3 perlembed - how to embed perl in your C program
13 =item B<Use C from Perl?>
15 Read L<perlxstut>, L<perlxs>, L<h2xs>, and L<perlguts>.
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 L<perlfunc/do> and L<perlfunc/eval> and L<perlfunc/require>
26 =item B<Use C from C?>
30 =item B<Use Perl from C?>
38 L<Compiling your C program>
40 L<Adding a Perl interpreter to your C program>
42 L<Calling a Perl subroutine from your C program>
44 L<Evaluating a Perl statement from your C program>
46 L<Performing Perl pattern matches and substitutions from your C program>
48 L<Fiddling with the Perl stack from your C program>
50 L<Maintaining a persistent interpreter>
52 L<Maintaining multiple interpreter instances>
54 L<Using Perl modules, which themselves use C libraries, from your C program>
56 L<Embedding Perl under Win32>
60 =head2 Compiling your C program
62 If you have trouble compiling the scripts in this documentation,
63 you're not alone. The cardinal rule: COMPILE THE PROGRAMS IN EXACTLY
64 THE SAME WAY THAT YOUR PERL WAS COMPILED. (Sorry for yelling.)
66 Also, every C program that uses Perl must link in the I<perl library>.
67 What's that, you ask? Perl is itself written in C; the perl library
68 is the collection of compiled C programs that were used to create your
69 perl executable (I</usr/bin/perl> or equivalent). (Corollary: you
70 can't use Perl from your C program unless Perl has been compiled on
71 your machine, or installed properly--that's why you shouldn't blithely
72 copy Perl executables from machine to machine without also copying the
75 When you use Perl from C, your C program will--usually--allocate,
76 "run", and deallocate a I<PerlInterpreter> object, which is defined by
79 If your copy of Perl is recent enough to contain this documentation
80 (version 5.002 or later), then the perl library (and I<EXTERN.h> and
81 I<perl.h>, which you'll also need) will reside in a directory
84 /usr/local/lib/perl5/your_architecture_here/CORE
88 /usr/local/lib/perl5/CORE
90 or maybe something like
94 Execute this statement for a hint about where to find CORE:
96 perl -MConfig -e 'print $Config{archlib}'
98 Here's how you'd compile the example in the next section,
99 L<Adding a Perl interpreter to your C program>, on my Linux box:
101 % gcc -O2 -Dbool=char -DHAS_BOOL -I/usr/local/include
102 -I/usr/local/lib/perl5/i586-linux/5.003/CORE
103 -L/usr/local/lib/perl5/i586-linux/5.003/CORE
104 -o interp interp.c -lperl -lm
106 (That's all one line.) On my DEC Alpha running old 5.003_05, the
107 incantation is a bit different:
109 % cc -O2 -Olimit 2900 -DSTANDARD_C -I/usr/local/include
110 -I/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE
111 -L/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE -L/usr/local/lib
112 -D__LANGUAGE_C__ -D_NO_PROTO -o interp interp.c -lperl -lm
114 How can you figure out what to add? Assuming your Perl is post-5.001,
115 execute a C<perl -V> command and pay special attention to the "cc" and
116 "ccflags" information.
118 You'll have to choose the appropriate compiler (I<cc>, I<gcc>, et al.) for
119 your machine: C<perl -MConfig -e 'print $Config{cc}'> will tell you what
122 You'll also have to choose the appropriate library directory
123 (I</usr/local/lib/...>) for your machine. If your compiler complains
124 that certain functions are undefined, or that it can't locate
125 I<-lperl>, then you need to change the path following the C<-L>. If it
126 complains that it can't find I<EXTERN.h> and I<perl.h>, you need to
127 change the path following the C<-I>.
