3 perlembed - how to embed perl in your C program
13 =item B<Use C from Perl?>
15 Read L<perlxs>, L<perlxstut> and L<h2xs>.
17 =item B<Use a Unix program from Perl?>
19 Read about back-quotes and about C<system> and C<exec> in L<perlfunc>.
21 =item B<Use Perl from Perl?>
23 Read about do(), eval(), require(), and use() in L<perlfunc>.
25 =item B<Use C from C?>
29 =item B<Use Perl from C?>
37 Compiling your C program
39 There's one example in each of the nine sections:
45 Adding a Perl interpreter to your C program
49 Calling a Perl subroutine from your C program
53 Evaluating a Perl statement from your C program
57 Performing Perl pattern matches and substitutions from your C program
61 Fiddling with the Perl stack from your C program
65 Maintaining a persistent interpreter
69 Maintaining multiple interpreter instances
73 Using Perl modules, which themselves use C libraries, from your C program
77 Embedding Perl under Win32
81 =head2 Compiling your C program
83 If you have trouble compiling the scripts in this documentation,
84 you're not alone. The cardinal rule: COMPILE THE PROGRAMS IN EXACTLY
85 THE SAME WAY THAT YOUR PERL WAS COMPILED. (Sorry for yelling.)
87 Also, every C program that uses Perl must link in the I<perl library>.
88 What's that, you ask? Perl is itself written in C; the perl library
89 is the collection of compiled C programs that were used to create your
90 perl executable (I</usr/bin/perl> or equivalent). (Corollary: you
91 can't use Perl from your C program unless Perl has been compiled on
92 your machine, or installed properly--that's why you shouldn't blithely
93 copy Perl executables from machine to machine without also copying the
96 When you use Perl from C, your C program will--usually--allocate,
97 "run", and deallocate a I<PerlInterpreter> object, which is defined by
100 If your copy of Perl is recent enough to contain this documentation
101 (version 5.002 or later), then the perl library (and I<EXTERN.h> and
102 I<perl.h>, which you'll also need) will reside in a directory
103 that looks like this:
105 /usr/local/lib/perl5/your_architecture_here/CORE
109 /usr/local/lib/perl5/CORE
111 or maybe something like
115 Execute this statement for a hint about where to find CORE:
117 perl -MConfig -e 'print $Config{archlib}'
119 Here's how you'd compile the example in the next section,
120 Adding a Perl interpreter to your C program, on my Linux box:
122 % gcc -O2 -Dbool=char -DHAS_BOOL -I/usr/local/include
123 -I/usr/local/lib/perl5/i586-linux/5.003/CORE
124 -L/usr/local/lib/perl5/i586-linux/5.003/CORE
125 -o interp interp.c -lperl -lm
127 (That's all one line.) On my DEC Alpha running 5.003_05, the incantation
130 % cc -O2 -Olimit 2900 -DSTANDARD_C -I/usr/local/include
131 -I/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE
132 -L/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE -L/usr/local/lib
133 -D__LANGUAGE_C__ -D_NO_PROTO -o interp interp.c -lperl -lm
135 How can you figure out what to add? Assuming your Perl is post-5.001,
136 execute a C<perl -V> command and pay special attention to the "cc" and
137 "ccflags" information.
139 You'll have to choose the appropriate compiler (I<cc>, I<gcc>, et al.) for
140 your machine: C<perl -MConfig -e 'print $Config{cc}'> will tell you what
143 You'll also have to choose the appropriate library directory
144 (I</usr/local/lib/...>) for your machine. If your compiler complains
145 that certain functions are undefined, or that it can't locate
146 I<-lperl>, then you need to change the path following the C<-L>. If it
147 complains that it can't find I<EXTERN.h> and I<perl.h>, you need to
148 change the path following the C<-I>.
