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?>
42 Compiling your C program
46 Adding a Perl interpreter to your C program
50 Calling a Perl subroutine from your C program
54 Evaluating a Perl statement from your C program
58 Performing Perl pattern matches and substitutions from your C program
62 Fiddling with the Perl stack from your C program
66 Maintaining a persistent interpreter
70 Maintaining multiple interpreter instances
74 Using Perl modules, which themselves use C libraries, from your C program
78 Embedding Perl under Win32
82 =head2 Compiling your C program
84 If you have trouble compiling the scripts in this documentation,
85 you're not alone. The cardinal rule: COMPILE THE PROGRAMS IN EXACTLY
86 THE SAME WAY THAT YOUR PERL WAS COMPILED. (Sorry for yelling.)
88 Also, every C program that uses Perl must link in the I<perl library>.
89 What's that, you ask? Perl is itself written in C; the perl library
90 is the collection of compiled C programs that were used to create your
91 perl executable (I</usr/bin/perl> or equivalent). (Corollary: you
92 can't use Perl from your C program unless Perl has been compiled on
93 your machine, or installed properly--that's why you shouldn't blithely
94 copy Perl executables from machine to machine without also copying the
97 When you use Perl from C, your C program will--usually--allocate,
98 "run", and deallocate a I<PerlInterpreter> object, which is defined by
101 If your copy of Perl is recent enough to contain this documentation
102 (version 5.002 or later), then the perl library (and I<EXTERN.h> and
103 I<perl.h>, which you'll also need) will reside in a directory
104 that looks like this:
106 /usr/local/lib/perl5/your_architecture_here/CORE
110 /usr/local/lib/perl5/CORE
112 or maybe something like
116 Execute this statement for a hint about where to find CORE:
118 perl -MConfig -e 'print $Config{archlib}'
120 Here's how you'd compile the example in the next section,
121 L<Adding a Perl interpreter to your C program>, on my Linux box:
123 % gcc -O2 -Dbool=char -DHAS_BOOL -I/usr/local/include
124 -I/usr/local/lib/perl5/i586-linux/5.003/CORE
125 -L/usr/local/lib/perl5/i586-linux/5.003/CORE
126 -o interp interp.c -lperl -lm
128 (That's all one line.) On my DEC Alpha running old 5.003_05, the
129 incantation is a bit different:
131 % cc -O2 -Olimit 2900 -DSTANDARD_C -I/usr/local/include
132 -I/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE
133 -L/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE -L/usr/local/lib
134 -D__LANGUAGE_C__ -D_NO_PROTO -o interp interp.c -lperl -lm
136 How can you figure out what to add? Assuming your Perl is post-5.001,
137 execute a C<perl -V> command and pay special attention to the "cc" and
138 "ccflags" information.
140 You'll have to choose the appropriate compiler (I<cc>, I<gcc>, et al.) for
141 your machine: C<perl -MConfig -e 'print $Config{cc}'> will tell you what
144 You'll also have to choose the appropriate library directory
145 (I</usr/local/lib/...>) for your machine. If your compiler complains
146 that certain functions are undefined, or that it can't locate
147 I<-lperl>, then you need to change the path following the C<-L>. If it
148 complains that it can't find I<EXTERN.h> and I<perl.h>, you need to
149 change the path following the C<-I>.
151 You may have to add extra libraries as well. Which ones?
152 Perhaps those printed by
154 perl -MConfig -e 'print $Config{libs}'
156 Provided your perl binary was properly configured and installed the
157 B<ExtUtils::Embed> module will determine all of this information for
160 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
162 If the B<ExtUtils::Embed> module isn't part of your Perl distribution,
163 you can retrieve it from
164 http://www.perl.com/perl/CPAN/modules/by-module/ExtUtils/
165 (If this documentation came from your Perl distribution, then you're
166 running 5.004 or better and you already have it.)
