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>, L<perlguts>, and L<perlapi>.
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?>
40 =item Compiling your C program
42 =item Adding a Perl interpreter to your C program
44 =item Calling a Perl subroutine from your C program
46 =item Evaluating a Perl statement from your C program
48 =item Performing Perl pattern matches and substitutions from your C program
50 =item Fiddling with the Perl stack from your C program
52 =item Maintaining a persistent interpreter
54 =item Maintaining multiple interpreter instances
56 =item Using Perl modules, which themselves use C libraries, from your C program
58 =item Embedding Perl under Win32
62 =head2 Compiling your C program
64 If you have trouble compiling the scripts in this documentation,
65 you're not alone. The cardinal rule: COMPILE THE PROGRAMS IN EXACTLY
66 THE SAME WAY THAT YOUR PERL WAS COMPILED. (Sorry for yelling.)
68 Also, every C program that uses Perl must link in the I<perl library>.
69 What's that, you ask? Perl is itself written in C; the perl library
70 is the collection of compiled C programs that were used to create your
71 perl executable (I</usr/bin/perl> or equivalent). (Corollary: you
72 can't use Perl from your C program unless Perl has been compiled on
73 your machine, or installed properly--that's why you shouldn't blithely
74 copy Perl executables from machine to machine without also copying the
77 When you use Perl from C, your C program will--usually--allocate,
78 "run", and deallocate a I<PerlInterpreter> object, which is defined by
81 If your copy of Perl is recent enough to contain this documentation
82 (version 5.002 or later), then the perl library (and I<EXTERN.h> and
83 I<perl.h>, which you'll also need) will reside in a directory
86 /usr/local/lib/perl5/your_architecture_here/CORE
90 /usr/local/lib/perl5/CORE
92 or maybe something like
96 Execute this statement for a hint about where to find CORE:
98 perl -MConfig -e 'print $Config{archlib}'
100 Here's how you'd compile the example in the next section,
101 L<Adding a Perl interpreter to your C program>, on my Linux box:
103 % gcc -O2 -Dbool=char -DHAS_BOOL -I/usr/local/include
104 -I/usr/local/lib/perl5/i586-linux/5.003/CORE
105 -L/usr/local/lib/perl5/i586-linux/5.003/CORE
106 -o interp interp.c -lperl -lm
108 (That's all one line.) On my DEC Alpha running old 5.003_05, the
109 incantation is a bit different:
111 % cc -O2 -Olimit 2900 -DSTANDARD_C -I/usr/local/include
112 -I/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE
113 -L/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE -L/usr/local/lib
114 -D__LANGUAGE_C__ -D_NO_PROTO -o interp interp.c -lperl -lm
116 How can you figure out what to add? Assuming your Perl is post-5.001,
117 execute a C<perl -V> command and pay special attention to the "cc" and
118 "ccflags" information.
120 You'll have to choose the appropriate compiler (I<cc>, I<gcc>, et al.) for
121 your machine: C<perl -MConfig -e 'print $Config{cc}'> will tell you what
124 You'll also have to choose the appropriate library directory
125 (I</usr/local/lib/...>) for your machine. If your compiler complains
126 that certain functions are undefined, or that it can't locate
127 I<-lperl>, then you need to change the path following the C<-L>. If it
128 complains that it can't find I<EXTERN.h> and I<perl.h>, you need to
129 change the path following the C<-I>.
131 You may have to add extra libraries as well. Which ones?
132 Perhaps those printed by
134 perl -MConfig -e 'print $Config{libs}'
136 Provided your perl binary was properly configured and installed the
137 B<ExtUtils::Embed> module will determine all of this information for
140 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
142 If the B<ExtUtils::Embed> module isn't part of your Perl distribution,
143 you can retrieve it from
144 http://www.perl.com/perl/CPAN/modules/by-module/ExtUtils/. (If
145 this documentation came from your Perl distribution, then you're
146 running 5.004 or better and you already have it.)
148 The B<ExtUtils::Embed> kit on CPAN also contains all source code for
149 the examples in this document, tests, additional examples and other
150 information you may find useful.