150 You may have to add extra libraries as well. Which ones?
151 Perhaps those printed by
153 perl -MConfig -e 'print $Config{libs}'
155 Provided your perl binary was properly configured and installed the
156 B<ExtUtils::Embed> module will determine all of this information for
159 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
161 If the B<ExtUtils::Embed> module isn't part of your Perl distribution,
162 you can retrieve it from
163 http://www.perl.com/perl/CPAN/modules/by-module/ExtUtils::Embed. (If
164 this documentation came from your Perl distribution, then you're
165 running 5.004 or better and you already have it.)
167 The B<ExtUtils::Embed> kit on CPAN also contains all source code for
168 the examples in this document, tests, additional examples and other
169 information you may find useful.
171 =head2 Adding a Perl interpreter to your C program
173 In a sense, perl (the C program) is a good example of embedding Perl
174 (the language), so I'll demonstrate embedding with I<miniperlmain.c>,
175 from the source distribution. Here's a bastardized, nonportable
176 version of I<miniperlmain.c> containing the essentials of embedding:
178 #include <EXTERN.h> /* from the Perl distribution */
179 #include <perl.h> /* from the Perl distribution */
181 static PerlInterpreter *my_perl; /*** The Perl interpreter ***/
183 int main(int argc, char **argv, char **env)
185 my_perl = perl_alloc();
186 perl_construct(my_perl);
187 perl_parse(my_perl, NULL, argc, argv, (char **)NULL);
189 perl_destruct(my_perl);
193 Notice that we don't use the C<env> pointer. Normally handed to
194 C<perl_parse> as its final argument, C<env> here is replaced by
195 C<NULL>, which means that the current environment will be used.
197 Now compile this program (I'll call it I<interp.c>) into an executable:
199 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
201 After a successful compilation, you'll be able to use I<interp> just
205 print "Pretty Good Perl \n";
206 print "10890 - 9801 is ", 10890 - 9801;
213 % interp -e 'printf("%x", 3735928559)'
216 You can also read and execute Perl statements from a file while in the
217 midst of your C program, by placing the filename in I<argv[1]> before
218 calling I<perl_run()>.
220 =head2 Calling a Perl subroutine from your C program
222 To call individual Perl subroutines, you can use any of the B<perl_call_*>
223 functions documented in L<perlcall>.
224 In this example we'll use perl_call_argv().
226 That's shown below, in a program I'll call I<showtime.c>.
231 static PerlInterpreter *my_perl;
233 int main(int argc, char **argv, char **env)
235 char *args[] = { NULL };
236 my_perl = perl_alloc();
237 perl_construct(my_perl);
239 perl_parse(my_perl, NULL, argc, argv, NULL);
241 /*** skipping perl_run() ***/
243 perl_call_argv("showtime", G_DISCARD | G_NOARGS, args);
245 perl_destruct(my_perl);
249 where I<showtime> is a Perl subroutine that takes no arguments (that's the
250 I<G_NOARGS>) and for which I'll ignore the return value (that's the
251 I<G_DISCARD>). Those flags, and others, are discussed in L<perlcall>.
253 I'll define the I<showtime> subroutine in a file called I<showtime.pl>:
255 print "I shan't be printed.";
261 Simple enough. Now compile and run:
263 % cc -o showtime showtime.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
265 % showtime showtime.pl
268 yielding the number of seconds that elapsed between January 1, 1970
269 (the beginning of the Unix epoch), and the moment I began writing this
272 In this particular case we don't have to call I<perl_run>, but in
273 general it's considered good practice to ensure proper initialization
274 of library code, including execution of all object C<DESTROY> methods
275 and package C<END {}> blocks.
277 If you want to pass arguments to the Perl subroutine, you can add
278 strings to the C<NULL>-terminated C<args> list passed to
279 I<perl_call_argv>. For other data types, or to examine return values,
280 you'll need to manipulate the Perl stack. That's demonstrated in the
281 last section of this document: Fiddling with the Perl stack from
284 =head2 Evaluating a Perl statement from your C program
286 Perl provides two API functions to evaluate pieces of Perl code.
287 These are perl_eval_sv() and perl_eval_pv().
289 Arguably, these are the only routines you'll ever need to execute
290 snippets of Perl code from within your C program. Your code can be
291 as long as you wish; it can contain multiple statements; it can employ
292 use(), require(), and do() to include external Perl files.