168 The B<ExtUtils::Embed> kit on CPAN also contains all source code for
169 the examples in this document, tests, additional examples and other
170 information you may find useful.
172 =head2 Adding a Perl interpreter to your C program
174 In a sense, perl (the C program) is a good example of embedding Perl
175 (the language), so I'll demonstrate embedding with I<miniperlmain.c>,
176 included in the source distribution. Here's a bastardized, non-portable
177 version of I<miniperlmain.c> containing the essentials of embedding:
179 #include <EXTERN.h> /* from the Perl distribution */
180 #include <perl.h> /* from the Perl distribution */
182 static PerlInterpreter *my_perl; /*** The Perl interpreter ***/
184 int main(int argc, char **argv, char **env)
186 PERL_SYS_INIT3(&argc,&argv,&env);
187 my_perl = perl_alloc();
188 perl_construct(my_perl);
189 PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
190 perl_parse(my_perl, NULL, argc, argv, (char **)NULL);
192 perl_destruct(my_perl);
197 Notice that we don't use the C<env> pointer. Normally handed to
198 C<perl_parse> as its final argument, C<env> here is replaced by
199 C<NULL>, which means that the current environment will be used.
201 The macros PERL_SYS_INIT3() and PERL_SYS_TERM() provide system-specific
202 tune up of the C runtime environment necessary to run Perl interpreters;
203 they should only be called once regardless of how many interpreters you
204 create or destroy. Call PERL_SYS_INIT3() before you create your first
205 interpreter, and PERL_SYS_TERM() after you free your last interpreter.
207 Since PERL_SYS_INIT3() may change C<env>, it may be more appropriate to
208 provide C<env> as an argument to perl_parse().
210 Also notice that no matter what arguments you pass to perl_parse(),
211 PERL_SYS_INIT3() must be invoked on the C main() argc, argv and env and
214 Now compile this program (I'll call it I<interp.c>) into an executable:
216 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
218 After a successful compilation, you'll be able to use I<interp> just
222 print "Pretty Good Perl \n";
223 print "10890 - 9801 is ", 10890 - 9801;
230 % interp -e 'printf("%x", 3735928559)'
233 You can also read and execute Perl statements from a file while in the
234 midst of your C program, by placing the filename in I<argv[1]> before
237 =head2 Calling a Perl subroutine from your C program
239 To call individual Perl subroutines, you can use any of the B<call_*>
240 functions documented in L<perlcall>.
241 In this example we'll use C<call_argv>.
243 That's shown below, in a program I'll call I<showtime.c>.
248 static PerlInterpreter *my_perl;
250 int main(int argc, char **argv, char **env)
252 char *args[] = { NULL };
253 PERL_SYS_INIT3(&argc,&argv,&env);
254 my_perl = perl_alloc();
255 perl_construct(my_perl);
257 perl_parse(my_perl, NULL, argc, argv, NULL);
258 PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
260 /*** skipping perl_run() ***/
262 call_argv("showtime", G_DISCARD | G_NOARGS, args);
264 perl_destruct(my_perl);
269 where I<showtime> is a Perl subroutine that takes no arguments (that's the
270 I<G_NOARGS>) and for which I'll ignore the return value (that's the
271 I<G_DISCARD>). Those flags, and others, are discussed in L<perlcall>.
273 I'll define the I<showtime> subroutine in a file called I<showtime.pl>:
275 print "I shan't be printed.";
281 Simple enough. Now compile and run:
283 % cc -o showtime showtime.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
