skip the program name */
interp=Parrot_new(NULL);
- Parrot_init(interp);
if(PARROT_JIT_CAPABLE) {
Parrot_set_run_core(interp, PARROT_JIT_CORE); /* activate JIT */
@@ -204,10 +203,6 @@
Allocates and returns a new Parrot interpreter. C<parent> is NULL for the main
interpreter that will be destroyed last.
-=item C<void Parrot_init(Parrot_Interp)>
-
-Initializes a Parrot interpreter.
-
=item C<void Parrot_setflag(Parrot_Interp, Parrot_Interp_flag,
Parrot_Interp_flag_val)>
Sets a flag in the Parrot interpreter.