152 =head2 Adding a Perl interpreter to your C program
154 In a sense, perl (the C program) is a good example of embedding Perl
155 (the language), so I'll demonstrate embedding with I<miniperlmain.c>,
156 included in the source distribution. Here's a bastardized, nonportable
157 version of I<miniperlmain.c> containing the essentials of embedding:
159 #include <EXTERN.h> /* from the Perl distribution */
160 #include <perl.h> /* from the Perl distribution */
162 static PerlInterpreter *my_perl; /*** The Perl interpreter ***/
164 int main(int argc, char **argv, char **env)
166 my_perl = perl_alloc();
167 perl_construct(my_perl);
168 perl_parse(my_perl, NULL, argc, argv, (char **)NULL);
170 perl_destruct(my_perl);
174 Notice that we don't use the C<env> pointer. Normally handed to
175 C<perl_parse> as its final argument, C<env> here is replaced by
176 C<NULL>, which means that the current environment will be used.
178 Now compile this program (I'll call it I<interp.c>) into an executable:
180 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
182 After a successful compilation, you'll be able to use I<interp> just
186 print "Pretty Good Perl \n";
187 print "10890 - 9801 is ", 10890 - 9801;
194 % interp -e 'printf("%x", 3735928559)'
197 You can also read and execute Perl statements from a file while in the
198 midst of your C program, by placing the filename in I<argv[1]> before
201 =head2 Calling a Perl subroutine from your C program
203 To call individual Perl subroutines, you can use any of the B<call_*>
204 functions documented in L<perlcall>.
205 In this example we'll use C<call_argv>.
207 That's shown below, in a program I'll call I<showtime.c>.
212 static PerlInterpreter *my_perl;
214 int main(int argc, char **argv, char **env)
216 char *args[] = { NULL };
217 my_perl = perl_alloc();
218 perl_construct(my_perl);
220 perl_parse(my_perl, NULL, argc, argv, NULL);
222 /*** skipping perl_run() ***/
224 call_argv("showtime", G_DISCARD | G_NOARGS, args);
226 perl_destruct(my_perl);
230 where I<showtime> is a Perl subroutine that takes no arguments (that's the
231 I<G_NOARGS>) and for which I'll ignore the return value (that's the
232 I<G_DISCARD>). Those flags, and others, are discussed in L<perlcall>.
234 I'll define the I<showtime> subroutine in a file called I<showtime.pl>:
236 print "I shan't be printed.";
242 Simple enough. Now compile and run:
244 % cc -o showtime showtime.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
246 % showtime showtime.pl
249 yielding the number of seconds that elapsed between January 1, 1970
250 (the beginning of the Unix epoch), and the moment I began writing this
253 In this particular case we don't have to call I<perl_run>, but in
254 general it's considered good practice to ensure proper initialization
255 of library code, including execution of all object C<DESTROY> methods
256 and package C<END {}> blocks.
258 If you want to pass arguments to the Perl subroutine, you can add
259 strings to the C<NULL>-terminated C<args> list passed to
260 I<call_argv>. For other data types, or to examine return values,
261 you'll need to manipulate the Perl stack. That's demonstrated in
262 L<Fiddling with the Perl stack from your C program>.
264 =head2 Evaluating a Perl statement from your C program
266 Perl provides two API functions to evaluate pieces of Perl code.
267 These are L<perlapi/eval_sv> and L<perlapi/eval_pv>.
269 Arguably, these are the only routines you'll ever need to execute
270 snippets of Perl code from within your C program. Your code can be as
271 long as you wish; it can contain multiple statements; it can employ
272 L<perlfunc/use>, L<perlfunc/require>, and L<perlfunc/do> to
273 include external Perl files.
275 I<eval_pv> lets us evaluate individual Perl strings, and then
276 extract variables for coercion into C types. The following program,
277 I<string.c>, executes three Perl strings, extracting an C<int> from
278 the first, a C<float> from the second, and a C<char *> from the third.
283 static PerlInterpreter *my_perl;
285 main (int argc, char **argv, char **env)
288 char *embedding[] = { "", "-e", "0" };
290 my_perl = perl_alloc();
291 perl_construct( my_perl );
293 perl_parse(my_perl, NULL, 3, embedding, NULL);
296 /** Treat $a as an integer **/
297 eval_pv("$a = 3; $a **= 2", TRUE);
298 printf("a = %d\n", SvIV(get_sv("a", FALSE)));
300 /** Treat $a as a float **/
301 eval_pv("$a = 3.14; $a **= 2", TRUE);
302 printf("a = %f\n", SvNV(get_sv("a", FALSE)));
304 /** Treat $a as a string **/
305 eval_pv("$a = 'rekcaH lreP rehtonA tsuJ'; $a = reverse($a);", TRUE);
306 printf("a = %s\n", SvPV(get_sv("a", FALSE), n_a));
308 perl_destruct(my_perl);
312 All of those strange functions with I<sv> in their names help convert Perl scalars to C types. They're described in L<perlguts> and L<perlapi>.