294 perl_eval_pv() lets us evaluate individual Perl strings, and then
295 extract variables for coercion into C types. The following program,
296 I<string.c>, executes three Perl strings, extracting an C<int> from
297 the first, a C<float> from the second, and a C<char *> from the third.
302 static PerlInterpreter *my_perl;
304 main (int argc, char **argv, char **env)
306 char *embedding[] = { "", "-e", "0" };
308 my_perl = perl_alloc();
309 perl_construct( my_perl );
311 perl_parse(my_perl, NULL, 3, embedding, NULL);
314 /** Treat $a as an integer **/
315 perl_eval_pv("$a = 3; $a **= 2", TRUE);
316 printf("a = %d\n", SvIV(perl_get_sv("a", FALSE)));
318 /** Treat $a as a float **/
319 perl_eval_pv("$a = 3.14; $a **= 2", TRUE);
320 printf("a = %f\n", SvNV(perl_get_sv("a", FALSE)));
322 /** Treat $a as a string **/
323 perl_eval_pv("$a = 'rekcaH lreP rehtonA tsuJ'; $a = reverse($a);", TRUE);
324 printf("a = %s\n", SvPV(perl_get_sv("a", FALSE), na));
326 perl_destruct(my_perl);
330 All of those strange functions with I<sv> in their names help convert Perl scalars to C types. They're described in L<perlguts>.
332 If you compile and run I<string.c>, you'll see the results of using
333 I<SvIV()> to create an C<int>, I<SvNV()> to create a C<float>, and
334 I<SvPV()> to create a string:
338 a = Just Another Perl Hacker
340 In the example above, we've created a global variable to temporarily
341 store the computed value of our eval'd expression. It is also
342 possible and in most cases a better strategy to fetch the return value
343 from perl_eval_pv() instead. Example:
346 SV *val = perl_eval_pv("reverse 'rekcaH lreP rehtonA tsuJ'", TRUE);
347 printf("%s\n", SvPV(val,na));
350 This way, we avoid namespace pollution by not creating global
351 variables and we've simplified our code as well.
353 =head2 Performing Perl pattern matches and substitutions from your C program
355 The I<perl_eval_sv()> function lets us evaluate chunks of Perl code, so we can
356 define some functions that use it to "specialize" in matches and
357 substitutions: I<match()>, I<substitute()>, and I<matches()>.
359 char match(SV *string, char *pattern);
361 Given a string and a pattern (e.g., C<m/clasp/> or C</\b\w*\b/>, which
362 in your C program might appear as "/\\b\\w*\\b/"), match()
363 returns 1 if the string matches the pattern and 0 otherwise.
365 int substitute(SV **string, char *pattern);
367 Given a pointer to an C<SV> and an C<=~> operation (e.g.,
368 C<s/bob/robert/g> or C<tr[A-Z][a-z]>), substitute() modifies the string
369 within the C<AV> at according to the operation, returning the number of substitutions
372 int matches(SV *string, char *pattern, AV **matches);
374 Given an C<SV>, a pattern, and a pointer to an empty C<AV>,
375 matches() evaluates C<$string =~ $pattern> in an array context, and
376 fills in I<matches> with the array elements, returning the number of matches found.
378 Here's a sample program, I<match.c>, that uses all three (long lines have
384 /** my_perl_eval_sv(code, error_check)
385 ** kinda like perl_eval_sv(),
386 ** but we pop the return value off the stack
388 SV* my_perl_eval_sv(SV *sv, I32 croak_on_error)
394 perl_eval_sv(sv, G_SCALAR);
400 if (croak_on_error && SvTRUE(GvSV(errgv)))
401 croak(SvPVx(GvSV(errgv), na));
406 /** match(string, pattern)
408 ** Used for matches in a scalar context.