129 You may have to add extra libraries as well. Which ones?
130 Perhaps those printed by
132 perl -MConfig -e 'print $Config{libs}'
134 Provided your perl binary was properly configured and installed the
135 B<ExtUtils::Embed> module will determine all of this information for
138 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
140 If the B<ExtUtils::Embed> module isn't part of your Perl distribution,
141 you can retrieve it from
142 http://www.perl.com/perl/CPAN/modules/by-module/ExtUtils::Embed. (If
143 this documentation came from your Perl distribution, then you're
144 running 5.004 or better and you already have it.)
146 The B<ExtUtils::Embed> kit on CPAN also contains all source code for
147 the examples in this document, tests, additional examples and other
148 information you may find useful.
150 =head2 Adding a Perl interpreter to your C program
152 In a sense, perl (the C program) is a good example of embedding Perl
153 (the language), so I'll demonstrate embedding with I<miniperlmain.c>,
154 included in the source distribution. Here's a bastardized, nonportable
155 version of I<miniperlmain.c> containing the essentials of embedding:
157 #include <EXTERN.h> /* from the Perl distribution */
158 #include <perl.h> /* from the Perl distribution */
160 static PerlInterpreter *my_perl; /*** The Perl interpreter ***/
162 int main(int argc, char **argv, char **env)
164 my_perl = perl_alloc();
165 perl_construct(my_perl);
166 perl_parse(my_perl, NULL, argc, argv, (char **)NULL);
168 perl_destruct(my_perl);
172 Notice that we don't use the C<env> pointer. Normally handed to
173 C<perl_parse> as its final argument, C<env> here is replaced by
174 C<NULL>, which means that the current environment will be used.
176 Now compile this program (I'll call it I<interp.c>) into an executable:
178 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
180 After a successful compilation, you'll be able to use I<interp> just
184 print "Pretty Good Perl \n";
185 print "10890 - 9801 is ", 10890 - 9801;
192 % interp -e 'printf("%x", 3735928559)'
195 You can also read and execute Perl statements from a file while in the
196 midst of your C program, by placing the filename in I<argv[1]> before
199 =head2 Calling a Perl subroutine from your C program
201 To call individual Perl subroutines, you can use any of the B<perl_call_*>
202 functions documented in L<perlcall>.
203 In this example we'll use C<perl_call_argv>.
205 That's shown below, in a program I'll call I<showtime.c>.
210 static PerlInterpreter *my_perl;
212 int main(int argc, char **argv, char **env)
214 char *args[] = { NULL };
215 my_perl = perl_alloc();
216 perl_construct(my_perl);
218 perl_parse(my_perl, NULL, argc, argv, NULL);
220 /*** skipping perl_run() ***/
222 perl_call_argv("showtime", G_DISCARD | G_NOARGS, args);
224 perl_destruct(my_perl);
228 where I<showtime> is a Perl subroutine that takes no arguments (that's the
229 I<G_NOARGS>) and for which I'll ignore the return value (that's the
230 I<G_DISCARD>). Those flags, and others, are discussed in L<perlcall>.
232 I'll define the I<showtime> subroutine in a file called I<showtime.pl>:
234 print "I shan't be printed.";
240 Simple enough. Now compile and run:
242 % cc -o showtime showtime.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
244 % showtime showtime.pl
247 yielding the number of seconds that elapsed between January 1, 1970
248 (the beginning of the Unix epoch), and the moment I began writing this
251 In this particular case we don't have to call I<perl_run>, but in
252 general it's considered good practice to ensure proper initialization
253 of library code, including execution of all object C<DESTROY> methods
254 and package C<END {}> blocks.
256 If you want to pass arguments to the Perl subroutine, you can add
257 strings to the C<NULL>-terminated C<args> list passed to
258 I<perl_call_argv>. For other data types, or to examine return values,
259 you'll need to manipulate the Perl stack. That's demonstrated in the
260 last section of this document: L<Fiddling with the Perl stack from
263 =head2 Evaluating a Perl statement from your C program
265 Perl provides two API functions to evaluate pieces of Perl code.
266 These are L<perlguts/perl_eval_sv> and L<perlguts/perl_eval_pv>.
268 Arguably, these are the only routines you'll ever need to execute
269 snippets of Perl code from within your C program. Your code can be as
270 long as you wish; it can contain multiple statements; it can employ
271 L<perlfunc/use>, L<perlfunc/require>, and L<perlfunc/do> to
272 include external Perl files.