314 If you compile and run I<string.c>, you'll see the results of using
315 I<SvIV()> to create an C<int>, I<SvNV()> to create a C<float>, and
316 I<SvPV()> to create a string:
320 a = Just Another Perl Hacker
322 In the example above, we've created a global variable to temporarily
323 store the computed value of our eval'd expression. It is also
324 possible and in most cases a better strategy to fetch the return value
325 from I<eval_pv()> instead. Example:
329 SV *val = eval_pv("reverse 'rekcaH lreP rehtonA tsuJ'", TRUE);
330 printf("%s\n", SvPV(val,n_a));
333 This way, we avoid namespace pollution by not creating global
334 variables and we've simplified our code as well.
336 =head2 Performing Perl pattern matches and substitutions from your C program
338 The I<eval_sv()> function lets us evaluate strings of Perl code, so we can
339 define some functions that use it to "specialize" in matches and
340 substitutions: I<match()>, I<substitute()>, and I<matches()>.
342 I32 match(SV *string, char *pattern);
344 Given a string and a pattern (e.g., C<m/clasp/> or C</\b\w*\b/>, which
345 in your C program might appear as "/\\b\\w*\\b/"), match()
346 returns 1 if the string matches the pattern and 0 otherwise.
348 int substitute(SV **string, char *pattern);
350 Given a pointer to an C<SV> and an C<=~> operation (e.g.,
351 C<s/bob/robert/g> or C<tr[A-Z][a-z]>), substitute() modifies the string
352 within the C<AV> at according to the operation, returning the number of substitutions
355 int matches(SV *string, char *pattern, AV **matches);
357 Given an C<SV>, a pattern, and a pointer to an empty C<AV>,
358 matches() evaluates C<$string =~ $pattern> in a list context, and
359 fills in I<matches> with the array elements, returning the number of matches found.
361 Here's a sample program, I<match.c>, that uses all three (long lines have
367 /** my_eval_sv(code, error_check)
368 ** kinda like eval_sv(),
369 ** but we pop the return value off the stack
371 SV* my_eval_sv(SV *sv, I32 croak_on_error)
378 eval_sv(sv, G_SCALAR);
384 if (croak_on_error && SvTRUE(ERRSV))
385 croak(SvPVx(ERRSV, n_a));
390 /** match(string, pattern)
392 ** Used for matches in a scalar context.
394 ** Returns 1 if the match was successful; 0 otherwise.
397 I32 match(SV *string, char *pattern)
399 SV *command = NEWSV(1099, 0), *retval;
402 sv_setpvf(command, "my $string = '%s'; $string =~ %s",
403 SvPV(string,n_a), pattern);
405 retval = my_eval_sv(command, TRUE);
406 SvREFCNT_dec(command);
411 /** substitute(string, pattern)
413 ** Used for =~ operations that modify their left-hand side (s/// and tr///)
415 ** Returns the number of successful matches, and
416 ** modifies the input string if there were any.
419 I32 substitute(SV **string, char *pattern)
421 SV *command = NEWSV(1099, 0), *retval;
424 sv_setpvf(command, "$string = '%s'; ($string =~ %s)",
425 SvPV(*string,n_a), pattern);
427 retval = my_eval_sv(command, TRUE);
428 SvREFCNT_dec(command);
430 *string = get_sv("string", FALSE);
434 /** matches(string, pattern, matches)
436 ** Used for matches in a list context.