410 ** Returns 1 if the match was successful; 0 otherwise.
413 I32 match(SV *string, char *pattern)
415 SV *command = NEWSV(1099, 0), *retval;
417 sv_setpvf(command, "my $string = '%s'; $string =~ %s",
418 SvPV(string,na), pattern);
420 retval = my_perl_eval_sv(command, TRUE);
421 SvREFCNT_dec(command);
426 /** substitute(string, pattern)
428 ** Used for =~ operations that modify their left-hand side (s/// and tr///)
430 ** Returns the number of successful matches, and
431 ** modifies the input string if there were any.
434 I32 substitute(SV **string, char *pattern)
436 SV *command = NEWSV(1099, 0), *retval;
438 sv_setpvf(command, "$string = '%s'; ($string =~ %s)",
439 SvPV(*string,na), pattern);
441 retval = my_perl_eval_sv(command, TRUE);
442 SvREFCNT_dec(command);
444 *string = perl_get_sv("string", FALSE);
448 /** matches(string, pattern, matches)
450 ** Used for matches in an array context.
452 ** Returns the number of matches,
453 ** and fills in **matches with the matching substrings
456 I32 matches(SV *string, char *pattern, AV **match_list)
458 SV *command = NEWSV(1099, 0);
461 sv_setpvf(command, "my $string = '%s'; @array = ($string =~ %s)",
462 SvPV(string,na), pattern);
464 my_perl_eval_sv(command, TRUE);
465 SvREFCNT_dec(command);
467 *match_list = perl_get_av("array", FALSE);
468 num_matches = av_len(*match_list) + 1; /** assume $[ is 0 **/
473 main (int argc, char **argv, char **env)
475 PerlInterpreter *my_perl = perl_alloc();
476 char *embedding[] = { "", "-e", "0" };
479 SV *text = NEWSV(1099,0);
481 perl_construct(my_perl);
482 perl_parse(my_perl, NULL, 3, embedding, NULL);
484 sv_setpv(text, "When he is at a convenience store and the bill comes to some amount like 76 cents, Maynard is aware that there is something he *should* do, something that will enable him to get back a quarter, but he has no idea *what*. He fumbles through his red squeezey changepurse and gives the boy three extra pennies with his dollar, hoping that he might luck into the correct amount. The boy gives him back two of his own pennies and then the big shiny quarter that is his prize. -RICHH");
486 if (match(text, "m/quarter/")) /** Does text contain 'quarter'? **/
487 printf("match: Text contains the word 'quarter'.\n\n");
489 printf("match: Text doesn't contain the word 'quarter'.\n\n");
491 if (match(text, "m/eighth/")) /** Does text contain 'eighth'? **/
492 printf("match: Text contains the word 'eighth'.\n\n");
494 printf("match: Text doesn't contain the word 'eighth'.\n\n");
496 /** Match all occurrences of /wi../ **/
497 num_matches = matches(text, "m/(wi..)/g", &match_list);
498 printf("matches: m/(wi..)/g found %d matches...\n", num_matches);
500 for (i = 0; i < num_matches; i++)
501 printf("match: %s\n", SvPV(*av_fetch(match_list, i, FALSE),na));
504 /** Remove all vowels from text **/
505 num_matches = substitute(&text, "s/[aeiou]//gi");
507 printf("substitute: s/[aeiou]//gi...%d substitutions made.\n",
509 printf("Now text is: %s\n\n", SvPV(text,na));
512 /** Attempt a substitution **/
513 if (!substitute(&text, "s/Perl/C/")) {
514 printf("substitute: s/Perl/C...No substitution made.\n\n");
518 perl_destruct_level = 1;
519 perl_destruct(my_perl);
523 which produces the output (again, long lines have been wrapped here)
525 match: Text contains the word 'quarter'.