274 I<perl_eval_pv> lets us evaluate individual Perl strings, and then
275 extract variables for coercion into C types. The following program,
276 I<string.c>, executes three Perl strings, extracting an C<int> from
277 the first, a C<float> from the second, and a C<char *> from the third.
282 static PerlInterpreter *my_perl;
284 main (int argc, char **argv, char **env)
286 char *embedding[] = { "", "-e", "0" };
288 my_perl = perl_alloc();
289 perl_construct( my_perl );
291 perl_parse(my_perl, NULL, 3, embedding, NULL);
294 /** Treat $a as an integer **/
295 perl_eval_pv("$a = 3; $a **= 2", TRUE);
296 printf("a = %d\n", SvIV(perl_get_sv("a", FALSE)));
298 /** Treat $a as a float **/
299 perl_eval_pv("$a = 3.14; $a **= 2", TRUE);
300 printf("a = %f\n", SvNV(perl_get_sv("a", FALSE)));
302 /** Treat $a as a string **/
303 perl_eval_pv("$a = 'rekcaH lreP rehtonA tsuJ'; $a = reverse($a);", TRUE);
304 printf("a = %s\n", SvPV(perl_get_sv("a", FALSE), na));
306 perl_destruct(my_perl);
310 All of those strange functions with I<sv> in their names help convert Perl scalars to C types. They're described in L<perlguts>.
312 If you compile and run I<string.c>, you'll see the results of using
313 I<SvIV()> to create an C<int>, I<SvNV()> to create a C<float>, and
314 I<SvPV()> to create a string:
318 a = Just Another Perl Hacker
320 In the example above, we've created a global variable to temporarily
321 store the computed value of our eval'd expression. It is also
322 possible and in most cases a better strategy to fetch the return value
323 from I<perl_eval_pv()> instead. Example:
326 SV *val = perl_eval_pv("reverse 'rekcaH lreP rehtonA tsuJ'", TRUE);
327 printf("%s\n", SvPV(val,na));
330 This way, we avoid namespace pollution by not creating global
331 variables and we've simplified our code as well.
333 =head2 Performing Perl pattern matches and substitutions from your C program
335 The I<perl_eval_sv()> function lets us evaluate strings of Perl code, so we can
336 define some functions that use it to "specialize" in matches and
337 substitutions: I<match()>, I<substitute()>, and I<matches()>.
339 I32 match(SV *string, char *pattern);
341 Given a string and a pattern (e.g., C<m/clasp/> or C</\b\w*\b/>, which
342 in your C program might appear as "/\\b\\w*\\b/"), match()
343 returns 1 if the string matches the pattern and 0 otherwise.
345 int substitute(SV **string, char *pattern);
347 Given a pointer to an C<SV> and an C<=~> operation (e.g.,
348 C<s/bob/robert/g> or C<tr[A-Z][a-z]>), substitute() modifies the string
349 within the C<AV> at according to the operation, returning the number of substitutions
352 int matches(SV *string, char *pattern, AV **matches);
354 Given an C<SV>, a pattern, and a pointer to an empty C<AV>,
355 matches() evaluates C<$string =~ $pattern> in an array context, and
356 fills in I<matches> with the array elements, returning the number of matches found.
358 Here's a sample program, I<match.c>, that uses all three (long lines have
364 /** my_perl_eval_sv(code, error_check)
365 ** kinda like perl_eval_sv(),
366 ** but we pop the return value off the stack
368 SV* my_perl_eval_sv(SV *sv, I32 croak_on_error)
374 perl_eval_sv(sv, G_SCALAR);
380 if (croak_on_error && SvTRUE(GvSV(errgv)))
381 croak(SvPVx(GvSV(errgv), na));
386 /** match(string, pattern)
388 ** Used for matches in a scalar context.