438 ** Returns the number of matches,
439 ** and fills in **matches with the matching substrings
442 I32 matches(SV *string, char *pattern, AV **match_list)
444 SV *command = NEWSV(1099, 0);
448 sv_setpvf(command, "my $string = '%s'; @array = ($string =~ %s)",
449 SvPV(string,n_a), pattern);
451 my_eval_sv(command, TRUE);
452 SvREFCNT_dec(command);
454 *match_list = get_av("array", FALSE);
455 num_matches = av_len(*match_list) + 1; /** assume $[ is 0 **/
460 main (int argc, char **argv, char **env)
462 PerlInterpreter *my_perl = perl_alloc();
463 char *embedding[] = { "", "-e", "0" };
466 SV *text = NEWSV(1099,0);
469 perl_construct(my_perl);
470 perl_parse(my_perl, NULL, 3, embedding, NULL);
472 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");
474 if (match(text, "m/quarter/")) /** Does text contain 'quarter'? **/
475 printf("match: Text contains the word 'quarter'.\n\n");
477 printf("match: Text doesn't contain the word 'quarter'.\n\n");
479 if (match(text, "m/eighth/")) /** Does text contain 'eighth'? **/
480 printf("match: Text contains the word 'eighth'.\n\n");
482 printf("match: Text doesn't contain the word 'eighth'.\n\n");
484 /** Match all occurrences of /wi../ **/
485 num_matches = matches(text, "m/(wi..)/g", &match_list);
486 printf("matches: m/(wi..)/g found %d matches...\n", num_matches);
488 for (i = 0; i < num_matches; i++)
489 printf("match: %s\n", SvPV(*av_fetch(match_list, i, FALSE),n_a));
492 /** Remove all vowels from text **/
493 num_matches = substitute(&text, "s/[aeiou]//gi");
495 printf("substitute: s/[aeiou]//gi...%d substitutions made.\n",
497 printf("Now text is: %s\n\n", SvPV(text,n_a));
500 /** Attempt a substitution **/
501 if (!substitute(&text, "s/Perl/C/")) {
502 printf("substitute: s/Perl/C...No substitution made.\n\n");
506 PL_perl_destruct_level = 1;
507 perl_destruct(my_perl);
511 which produces the output (again, long lines have been wrapped here)
513 match: Text contains the word 'quarter'.
515 match: Text doesn't contain the word 'eighth'.
517 matches: m/(wi..)/g found 2 matches...
521 substitute: s/[aeiou]//gi...139 substitutions made.
522 Now text is: Whn h s t cnvnnc str nd th bll cms t sm mnt lk 76 cnts,
523 Mynrd s wr tht thr s smthng h *shld* d, smthng tht wll nbl hm t gt bck
524 qrtr, bt h hs n d *wht*. H fmbls thrgh hs rd sqzy chngprs nd gvs th by
525 thr xtr pnns wth hs dllr, hpng tht h mght lck nt th crrct mnt. Th by gvs
526 hm bck tw f hs wn pnns nd thn th bg shny qrtr tht s hs prz. -RCHH
528 substitute: s/Perl/C...No substitution made.
530 =head2 Fiddling with the Perl stack from your C program
532 When trying to explain stacks, most computer science textbooks mumble
533 something about spring-loaded columns of cafeteria plates: the last
534 thing you pushed on the stack is the first thing you pop off. That'll
535 do for our purposes: your C program will push some arguments onto "the Perl
536 stack", shut its eyes while some magic happens, and then pop the
537 results--the return value of your Perl subroutine--off the stack.
539 First you'll need to know how to convert between C types and Perl
540 types, with newSViv() and sv_setnv() and newAV() and all their
541 friends. They're described in L<perlguts> and L<perlapi>.
543 Then you'll need to know how to manipulate the Perl stack. That's
544 described in L<perlcall>.