527 match: Text doesn't contain the word 'eighth'.
529 matches: m/(wi..)/g found 2 matches...
533 substitute: s/[aeiou]//gi...139 substitutions made.
534 Now text is: Whn h s t cnvnnc str nd th bll cms t sm mnt lk 76 cnts,
535 Mynrd s wr tht thr s smthng h *shld* d, smthng tht wll nbl hm t gt bck
536 qrtr, bt h hs n d *wht*. H fmbls thrgh hs rd sqzy chngprs nd gvs th by
537 thr xtr pnns wth hs dllr, hpng tht h mght lck nt th crrct mnt. Th by gvs
538 hm bck tw f hs wn pnns nd thn th bg shny qrtr tht s hs prz. -RCHH
540 substitute: s/Perl/C...No substitution made.
542 =head2 Fiddling with the Perl stack from your C program
544 When trying to explain stacks, most computer science textbooks mumble
545 something about spring-loaded columns of cafeteria plates: the last
546 thing you pushed on the stack is the first thing you pop off. That'll
547 do for our purposes: your C program will push some arguments onto "the Perl
548 stack", shut its eyes while some magic happens, and then pop the
549 results--the return value of your Perl subroutine--off the stack.
551 First you'll need to know how to convert between C types and Perl
552 types, with newSViv() and sv_setnv() and newAV() and all their
553 friends. They're described in L<perlguts>.
555 Then you'll need to know how to manipulate the Perl stack. That's
556 described in L<perlcall>.
558 Once you've understood those, embedding Perl in C is easy.
560 Because C has no builtin function for integer exponentiation, let's
561 make Perl's ** operator available to it (this is less useful than it
562 sounds, because Perl implements ** with C's I<pow()> function). First
563 I'll create a stub exponentiation function in I<power.pl>:
570 Now I'll create a C program, I<power.c>, with a function
571 I<PerlPower()> that contains all the perlguts necessary to push the
572 two arguments into I<expo()> and to pop the return value out. Take a
578 static PerlInterpreter *my_perl;
581 PerlPower(int a, int b)
583 dSP; /* initialize stack pointer */
584 ENTER; /* everything created after here */
585 SAVETMPS; /* ...is a temporary variable. */
586 PUSHMARK(SP); /* remember the stack pointer */
587 XPUSHs(sv_2mortal(newSViv(a))); /* push the base onto the stack */
588 XPUSHs(sv_2mortal(newSViv(b))); /* push the exponent onto stack */
589 PUTBACK; /* make local stack pointer global */
590 perl_call_pv("expo", G_SCALAR); /* call the function */
591 SPAGAIN; /* refresh stack pointer */
592 /* pop the return value from stack */
593 printf ("%d to the %dth power is %d.\n", a, b, POPi);
595 FREETMPS; /* free that return value */
596 LEAVE; /* ...and the XPUSHed "mortal" args.*/
599 int main (int argc, char **argv, char **env)
601 char *my_argv[] = { "", "power.pl" };
603 my_perl = perl_alloc();
604 perl_construct( my_perl );
606 perl_parse(my_perl, NULL, 2, my_argv, (char **)NULL);
609 PerlPower(3, 4); /*** Compute 3 ** 4 ***/
611 perl_destruct(my_perl);
619 % cc -o power power.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
622 3 to the 4th power is 81.
624 =head2 Maintaining a persistent interpreter
626 When developing interactive and/or potentially long-running
627 applications, it's a good idea to maintain a persistent interpreter
628 rather than allocating and constructing a new interpreter multiple
629 times. The major reason is speed: since Perl will only be loaded into
632 However, you have to be more cautious with namespace and variable
633 scoping when using a persistent interpreter. In previous examples
634 we've been using global variables in the default package C<main>. We
635 knew exactly what code would be run, and assumed we could avoid
636 variable collisions and outrageous symbol table growth.