285 % showtime showtime.pl
288 yielding the number of seconds that elapsed between January 1, 1970
289 (the beginning of the Unix epoch), and the moment I began writing this
292 In this particular case we don't have to call I<perl_run>, as we set
293 the PL_exit_flag PERL_EXIT_DESTRUCT_END which executes END blocks in
296 If you want to pass arguments to the Perl subroutine, you can add
297 strings to the C<NULL>-terminated C<args> list passed to
298 I<call_argv>. For other data types, or to examine return values,
299 you'll need to manipulate the Perl stack. That's demonstrated in
300 L<Fiddling with the Perl stack from your C program>.
302 =head2 Evaluating a Perl statement from your C program
304 Perl provides two API functions to evaluate pieces of Perl code.
305 These are L<perlapi/eval_sv> and L<perlapi/eval_pv>.
307 Arguably, these are the only routines you'll ever need to execute
308 snippets of Perl code from within your C program. Your code can be as
309 long as you wish; it can contain multiple statements; it can employ
310 L<perlfunc/use>, L<perlfunc/require>, and L<perlfunc/do> to
311 include external Perl files.
313 I<eval_pv> lets us evaluate individual Perl strings, and then
314 extract variables for coercion into C types. The following program,
315 I<string.c>, executes three Perl strings, extracting an C<int> from
316 the first, a C<float> from the second, and a C<char *> from the third.
321 static PerlInterpreter *my_perl;
323 main (int argc, char **argv, char **env)
325 char *embedding[] = { "", "-e", "0" };
327 PERL_SYS_INIT3(&argc,&argv,&env);
328 my_perl = perl_alloc();
329 perl_construct( my_perl );
331 perl_parse(my_perl, NULL, 3, embedding, NULL);
332 PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
335 /** Treat $a as an integer **/
336 eval_pv("$a = 3; $a **= 2", TRUE);
337 printf("a = %d\n", SvIV(get_sv("a", 0)));
339 /** Treat $a as a float **/
340 eval_pv("$a = 3.14; $a **= 2", TRUE);
341 printf("a = %f\n", SvNV(get_sv("a", 0)));
343 /** Treat $a as a string **/
344 eval_pv("$a = 'rekcaH lreP rehtonA tsuJ'; $a = reverse($a);", TRUE);
345 printf("a = %s\n", SvPV_nolen(get_sv("a", 0)));
347 perl_destruct(my_perl);
352 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>.
354 If you compile and run I<string.c>, you'll see the results of using
355 I<SvIV()> to create an C<int>, I<SvNV()> to create a C<float>, and
356 I<SvPV()> to create a string:
360 a = Just Another Perl Hacker
362 In the example above, we've created a global variable to temporarily
363 store the computed value of our eval'ed expression. It is also
364 possible and in most cases a better strategy to fetch the return value
365 from I<eval_pv()> instead. Example:
368 SV *val = eval_pv("reverse 'rekcaH lreP rehtonA tsuJ'", TRUE);
369 printf("%s\n", SvPV_nolen(val));
372 This way, we avoid namespace pollution by not creating global
373 variables and we've simplified our code as well.
375 =head2 Performing Perl pattern matches and substitutions from your C program
377 The I<eval_sv()> function lets us evaluate strings of Perl code, so we can
378 define some functions that use it to "specialize" in matches and
379 substitutions: I<match()>, I<substitute()>, and I<matches()>.
381 I32 match(SV *string, char *pattern);
383 Given a string and a pattern (e.g., C<m/clasp/> or C</\b\w*\b/>, which
384 in your C program might appear as "/\\b\\w*\\b/"), match()
385 returns 1 if the string matches the pattern and 0 otherwise.
387 int substitute(SV **string, char *pattern);
389 Given a pointer to an C<SV> and an C<=~> operation (e.g.,
390 C<s/bob/robert/g> or C<tr[A-Z][a-z]>), substitute() modifies the string
391 within the C<SV> as according to the operation, returning the number of substitutions
394 int matches(SV *string, char *pattern, AV **matches);
396 Given an C<SV>, a pattern, and a pointer to an empty C<AV>,
397 matches() evaluates C<$string =~ $pattern> in a list context, and
398 fills in I<matches> with the array elements, returning the number of matches found.