546 Once you've understood those, embedding Perl in C is easy.
548 Because C has no builtin function for integer exponentiation, let's
549 make Perl's ** operator available to it (this is less useful than it
550 sounds, because Perl implements ** with C's I<pow()> function). First
551 I'll create a stub exponentiation function in I<power.pl>:
558 Now I'll create a C program, I<power.c>, with a function
559 I<PerlPower()> that contains all the perlguts necessary to push the
560 two arguments into I<expo()> and to pop the return value out. Take a
566 static PerlInterpreter *my_perl;
569 PerlPower(int a, int b)
571 dSP; /* initialize stack pointer */
572 ENTER; /* everything created after here */
573 SAVETMPS; /* ...is a temporary variable. */
574 PUSHMARK(SP); /* remember the stack pointer */
575 XPUSHs(sv_2mortal(newSViv(a))); /* push the base onto the stack */
576 XPUSHs(sv_2mortal(newSViv(b))); /* push the exponent onto stack */
577 PUTBACK; /* make local stack pointer global */
578 call_pv("expo", G_SCALAR); /* call the function */
579 SPAGAIN; /* refresh stack pointer */
580 /* pop the return value from stack */
581 printf ("%d to the %dth power is %d.\n", a, b, POPi);
583 FREETMPS; /* free that return value */
584 LEAVE; /* ...and the XPUSHed "mortal" args.*/
587 int main (int argc, char **argv, char **env)
589 char *my_argv[] = { "", "power.pl" };
591 my_perl = perl_alloc();
592 perl_construct( my_perl );
594 perl_parse(my_perl, NULL, 2, my_argv, (char **)NULL);
597 PerlPower(3, 4); /*** Compute 3 ** 4 ***/
599 perl_destruct(my_perl);
607 % cc -o power power.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
610 3 to the 4th power is 81.
612 =head2 Maintaining a persistent interpreter
614 When developing interactive and/or potentially long-running
615 applications, it's a good idea to maintain a persistent interpreter
616 rather than allocating and constructing a new interpreter multiple
617 times. The major reason is speed: since Perl will only be loaded into
620 However, you have to be more cautious with namespace and variable
621 scoping when using a persistent interpreter. In previous examples
622 we've been using global variables in the default package C<main>. We
623 knew exactly what code would be run, and assumed we could avoid
624 variable collisions and outrageous symbol table growth.
626 Let's say your application is a server that will occasionally run Perl
627 code from some arbitrary file. Your server has no way of knowing what
628 code it's going to run. Very dangerous.
630 If the file is pulled in by C<perl_parse()>, compiled into a newly
631 constructed interpreter, and subsequently cleaned out with
632 C<perl_destruct()> afterwards, you're shielded from most namespace
635 One way to avoid namespace collisions in this scenario is to translate
636 the filename into a guaranteed-unique package name, and then compile
637 the code into that package using L<perlfunc/eval>. In the example
638 below, each file will only be compiled once. Or, the application
639 might choose to clean out the symbol table associated with the file
640 after it's no longer needed. Using L<perlapi/call_argv>, We'll
641 call the subroutine C<Embed::Persistent::eval_file> which lives in the
642 file C<persistent.pl> and pass the filename and boolean cleanup/cache
645 Note that the process will continue to grow for each file that it
646 uses. In addition, there might be C<AUTOLOAD>ed subroutines and other
647 conditions that cause Perl's symbol table to grow. You might want to
648 add some logic that keeps track of the process size, or restarts
649 itself after a certain number of requests, to ensure that memory
650 consumption is minimized. You'll also want to scope your variables
651 with L<perlfunc/my> whenever possible.
654 package Embed::Persistent;
659 use Symbol qw(delete_package);
661 sub valid_package_name {
663 $string =~ s/([^A-Za-z0-9\/])/sprintf("_%2x",unpack("C",$1))/eg;
664 # second pass only for words starting with a digit
665 $string =~ s|/(\d)|sprintf("/_%2x",unpack("C",$1))|eg;
667 # Dress it up as a real package name
669 return "Embed" . $string;
673 my($filename, $delete) = @_;
674 my $package = valid_package_name($filename);
675 my $mtime = -M $filename;
676 if(defined $Cache{$package}{mtime}
678 $Cache{$package}{mtime} <= $mtime)
680 # we have compiled this subroutine already,
681 # it has not been updated on disk, nothing left to do
682 print STDERR "already compiled $package->handler\n";
686 open FH, $filename or die "open '$filename' $!";
691 #wrap the code into a subroutine inside our unique package
692 my $eval = qq{package $package; sub handler { $sub; }};
694 # hide our variables within this block
695 my($filename,$mtime,$package,$sub);
700 #cache it unless we're cleaning out each time
701 $Cache{$package}{mtime} = $mtime unless $delete;
704 eval {$package->handler;};
707 delete_package($package) if $delete;
709 #take a look if you want
710 #print Devel::Symdump->rnew($package)->as_string, $/;
721 /* 1 = clean out filename's symbol table after each request, 0 = don't */
726 static PerlInterpreter *perl = NULL;
729 main(int argc, char **argv, char **env)
731 char *embedding[] = { "", "persistent.pl" };
732 char *args[] = { "", DO_CLEAN, NULL };
733 char filename [1024];
737 if((perl = perl_alloc()) == NULL) {
738 fprintf(stderr, "no memory!");
741 perl_construct(perl);
743 exitstatus = perl_parse(perl, NULL, 2, embedding, NULL);
746 exitstatus = perl_run(perl);
748 while(printf("Enter file name: ") && gets(filename)) {
750 /* call the subroutine, passing it the filename as an argument */
752 call_argv("Embed::Persistent::eval_file",
753 G_DISCARD | G_EVAL, args);
757 fprintf(stderr, "eval error: %s\n", SvPV(ERRSV,n_a));
761 PL_perl_destruct_level = 0;
769 % cc -o persistent persistent.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
771 Here's a example script file:
774 my $string = "hello";
778 print "foo says: @_\n";
784 Enter file name: test.pl
786 Enter file name: test.pl
787 already compiled Embed::test_2epl->handler
791 =head2 Maintaining multiple interpreter instances
793 Some rare applications will need to create more than one interpreter
794 during a session. Such an application might sporadically decide to
795 release any resources associated with the interpreter.