638 Let's say your application is a server that will occasionally run Perl
639 code from some arbitrary file. Your server has no way of knowing what
640 code it's going to run. Very dangerous.
642 If the file is pulled in by C<perl_parse()>, compiled into a newly
643 constructed interpreter, and subsequently cleaned out with
644 C<perl_destruct()> afterwards, you're shielded from most namespace
647 One way to avoid namespace collisions in this scenario is to translate
648 the filename into a guaranteed-unique package name, and then compile
649 the code into that package using eval(). In the example
650 below, each file will only be compiled once. Or, the application
651 might choose to clean out the symbol table associated with the file
652 after it's no longer needed. Using perl_call_argv(), We'll
653 call the subroutine C<Embed::Persistent::eval_file> which lives in the
654 file C<persistent.pl> and pass the filename and boolean cleanup/cache
657 Note that the process will continue to grow for each file that it
658 uses. In addition, there might be C<AUTOLOAD>ed subroutines and other
659 conditions that cause Perl's symbol table to grow. You might want to
660 add some logic that keeps track of the process size, or restarts
661 itself after a certain number of requests, to ensure that memory
662 consumption is minimized. You'll also want to scope your variables
663 with my() whenever possible.
666 package Embed::Persistent;
672 sub valid_package_name {
674 $string =~ s/([^A-Za-z0-9\/])/sprintf("_%2x",unpack("C",$1))/eg;
675 # second pass only for words starting with a digit
676 $string =~ s|/(\d)|sprintf("/_%2x",unpack("C",$1))|eg;
678 # Dress it up as a real package name
680 return "Embed" . $string;
683 #borrowed from Safe.pm
689 $pkg = "main::$pkg\::"; # expand to full symbol table name
690 ($stem, $leaf) = $pkg =~ m/(.*::)(\w+::)$/;
692 my $stem_symtab = *{$stem}{HASH};
694 delete $stem_symtab->{$leaf};
698 my($filename, $delete) = @_;
699 my $package = valid_package_name($filename);
700 my $mtime = -M $filename;
701 if(defined $Cache{$package}{mtime}
703 $Cache{$package}{mtime} <= $mtime)
705 # we have compiled this subroutine already,
706 # it has not been updated on disk, nothing left to do
707 print STDERR "already compiled $package->handler\n";
711 open FH, $filename or die "open '$filename' $!";
716 #wrap the code into a subroutine inside our unique package
717 my $eval = qq{package $package; sub handler { $sub; }};
719 # hide our variables within this block
720 my($filename,$mtime,$package,$sub);
725 #cache it unless we're cleaning out each time
726 $Cache{$package}{mtime} = $mtime unless $delete;
729 eval {$package->handler;};
732 delete_package($package) if $delete;
734 #take a look if you want
735 #print Devel::Symdump->rnew($package)->as_string, $/;
746 /* 1 = clean out filename's symbol table after each request, 0 = don't */
751 static PerlInterpreter *perl = NULL;
754 main(int argc, char **argv, char **env)
756 char *embedding[] = { "", "persistent.pl" };
757 char *args[] = { "", DO_CLEAN, NULL };
758 char filename [1024];
761 if((perl = perl_alloc()) == NULL) {
762 fprintf(stderr, "no memory!");
765 perl_construct(perl);
767 exitstatus = perl_parse(perl, NULL, 2, embedding, NULL);
770 exitstatus = perl_run(perl);
772 while(printf("Enter file name: ") && gets(filename)) {
774 /* call the subroutine, passing it the filename as an argument */
776 perl_call_argv("Embed::Persistent::eval_file",
777 G_DISCARD | G_EVAL, args);
780 if(SvTRUE(GvSV(errgv)))
781 fprintf(stderr, "eval error: %s\n", SvPV(GvSV(errgv),na));
785 perl_destruct_level = 0;
793 % cc -o persistent persistent.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
795 Here's a example script file:
798 my $string = "hello";
802 print "foo says: @_\n";
808 Enter file name: test.pl
810 Enter file name: test.pl
811 already compiled Embed::test_2epl->handler
815 =head2 Maintaining multiple interpreter instances
817 Some rare applications will need to create more than one interpreter
818 during a session. Such an application might sporadically decide to
819 release any resources associated with the interpreter.