390 ** Returns 1 if the match was successful; 0 otherwise.
393 I32 match(SV *string, char *pattern)
395 SV *command = NEWSV(1099, 0), *retval;
397 sv_setpvf(command, "my $string = '%s'; $string =~ %s",
398 SvPV(string,na), pattern);
400 retval = my_perl_eval_sv(command, TRUE);
401 SvREFCNT_dec(command);
406 /** substitute(string, pattern)
408 ** Used for =~ operations that modify their left-hand side (s/// and tr///)
410 ** Returns the number of successful matches, and
411 ** modifies the input string if there were any.
414 I32 substitute(SV **string, char *pattern)
416 SV *command = NEWSV(1099, 0), *retval;
418 sv_setpvf(command, "$string = '%s'; ($string =~ %s)",
419 SvPV(*string,na), pattern);
421 retval = my_perl_eval_sv(command, TRUE);
422 SvREFCNT_dec(command);
424 *string = perl_get_sv("string", FALSE);
428 /** matches(string, pattern, matches)
430 ** Used for matches in an array context.
432 ** Returns the number of matches,
433 ** and fills in **matches with the matching substrings
436 I32 matches(SV *string, char *pattern, AV **match_list)
438 SV *command = NEWSV(1099, 0);
441 sv_setpvf(command, "my $string = '%s'; @array = ($string =~ %s)",
442 SvPV(string,na), pattern);
444 my_perl_eval_sv(command, TRUE);
445 SvREFCNT_dec(command);
447 *match_list = perl_get_av("array", FALSE);
448 num_matches = av_len(*match_list) + 1; /** assume $[ is 0 **/
453 main (int argc, char **argv, char **env)
455 PerlInterpreter *my_perl = perl_alloc();
456 char *embedding[] = { "", "-e", "0" };
459 SV *text = NEWSV(1099,0);
461 perl_construct(my_perl);
462 perl_parse(my_perl, NULL, 3, embedding, NULL);
464 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");
466 if (match(text, "m/quarter/")) /** Does text contain 'quarter'? **/
467 printf("match: Text contains the word 'quarter'.\n\n");
469 printf("match: Text doesn't contain the word 'quarter'.\n\n");
471 if (match(text, "m/eighth/")) /** Does text contain 'eighth'? **/
472 printf("match: Text contains the word 'eighth'.\n\n");
474 printf("match: Text doesn't contain the word 'eighth'.\n\n");
476 /** Match all occurrences of /wi../ **/
477 num_matches = matches(text, "m/(wi..)/g", &match_list);
478 printf("matches: m/(wi..)/g found %d matches...\n", num_matches);
480 for (i = 0; i < num_matches; i++)
481 printf("match: %s\n", SvPV(*av_fetch(match_list, i, FALSE),na));
484 /** Remove all vowels from text **/
485 num_matches = substitute(&text, "s/[aeiou]//gi");
487 printf("substitute: s/[aeiou]//gi...%d substitutions made.\n",
489 printf("Now text is: %s\n\n", SvPV(text,na));
492 /** Attempt a substitution **/
493 if (!substitute(&text, "s/Perl/C/")) {
494 printf("substitute: s/Perl/C...No substitution made.\n\n");
498 perl_destruct_level = 1;
499 perl_destruct(my_perl);
503 which produces the output (again, long lines have been wrapped here)
505 match: Text contains the word 'quarter'.
507 match: Text doesn't contain the word 'eighth'.
509 matches: m/(wi..)/g found 2 matches...
513 substitute: s/[aeiou]//gi...139 substitutions made.