400 Here's a sample program, I<match.c>, that uses all three (long lines have
406 static PerlInterpreter *my_perl;
408 /** my_eval_sv(code, error_check)
409 ** kinda like eval_sv(),
410 ** but we pop the return value off the stack
412 SV* my_eval_sv(SV *sv, I32 croak_on_error)
419 eval_sv(sv, G_SCALAR);
425 if (croak_on_error && SvTRUE(ERRSV))
426 croak(SvPVx_nolen(ERRSV));
431 /** match(string, pattern)
433 ** Used for matches in a scalar context.
435 ** Returns 1 if the match was successful; 0 otherwise.
438 I32 match(SV *string, char *pattern)
440 SV *command = newSV(0), *retval;
442 sv_setpvf(command, "my $string = '%s'; $string =~ %s",
443 SvPV_nolen(string), pattern);
445 retval = my_eval_sv(command, TRUE);
446 SvREFCNT_dec(command);
451 /** substitute(string, pattern)
453 ** Used for =~ operations that modify their left-hand side (s/// and tr///)
455 ** Returns the number of successful matches, and
456 ** modifies the input string if there were any.
459 I32 substitute(SV **string, char *pattern)
461 SV *command = newSV(0), *retval;
463 sv_setpvf(command, "$string = '%s'; ($string =~ %s)",
464 SvPV_nolen(*string), pattern);
466 retval = my_eval_sv(command, TRUE);
467 SvREFCNT_dec(command);
469 *string = get_sv("string", 0);
473 /** matches(string, pattern, matches)
475 ** Used for matches in a list context.
477 ** Returns the number of matches,
478 ** and fills in **matches with the matching substrings
481 I32 matches(SV *string, char *pattern, AV **match_list)
483 SV *command = newSV(0);
486 sv_setpvf(command, "my $string = '%s'; @array = ($string =~ %s)",
487 SvPV_nolen(string), pattern);
489 my_eval_sv(command, TRUE);
490 SvREFCNT_dec(command);
492 *match_list = get_av("array", 0);
493 num_matches = av_len(*match_list) + 1;
498 main (int argc, char **argv, char **env)
500 char *embedding[] = { "", "-e", "0" };
505 PERL_SYS_INIT3(&argc,&argv,&env);
506 my_perl = perl_alloc();
507 perl_construct(my_perl);
508 perl_parse(my_perl, NULL, 3, embedding, NULL);
509 PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
512 sv_setpv(text, "When he is at a convenience store and the "
513 "bill comes to some amount like 76 cents, Maynard is "
514 "aware that there is something he *should* do, something "
515 "that will enable him to get back a quarter, but he has "
516 "no idea *what*. He fumbles through his red squeezey "
517 "changepurse and gives the boy three extra pennies with "
518 "his dollar, hoping that he might luck into the correct "
519 "amount. The boy gives him back two of his own pennies "
520 "and then the big shiny quarter that is his prize. "
523 if (match(text, "m/quarter/")) /** Does text contain 'quarter'? **/
524 printf("match: Text contains the word 'quarter'.\n\n");
526 printf("match: Text doesn't contain the word 'quarter'.\n\n");
528 if (match(text, "m/eighth/")) /** Does text contain 'eighth'? **/
529 printf("match: Text contains the word 'eighth'.\n\n");
531 printf("match: Text doesn't contain the word 'eighth'.\n\n");
533 /** Match all occurrences of /wi../ **/
534 num_matches = matches(text, "m/(wi..)/g", &match_list);
535 printf("matches: m/(wi..)/g found %d matches...\n", num_matches);
537 for (i = 0; i < num_matches; i++)
538 printf("match: %s\n", SvPV_nolen(*av_fetch(match_list, i, FALSE)));
541 /** Remove all vowels from text **/
542 num_matches = substitute(&text, "s/[aeiou]//gi");
544 printf("substitute: s/[aeiou]//gi...%d substitutions made.\n",
546 printf("Now text is: %s\n\n", SvPV_nolen(text));
549 /** Attempt a substitution **/
550 if (!substitute(&text, "s/Perl/C/")) {
551 printf("substitute: s/Perl/C...No substitution made.\n\n");
555 PL_perl_destruct_level = 1;
556 perl_destruct(my_perl);
561 which produces the output (again, long lines have been wrapped here)
563 match: Text contains the word 'quarter'.