821 The program must take care to ensure that this takes place I<before>
822 the next interpreter is constructed. By default, the global variable
823 C<perl_destruct_level> is set to C<0>, since extra cleaning isn't
824 needed when a program has only one interpreter.
826 Setting C<perl_destruct_level> to C<1> makes everything squeaky clean:
828 perl_destruct_level = 1;
832 /* reset global variables here with perl_destruct_level = 1 */
833 perl_construct(my_perl);
835 /* clean and reset _everything_ during perl_destruct */
836 perl_destruct(my_perl);
839 /* let's go do it again! */
842 When I<perl_destruct()> is called, the interpreter's syntax parse tree
843 and symbol tables are cleaned up, and global variables are reset.
845 Now suppose we have more than one interpreter instance running at the
846 same time. This is feasible, but only if you used the
847 C<-DMULTIPLICITY> flag when building Perl. By default, that sets
848 C<perl_destruct_level> to C<1>.
856 /* we're going to embed two interpreters */
857 /* we're going to embed two interpreters */
859 #define SAY_HELLO "-e", "print qq(Hi, I'm $^X\n)"
861 int main(int argc, char **argv, char **env)
864 *one_perl = perl_alloc(),
865 *two_perl = perl_alloc();
866 char *one_args[] = { "one_perl", SAY_HELLO };
867 char *two_args[] = { "two_perl", SAY_HELLO };
869 perl_construct(one_perl);
870 perl_construct(two_perl);
872 perl_parse(one_perl, NULL, 3, one_args, (char **)NULL);
873 perl_parse(two_perl, NULL, 3, two_args, (char **)NULL);
878 perl_destruct(one_perl);
879 perl_destruct(two_perl);
888 % cc -o multiplicity multiplicity.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
896 =head2 Using Perl modules, which themselves use C libraries, from your C program
898 If you've played with the examples above and tried to embed a script
899 that I<use()>s a Perl module (such as I<Socket>) which itself uses a C or C++ library,
900 this probably happened:
903 Can't load module Socket, dynamic loading not available in this perl.
904 (You may need to build a new perl executable which either supports
905 dynamic loading or has the Socket module statically linked into it.)
910 Your interpreter doesn't know how to communicate with these extensions
911 on its own. A little glue will help. Up until now you've been
912 calling I<perl_parse()>, handing it NULL for the second argument:
914 perl_parse(my_perl, NULL, argc, my_argv, NULL);
916 That's where the glue code can be inserted to create the initial contact between
917 Perl and linked C/C++ routines. Let's take a look some pieces of I<perlmain.c>
918 to see how Perl does this:
922 # define EXTERN_C extern "C"
924 # define EXTERN_C extern
927 static void xs_init _((void));
929 EXTERN_C void boot_DynaLoader _((CV* cv));
930 EXTERN_C void boot_Socket _((CV* cv));
936 char *file = __FILE__;
937 /* DynaLoader is a special case */
938 newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
939 newXS("Socket::bootstrap", boot_Socket, file);
942 Simply put: for each extension linked with your Perl executable
943 (determined during its initial configuration on your
944 computer or when adding a new extension),
945 a Perl subroutine is created to incorporate the extension's
946 routines. Normally, that subroutine is named
947 I<Module::bootstrap()> and is invoked when you say I<use Module>. In
948 turn, this hooks into an XSUB, I<boot_Module>, which creates a Perl
949 counterpart for each of the extension's XSUBs. Don't worry about this
950 part; leave that to the I<xsubpp> and extension authors. If your
951 extension is dynamically loaded, DynaLoader creates I<Module::bootstrap()>
952 for you on the fly. In fact, if you have a working DynaLoader then there
953 is rarely any need to link in any other extensions statically.