514 Now text is: Whn h s t cnvnnc str nd th bll cms t sm mnt lk 76 cnts,
515 Mynrd s wr tht thr s smthng h *shld* d, smthng tht wll nbl hm t gt bck
516 qrtr, bt h hs n d *wht*. H fmbls thrgh hs rd sqzy chngprs nd gvs th by
517 thr xtr pnns wth hs dllr, hpng tht h mght lck nt th crrct mnt. Th by gvs
518 hm bck tw f hs wn pnns nd thn th bg shny qrtr tht s hs prz. -RCHH
520 substitute: s/Perl/C...No substitution made.
522 =head2 Fiddling with the Perl stack from your C program
524 When trying to explain stacks, most computer science textbooks mumble
525 something about spring-loaded columns of cafeteria plates: the last
526 thing you pushed on the stack is the first thing you pop off. That'll
527 do for our purposes: your C program will push some arguments onto "the Perl
528 stack", shut its eyes while some magic happens, and then pop the
529 results--the return value of your Perl subroutine--off the stack.
531 First you'll need to know how to convert between C types and Perl
532 types, with newSViv() and sv_setnv() and newAV() and all their
533 friends. They're described in L<perlguts>.
535 Then you'll need to know how to manipulate the Perl stack. That's
536 described in L<perlcall>.
538 Once you've understood those, embedding Perl in C is easy.
540 Because C has no builtin function for integer exponentiation, let's
541 make Perl's ** operator available to it (this is less useful than it
542 sounds, because Perl implements ** with C's I<pow()> function). First
543 I'll create a stub exponentiation function in I<power.pl>:
550 Now I'll create a C program, I<power.c>, with a function
551 I<PerlPower()> that contains all the perlguts necessary to push the
552 two arguments into I<expo()> and to pop the return value out. Take a
558 static PerlInterpreter *my_perl;
561 PerlPower(int a, int b)
563 dSP; /* initialize stack pointer */
564 ENTER; /* everything created after here */
565 SAVETMPS; /* ...is a temporary variable. */
566 PUSHMARK(SP); /* remember the stack pointer */
567 XPUSHs(sv_2mortal(newSViv(a))); /* push the base onto the stack */
568 XPUSHs(sv_2mortal(newSViv(b))); /* push the exponent onto stack */
569 PUTBACK; /* make local stack pointer global */
570 perl_call_pv("expo", G_SCALAR); /* call the function */
571 SPAGAIN; /* refresh stack pointer */
572 /* pop the return value from stack */
573 printf ("%d to the %dth power is %d.\n", a, b, POPi);
575 FREETMPS; /* free that return value */
576 LEAVE; /* ...and the XPUSHed "mortal" args.*/
579 int main (int argc, char **argv, char **env)
581 char *my_argv[] = { "", "power.pl" };
583 my_perl = perl_alloc();
584 perl_construct( my_perl );
586 perl_parse(my_perl, NULL, 2, my_argv, (char **)NULL);
589 PerlPower(3, 4); /*** Compute 3 ** 4 ***/
591 perl_destruct(my_perl);
599 % cc -o power power.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
602 3 to the 4th power is 81.
604 =head2 Maintaining a persistent interpreter
606 When developing interactive and/or potentially long-running
607 applications, it's a good idea to maintain a persistent interpreter
608 rather than allocating and constructing a new interpreter multiple
609 times. The major reason is speed: since Perl will only be loaded into
612 However, you have to be more cautious with namespace and variable
613 scoping when using a persistent interpreter. In previous examples
614 we've been using global variables in the default package C<main>. We
615 knew exactly what code would be run, and assumed we could avoid
616 variable collisions and outrageous symbol table growth.
618 Let's say your application is a server that will occasionally run Perl
619 code from some arbitrary file. Your server has no way of knowing what
620 code it's going to run. Very dangerous.