565 match: Text doesn't contain the word 'eighth'.
567 matches: m/(wi..)/g found 2 matches...
571 substitute: s/[aeiou]//gi...139 substitutions made.
572 Now text is: Whn h s t cnvnnc str nd th bll cms t sm mnt lk 76 cnts,
573 Mynrd s wr tht thr s smthng h *shld* d, smthng tht wll nbl hm t gt bck
574 qrtr, bt h hs n d *wht*. H fmbls thrgh hs rd sqzy chngprs nd gvs th by
575 thr xtr pnns wth hs dllr, hpng tht h mght lck nt th crrct mnt. Th by gvs
576 hm bck tw f hs wn pnns nd thn th bg shny qrtr tht s hs prz. -RCHH
578 substitute: s/Perl/C...No substitution made.
580 =head2 Fiddling with the Perl stack from your C program
582 When trying to explain stacks, most computer science textbooks mumble
583 something about spring-loaded columns of cafeteria plates: the last
584 thing you pushed on the stack is the first thing you pop off. That'll
585 do for our purposes: your C program will push some arguments onto "the Perl
586 stack", shut its eyes while some magic happens, and then pop the
587 results--the return value of your Perl subroutine--off the stack.
589 First you'll need to know how to convert between C types and Perl
590 types, with newSViv() and sv_setnv() and newAV() and all their
591 friends. They're described in L<perlguts> and L<perlapi>.
593 Then you'll need to know how to manipulate the Perl stack. That's
594 described in L<perlcall>.
596 Once you've understood those, embedding Perl in C is easy.
598 Because C has no builtin function for integer exponentiation, let's
599 make Perl's ** operator available to it (this is less useful than it
600 sounds, because Perl implements ** with C's I<pow()> function). First
601 I'll create a stub exponentiation function in I<power.pl>:
608 Now I'll create a C program, I<power.c>, with a function
609 I<PerlPower()> that contains all the perlguts necessary to push the
610 two arguments into I<expo()> and to pop the return value out. Take a
616 static PerlInterpreter *my_perl;
619 PerlPower(int a, int b)
621 dSP; /* initialize stack pointer */
622 ENTER; /* everything created after here */
623 SAVETMPS; /* ...is a temporary variable. */
624 PUSHMARK(SP); /* remember the stack pointer */
625 XPUSHs(sv_2mortal(newSViv(a))); /* push the base onto the stack */
626 XPUSHs(sv_2mortal(newSViv(b))); /* push the exponent onto stack */
627 PUTBACK; /* make local stack pointer global */
628 call_pv("expo", G_SCALAR); /* call the function */
629 SPAGAIN; /* refresh stack pointer */
630 /* pop the return value from stack */
631 printf ("%d to the %dth power is %d.\n", a, b, POPi);
633 FREETMPS; /* free that return value */
634 LEAVE; /* ...and the XPUSHed "mortal" args.*/
637 int main (int argc, char **argv, char **env)
639 char *my_argv[] = { "", "power.pl" };
641 PERL_SYS_INIT3(&argc,&argv,&env);
642 my_perl = perl_alloc();
643 perl_construct( my_perl );
645 perl_parse(my_perl, NULL, 2, my_argv, (char **)NULL);
646 PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
649 PerlPower(3, 4); /*** Compute 3 ** 4 ***/
651 perl_destruct(my_perl);
660 % cc -o power power.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
663 3 to the 4th power is 81.