956 Once you have this code, slap it into the second argument of I<perl_parse()>:
959 perl_parse(my_perl, xs_init, argc, my_argv, NULL);
964 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
968 use SomeDynamicallyLoadedModule;
970 print "Now I can use extensions!\n"'
972 B<ExtUtils::Embed> can also automate writing the I<xs_init> glue code.
974 % perl -MExtUtils::Embed -e xsinit -- -o perlxsi.c
975 % cc -c perlxsi.c `perl -MExtUtils::Embed -e ccopts`
976 % cc -c interp.c `perl -MExtUtils::Embed -e ccopts`
977 % cc -o interp perlxsi.o interp.o `perl -MExtUtils::Embed -e ldopts`
979 Consult L<perlxs> and L<perlguts> for more details.
981 =head1 Embedding Perl under Win32
983 At the time of this writing, there are two versions of Perl which run
984 under Win32. Interfacing to Activeware's Perl library is quite
985 different from the examples in this documentation, as significant
986 changes were made to the internal Perl API. However, it is possible
987 to embed Activeware's Perl runtime, see the Perl for Win32 FAQ:
988 http://www.perl.com/perl/faq/win32/Perl_for_Win32_FAQ.html
990 With the "official" Perl version 5.004 or higher, all the examples
991 within this documentation will compile and run untouched, although,
992 the build process is slightly different between Unix and Win32.
994 For starters, backticks don't work under the Win32 native command shell!
995 The ExtUtils::Embed kit on CPAN ships with a script called
996 B<genmake>, which generates a simple makefile to build a program from
997 a single C source file. It can be used like so:
999 C:\ExtUtils-Embed\eg> perl genmake interp.c
1000 C:\ExtUtils-Embed\eg> nmake
1001 C:\ExtUtils-Embed\eg> interp -e "print qq{I'm embedded in Win32!\n}"
1003 You may wish to use a more robust environment such as the MS Developer
1004 stdio. In this case, to generate perlxsi.c run:
1006 perl -MExtUtils::Embed -e xsinit
1008 Create a new project, Insert -> Files into Project: perlxsi.c, perl.lib,
1009 and your own source files, e.g. interp.c. Typically you'll find
1010 perl.lib in B<C:\perl\lib\CORE>, if not, you should see the B<CORE>
1011 directory relative to C<perl -V:archlib>.
1012 The studio will also need this path so it knows where to find Perl
1013 include files. This path can be added via the Tools -> Options ->
1014 Directories menu. Finnally, select Build -> Build interp.exe and
1019 You can sometimes I<write faster code> in C, but
1020 you can always I<write code faster> in Perl. Because you can use
1021 each from the other, combine them as you wish.
1026 Jon Orwant and <F<orwant@tpj.com>> and Doug MacEachern <F<dougm@osf.org>>,
1027 with small contributions from Tim Bunce, Tom Christiansen, Hallvard Furuseth,
1028 Dov Grobgeld, and Ilya Zakharevich.
1030 Check out Doug's article on embedding in Volume 1, Issue 4 of The Perl
1031 Journal. Info about TPJ is available from http://tpj.com.
1035 Some of this material is excerpted from Jon Orwant's book: I<Perl 5
1036 Interactive>, Waite Group Press, 1996 (ISBN 1-57169-064-6) and appears
1037 courtesy of Waite Group Press.
1041 Copyright (C) 1995, 1996, 1997 Doug MacEachern and Jon Orwant. All
1044 Although destined for release with the standard Perl distribution,
1045 this document is not public domain, nor is any of Perl and its
1046 documentation. Permission is granted to freely distribute verbatim
1047 copies of this document provided that no modifications outside of
1048 formatting be made, and that this notice remain intact. You are
1049 permitted and encouraged to use its code and derivatives thereof in
1050 your own source code for fun or for profit as you see fit.