622 If the file is pulled in by C<perl_parse()>, compiled into a newly
623 constructed interpreter, and subsequently cleaned out with
624 C<perl_destruct()> afterwards, you're shielded from most namespace
627 One way to avoid namespace collisions in this scenario is to translate
628 the filename into a guaranteed-unique package name, and then compile
629 the code into that package using L<perlfunc/eval>. In the example
630 below, each file will only be compiled once. Or, the application
631 might choose to clean out the symbol table associated with the file
632 after it's no longer needed. Using L<perlcall/perl_call_argv>, We'll
633 call the subroutine C<Embed::Persistent::eval_file> which lives in the
634 file C<persistent.pl> and pass the filename and boolean cleanup/cache
637 Note that the process will continue to grow for each file that it
638 uses. In addition, there might be C<AUTOLOAD>ed subroutines and other
639 conditions that cause Perl's symbol table to grow. You might want to
640 add some logic that keeps track of the process size, or restarts
641 itself after a certain number of requests, to ensure that memory
642 consumption is minimized. You'll also want to scope your variables
643 with L<perlfunc/my> whenever possible.
646 package Embed::Persistent;
651 use Symbol qw(delete_package);
653 sub valid_package_name {
655 $string =~ s/([^A-Za-z0-9\/])/sprintf("_%2x",unpack("C",$1))/eg;
656 # second pass only for words starting with a digit
657 $string =~ s|/(\d)|sprintf("/_%2x",unpack("C",$1))|eg;
659 # Dress it up as a real package name
661 return "Embed" . $string;
665 my($filename, $delete) = @_;
666 my $package = valid_package_name($filename);
667 my $mtime = -M $filename;
668 if(defined $Cache{$package}{mtime}
670 $Cache{$package}{mtime} <= $mtime)
672 # we have compiled this subroutine already,
673 # it has not been updated on disk, nothing left to do
674 print STDERR "already compiled $package->handler\n";
678 open FH, $filename or die "open '$filename' $!";
683 #wrap the code into a subroutine inside our unique package
684 my $eval = qq{package $package; sub handler { $sub; }};
686 # hide our variables within this block
687 my($filename,$mtime,$package,$sub);
692 #cache it unless we're cleaning out each time
693 $Cache{$package}{mtime} = $mtime unless $delete;
696 eval {$package->handler;};
699 delete_package($package) if $delete;
701 #take a look if you want
702 #print Devel::Symdump->rnew($package)->as_string, $/;
713 /* 1 = clean out filename's symbol table after each request, 0 = don't */
718 static PerlInterpreter *perl = NULL;
721 main(int argc, char **argv, char **env)
723 char *embedding[] = { "", "persistent.pl" };
724 char *args[] = { "", DO_CLEAN, NULL };
725 char filename [1024];
728 if((perl = perl_alloc()) == NULL) {
729 fprintf(stderr, "no memory!");
732 perl_construct(perl);
734 exitstatus = perl_parse(perl, NULL, 2, embedding, NULL);
737 exitstatus = perl_run(perl);
739 while(printf("Enter file name: ") && gets(filename)) {
741 /* call the subroutine, passing it the filename as an argument */
743 perl_call_argv("Embed::Persistent::eval_file",
744 G_DISCARD | G_EVAL, args);
747 if(SvTRUE(GvSV(errgv)))
748 fprintf(stderr, "eval error: %s\n", SvPV(GvSV(errgv),na));
752 perl_destruct_level = 0;
760 % cc -o persistent persistent.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
762 Here's a example script file:
765 my $string = "hello";
769 print "foo says: @_\n";
775 Enter file name: test.pl
777 Enter file name: test.pl
778 already compiled Embed::test_2epl->handler
782 =head2 Maintaining multiple interpreter instances
784 Some rare applications will need to create more than one interpreter
785 during a session. Such an application might sporadically decide to
786 release any resources associated with the interpreter.
788 The program must take care to ensure that this takes place I<before>
789 the next interpreter is constructed. By default, the global variable
790 C<perl_destruct_level> is set to C<0>, since extra cleaning isn't
791 needed when a program has only one interpreter.