665 =head2 Maintaining a persistent interpreter
667 When developing interactive and/or potentially long-running
668 applications, it's a good idea to maintain a persistent interpreter
669 rather than allocating and constructing a new interpreter multiple
670 times. The major reason is speed: since Perl will only be loaded into
673 However, you have to be more cautious with namespace and variable
674 scoping when using a persistent interpreter. In previous examples
675 we've been using global variables in the default package C<main>. We
676 knew exactly what code would be run, and assumed we could avoid
677 variable collisions and outrageous symbol table growth.
679 Let's say your application is a server that will occasionally run Perl
680 code from some arbitrary file. Your server has no way of knowing what
681 code it's going to run. Very dangerous.
683 If the file is pulled in by C<perl_parse()>, compiled into a newly
684 constructed interpreter, and subsequently cleaned out with
685 C<perl_destruct()> afterwards, you're shielded from most namespace
688 One way to avoid namespace collisions in this scenario is to translate
689 the filename into a guaranteed-unique package name, and then compile
690 the code into that package using L<perlfunc/eval>. In the example
691 below, each file will only be compiled once. Or, the application
692 might choose to clean out the symbol table associated with the file
693 after it's no longer needed. Using L<perlapi/call_argv>, We'll
694 call the subroutine C<Embed::Persistent::eval_file> which lives in the
695 file C<persistent.pl> and pass the filename and boolean cleanup/cache
698 Note that the process will continue to grow for each file that it
699 uses. In addition, there might be C<AUTOLOAD>ed subroutines and other
700 conditions that cause Perl's symbol table to grow. You might want to
701 add some logic that keeps track of the process size, or restarts
702 itself after a certain number of requests, to ensure that memory
703 consumption is minimized. You'll also want to scope your variables
704 with L<perlfunc/my> whenever possible.
707 package Embed::Persistent;
712 use Symbol qw(delete_package);
714 sub valid_package_name {
716 $string =~ s/([^A-Za-z0-9\/])/sprintf("_%2x",unpack("C",$1))/eg;
717 # second pass only for words starting with a digit
718 $string =~ s|/(\d)|sprintf("/_%2x",unpack("C",$1))|eg;
720 # Dress it up as a real package name
722 return "Embed" . $string;
726 my($filename, $delete) = @_;
727 my $package = valid_package_name($filename);
728 my $mtime = -M $filename;
729 if(defined $Cache{$package}{mtime}
731 $Cache{$package}{mtime} <= $mtime)
733 # we have compiled this subroutine already,
734 # it has not been updated on disk, nothing left to do
735 print STDERR "already compiled $package->handler\n";
739 open FH, $filename or die "open '$filename' $!";
744 #wrap the code into a subroutine inside our unique package
745 my $eval = qq{package $package; sub handler { $sub; }};
747 # hide our variables within this block
748 my($filename,$mtime,$package,$sub);
753 #cache it unless we're cleaning out each time
754 $Cache{$package}{mtime} = $mtime unless $delete;
757 eval {$package->handler;};
760 delete_package($package) if $delete;
762 #take a look if you want
763 #print Devel::Symdump->rnew($package)->as_string, $/;
774 /* 1 = clean out filename's symbol table after each request, 0 = don't */
779 #define BUFFER_SIZE 1024
781 static PerlInterpreter *my_perl = NULL;
784 main(int argc, char **argv, char **env)
786 char *embedding[] = { "", "persistent.pl" };
787 char *args[] = { "", DO_CLEAN, NULL };
788 char filename[BUFFER_SIZE];
791 PERL_SYS_INIT3(&argc,&argv,&env);
792 if((my_perl = perl_alloc()) == NULL) {
793 fprintf(stderr, "no memory!");
796 perl_construct(my_perl);
798 PL_origalen = 1; /* don't let $0 assignment update the proctitle or embedding[0] */
799 exitstatus = perl_parse(my_perl, NULL, 2, embedding, NULL);
800 PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
802 exitstatus = perl_run(my_perl);
804 while(printf("Enter file name: ") &&
805 fgets(filename, BUFFER_SIZE, stdin)) {
807 filename[strlen(filename)-1] = '\0'; /* strip \n */
808 /* call the subroutine, passing it the filename as an argument */
810 call_argv("Embed::Persistent::eval_file",
811 G_DISCARD | G_EVAL, args);
815 fprintf(stderr, "eval error: %s\n", SvPV_nolen(ERRSV));
819 PL_perl_destruct_level = 0;
820 perl_destruct(my_perl);
828 % cc -o persistent persistent.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
830 Here's an example script file:
833 my $string = "hello";
837 print "foo says: @_\n";