797 The program must take care to ensure that this takes place I<before>
798 the next interpreter is constructed. By default, the global variable
799 C<PL_perl_destruct_level> is set to C<0>, since extra cleaning isn't
800 needed when a program has only one interpreter.
802 Setting C<PL_perl_destruct_level> to C<1> makes everything squeaky clean:
804 PL_perl_destruct_level = 1;
808 /* reset global variables here with PL_perl_destruct_level = 1 */
809 perl_construct(my_perl);
811 /* clean and reset _everything_ during perl_destruct */
812 perl_destruct(my_perl);
815 /* let's go do it again! */
818 When I<perl_destruct()> is called, the interpreter's syntax parse tree
819 and symbol tables are cleaned up, and global variables are reset.
821 Now suppose we have more than one interpreter instance running at the
822 same time. This is feasible, but only if you used the
823 C<-DMULTIPLICITY> flag when building Perl. By default, that sets
824 C<PL_perl_destruct_level> to C<1>.
832 /* we're going to embed two interpreters */
833 /* we're going to embed two interpreters */
835 #define SAY_HELLO "-e", "print qq(Hi, I'm $^X\n)"
837 int main(int argc, char **argv, char **env)
840 *one_perl = perl_alloc(),
841 *two_perl = perl_alloc();
842 char *one_args[] = { "one_perl", SAY_HELLO };
843 char *two_args[] = { "two_perl", SAY_HELLO };
845 perl_construct(one_perl);
846 perl_construct(two_perl);
848 perl_parse(one_perl, NULL, 3, one_args, (char **)NULL);
849 perl_parse(two_perl, NULL, 3, two_args, (char **)NULL);
854 perl_destruct(one_perl);
855 perl_destruct(two_perl);
864 % cc -o multiplicity multiplicity.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
872 =head2 Using Perl modules, which themselves use C libraries, from your C program
874 If you've played with the examples above and tried to embed a script
875 that I<use()>s a Perl module (such as I<Socket>) which itself uses a C or C++ library,
876 this probably happened:
879 Can't load module Socket, dynamic loading not available in this perl.
880 (You may need to build a new perl executable which either supports
881 dynamic loading or has the Socket module statically linked into it.)
886 Your interpreter doesn't know how to communicate with these extensions
887 on its own. A little glue will help. Up until now you've been
888 calling I<perl_parse()>, handing it NULL for the second argument:
890 perl_parse(my_perl, NULL, argc, my_argv, NULL);
892 That's where the glue code can be inserted to create the initial contact between
893 Perl and linked C/C++ routines. Let's take a look some pieces of I<perlmain.c>
894 to see how Perl does this:
896 static void xs_init (pTHX);
898 EXTERN_C void boot_DynaLoader (pTHX_ CV* cv);
899 EXTERN_C void boot_Socket (pTHX_ CV* cv);
905 char *file = __FILE__;
906 /* DynaLoader is a special case */
907 newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
908 newXS("Socket::bootstrap", boot_Socket, file);
911 Simply put: for each extension linked with your Perl executable
912 (determined during its initial configuration on your
913 computer or when adding a new extension),
914 a Perl subroutine is created to incorporate the extension's
915 routines. Normally, that subroutine is named
916 I<Module::bootstrap()> and is invoked when you say I<use Module>. In
917 turn, this hooks into an XSUB, I<boot_Module>, which creates a Perl
918 counterpart for each of the extension's XSUBs. Don't worry about this
919 part; leave that to the I<xsubpp> and extension authors. If your
920 extension is dynamically loaded, DynaLoader creates I<Module::bootstrap()>
921 for you on the fly. In fact, if you have a working DynaLoader then there
922 is rarely any need to link in any other extensions statically.