793 Setting C<perl_destruct_level> to C<1> makes everything squeaky clean:
795 perl_destruct_level = 1;
799 /* reset global variables here with perl_destruct_level = 1 */
800 perl_construct(my_perl);
802 /* clean and reset _everything_ during perl_destruct */
803 perl_destruct(my_perl);
806 /* let's go do it again! */
809 When I<perl_destruct()> is called, the interpreter's syntax parse tree
810 and symbol tables are cleaned up, and global variables are reset.
812 Now suppose we have more than one interpreter instance running at the
813 same time. This is feasible, but only if you used the
814 C<-DMULTIPLICITY> flag when building Perl. By default, that sets
815 C<perl_destruct_level> to C<1>.
823 /* we're going to embed two interpreters */
824 /* we're going to embed two interpreters */
826 #define SAY_HELLO "-e", "print qq(Hi, I'm $^X\n)"
828 int main(int argc, char **argv, char **env)
831 *one_perl = perl_alloc(),
832 *two_perl = perl_alloc();
833 char *one_args[] = { "one_perl", SAY_HELLO };
834 char *two_args[] = { "two_perl", SAY_HELLO };
836 perl_construct(one_perl);
837 perl_construct(two_perl);
839 perl_parse(one_perl, NULL, 3, one_args, (char **)NULL);
840 perl_parse(two_perl, NULL, 3, two_args, (char **)NULL);
845 perl_destruct(one_perl);
846 perl_destruct(two_perl);
855 % cc -o multiplicity multiplicity.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
863 =head2 Using Perl modules, which themselves use C libraries, from your C program
865 If you've played with the examples above and tried to embed a script
866 that I<use()>s a Perl module (such as I<Socket>) which itself uses a C or C++ library,
867 this probably happened:
870 Can't load module Socket, dynamic loading not available in this perl.
871 (You may need to build a new perl executable which either supports
872 dynamic loading or has the Socket module statically linked into it.)
877 Your interpreter doesn't know how to communicate with these extensions
878 on its own. A little glue will help. Up until now you've been
879 calling I<perl_parse()>, handing it NULL for the second argument:
881 perl_parse(my_perl, NULL, argc, my_argv, NULL);
883 That's where the glue code can be inserted to create the initial contact between
884 Perl and linked C/C++ routines. Let's take a look some pieces of I<perlmain.c>
885 to see how Perl does this:
889 # define EXTERN_C extern "C"
891 # define EXTERN_C extern
894 static void xs_init _((void));
896 EXTERN_C void boot_DynaLoader _((CV* cv));
897 EXTERN_C void boot_Socket _((CV* cv));
903 char *file = __FILE__;
904 /* DynaLoader is a special case */
905 newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
906 newXS("Socket::bootstrap", boot_Socket, file);
909 Simply put: for each extension linked with your Perl executable
910 (determined during its initial configuration on your
911 computer or when adding a new extension),
912 a Perl subroutine is created to incorporate the extension's
913 routines. Normally, that subroutine is named
914 I<Module::bootstrap()> and is invoked when you say I<use Module>. In
915 turn, this hooks into an XSUB, I<boot_Module>, which creates a Perl
916 counterpart for each of the extension's XSUBs. Don't worry about this
917 part; leave that to the I<xsubpp> and extension authors. If your
918 extension is dynamically loaded, DynaLoader creates I<Module::bootstrap()>
919 for you on the fly. In fact, if you have a working DynaLoader then there
920 is rarely any need to link in any other extensions statically.
923 Once you have this code, slap it into the second argument of I<perl_parse()>:
926 perl_parse(my_perl, xs_init, argc, my_argv, NULL);
931 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
935 use SomeDynamicallyLoadedModule;
937 print "Now I can use extensions!\n"'
939 B<ExtUtils::Embed> can also automate writing the I<xs_init> glue code.
941 % perl -MExtUtils::Embed -e xsinit -- -o perlxsi.c
942 % cc -c perlxsi.c `perl -MExtUtils::Embed -e ccopts`
943 % cc -c interp.c `perl -MExtUtils::Embed -e ccopts`
944 % cc -o interp perlxsi.o interp.o `perl -MExtUtils::Embed -e ldopts`
946 Consult L<perlxs> and L<perlguts> for more details.
948 =head1 Embedding Perl under Win32
950 At the time of this writing (5.004), there are two versions of Perl
951 which run under Win32. (The two versions are merging in 5.005.)
952 Interfacing to ActiveState's Perl library is quite different from the
953 examples in this documentation, as significant changes were made to
954 the internal Perl API. However, it is possible to embed ActiveState's
955 Perl runtime. For details, see the Perl for Win32 FAQ at
956 http://www.perl.com/perl/faq/win32/Perl_for_Win32_FAQ.html.
958 With the "official" Perl version 5.004 or higher, all the examples
959 within this documentation will compile and run untouched, although
960 the build process is slightly different between Unix and Win32.
962 For starters, backticks don't work under the Win32 native command shell.
963 The ExtUtils::Embed kit on CPAN ships with a script called
964 B<genmake>, which generates a simple makefile to build a program from
965 a single C source file. It can be used like this:
967 C:\ExtUtils-Embed\eg> perl genmake interp.c
968 C:\ExtUtils-Embed\eg> nmake
969 C:\ExtUtils-Embed\eg> interp -e "print qq{I'm embedded in Win32!\n}"
971 You may wish to use a more robust environment such as the Microsoft
972 Developer Studio. In this case, run this to generate perlxsi.c:
974 perl -MExtUtils::Embed -e xsinit
976 Create a new project and Insert -> Files into Project: perlxsi.c,
977 perl.lib, and your own source files, e.g. interp.c. Typically you'll
978 find perl.lib in B<C:\perl\lib\CORE>, if not, you should see the
979 B<CORE> directory relative to C<perl -V:archlib>. The studio will
980 also need this path so it knows where to find Perl include files.
981 This path can be added via the Tools -> Options -> Directories menu.
982 Finally, select Build -> Build interp.exe and you're ready to go.
986 You can sometimes I<write faster code> in C, but
987 you can always I<write code faster> in Perl. Because you can use
988 each from the other, combine them as you wish.
993 Jon Orwant <F<orwant@tpj.com>> and Doug MacEachern
994 <F<dougm@osf.org>>, with small contributions from Tim Bunce, Tom
995 Christiansen, Guy Decoux, Hallvard Furuseth, Dov Grobgeld, and Ilya
998 Doug MacEachern has an article on embedding in Volume 1, Issue 4 of
999 The Perl Journal (http://tpj.com). Doug is also the developer of the
1000 most widely-used Perl embedding: the mod_perl system
1001 (perl.apache.org), which embeds Perl in the Apache web server.
1002 Oracle, Binary Evolution, ActiveState, and Ben Sugars's nsapi_perl
1003 have used this model for Oracle, Netscape and Internet Information
1004 Server Perl plugins.
1010 Copyright (C) 1995, 1996, 1997, 1998 Doug MacEachern and Jon Orwant. All
1013 Permission is granted to make and distribute verbatim copies of this
1014 documentation provided the copyright notice and this permission notice are
1015 preserved on all copies.
1017 Permission is granted to copy and distribute modified versions of this
1018 documentation under the conditions for verbatim copying, provided also
1019 that they are marked clearly as modified versions, that the authors'
1020 names and title are unchanged (though subtitles and additional
1021 authors' names may be added), and that the entire resulting derived
1022 work is distributed under the terms of a permission notice identical
1025 Permission is granted to copy and distribute translations of this
1026 documentation into another language, under the above conditions for