925 Once you have this code, slap it into the second argument of I<perl_parse()>:
928 perl_parse(my_perl, xs_init, argc, my_argv, NULL);
933 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
937 use SomeDynamicallyLoadedModule;
939 print "Now I can use extensions!\n"'
941 B<ExtUtils::Embed> can also automate writing the I<xs_init> glue code.
943 % perl -MExtUtils::Embed -e xsinit -- -o perlxsi.c
944 % cc -c perlxsi.c `perl -MExtUtils::Embed -e ccopts`
945 % cc -c interp.c `perl -MExtUtils::Embed -e ccopts`
946 % cc -o interp perlxsi.o interp.o `perl -MExtUtils::Embed -e ldopts`
948 Consult L<perlxs>, L<perlguts>, and L<perlapi> for more details.
950 =head1 Embedding Perl under Win32
952 In general, all of the source code shown here should work unmodified under
955 However, there are some caveats about the command-line examples shown.
956 For starters, backticks won't work under the Win32 native command shell.
957 The ExtUtils::Embed kit on CPAN ships with a script called
958 B<genmake>, which generates a simple makefile to build a program from
959 a single C source file. It can be used like this:
961 C:\ExtUtils-Embed\eg> perl genmake interp.c
962 C:\ExtUtils-Embed\eg> nmake
963 C:\ExtUtils-Embed\eg> interp -e "print qq{I'm embedded in Win32!\n}"
965 You may wish to use a more robust environment such as the Microsoft
966 Developer Studio. In this case, run this to generate perlxsi.c:
968 perl -MExtUtils::Embed -e xsinit
970 Create a new project and Insert -> Files into Project: perlxsi.c,
971 perl.lib, and your own source files, e.g. interp.c. Typically you'll
972 find perl.lib in B<C:\perl\lib\CORE>, if not, you should see the
973 B<CORE> directory relative to C<perl -V:archlib>. The studio will
974 also need this path so it knows where to find Perl include files.
975 This path can be added via the Tools -> Options -> Directories menu.
976 Finally, select Build -> Build interp.exe and you're ready to go.
980 You can sometimes I<write faster code> in C, but
981 you can always I<write code faster> in Perl. Because you can use
982 each from the other, combine them as you wish.
987 Jon Orwant <F<orwant@tpj.com>> and Doug MacEachern
988 <F<dougm@osf.org>>, with small contributions from Tim Bunce, Tom
989 Christiansen, Guy Decoux, Hallvard Furuseth, Dov Grobgeld, and Ilya
992 Doug MacEachern has an article on embedding in Volume 1, Issue 4 of
993 The Perl Journal (http://tpj.com). Doug is also the developer of the
994 most widely-used Perl embedding: the mod_perl system
995 (perl.apache.org), which embeds Perl in the Apache web server.
996 Oracle, Binary Evolution, ActiveState, and Ben Sugars's nsapi_perl
997 have used this model for Oracle, Netscape and Internet Information
1004 Copyright (C) 1995, 1996, 1997, 1998 Doug MacEachern and Jon Orwant. All
1007 Permission is granted to make and distribute verbatim copies of this
1008 documentation provided the copyright notice and this permission notice are
1009 preserved on all copies.
1011 Permission is granted to copy and distribute modified versions of this
1012 documentation under the conditions for verbatim copying, provided also
1013 that they are marked clearly as modified versions, that the authors'
1014 names and title are unchanged (though subtitles and additional
1015 authors' names may be added), and that the entire resulting derived
1016 work is distributed under the terms of a permission notice identical
1019 Permission is granted to copy and distribute translations of this
1020 documentation into another language, under the above conditions for