843 Enter file name: test.pl
845 Enter file name: test.pl
846 already compiled Embed::test_2epl->handler
850 =head2 Execution of END blocks
852 Traditionally END blocks have been executed at the end of the perl_run.
853 This causes problems for applications that never call perl_run. Since
854 perl 5.7.2 you can specify C<PL_exit_flags |= PERL_EXIT_DESTRUCT_END>
855 to get the new behaviour. This also enables the running of END blocks if
856 the perl_parse fails and C<perl_destruct> will return the exit value.
858 =head2 $0 assignments
860 When a perl script assigns a value to $0 then the perl runtime will
861 try to make this value show up as the program name reported by "ps" by
862 updating the memory pointed to by the argv passed to perl_parse() and
863 also calling API functions like setproctitle() where available. This
864 behaviour might not be appropriate when embedding perl and can be
865 disabled by assigning the value C<1> to the variable C<PL_origalen>
866 before perl_parse() is called.
868 The F<persistent.c> example above is for instance likely to segfault
869 when $0 is assigned to if the C<PL_origalen = 1;> assignment is
870 removed. This because perl will try to write to the read only memory
871 of the C<embedding[]> strings.
873 =head2 Maintaining multiple interpreter instances
875 Some rare applications will need to create more than one interpreter
876 during a session. Such an application might sporadically decide to
877 release any resources associated with the interpreter.
879 The program must take care to ensure that this takes place I<before>
880 the next interpreter is constructed. By default, when perl is not
881 built with any special options, the global variable
882 C<PL_perl_destruct_level> is set to C<0>, since extra cleaning isn't
883 usually needed when a program only ever creates a single interpreter
884 in its entire lifetime.
886 Setting C<PL_perl_destruct_level> to C<1> makes everything squeaky clean:
890 /* reset global variables here with PL_perl_destruct_level = 1 */
891 PL_perl_destruct_level = 1;
892 perl_construct(my_perl);
894 /* clean and reset _everything_ during perl_destruct */
895 PL_perl_destruct_level = 1;
896 perl_destruct(my_perl);
899 /* let's go do it again! */
902 When I<perl_destruct()> is called, the interpreter's syntax parse tree
903 and symbol tables are cleaned up, and global variables are reset. The
904 second assignment to C<PL_perl_destruct_level> is needed because
905 perl_construct resets it to C<0>.
907 Now suppose we have more than one interpreter instance running at the
908 same time. This is feasible, but only if you used the Configure option
909 C<-Dusemultiplicity> or the options C<-Dusethreads -Duseithreads> when
910 building perl. By default, enabling one of these Configure options
911 sets the per-interpreter global variable C<PL_perl_destruct_level> to
912 C<1>, so that thorough cleaning is automatic and interpreter variables
913 are initialized correctly. Even if you don't intend to run two or
914 more interpreters at the same time, but to run them sequentially, like
915 in the above example, it is recommended to build perl with the
916 C<-Dusemultiplicity> option otherwise some interpreter variables may
917 not be initialized correctly between consecutive runs and your
918 application may crash.
920 See also L<perlxs/Thread-aware system interfaces>.
922 Using C<-Dusethreads -Duseithreads> rather than C<-Dusemultiplicity>
923 is more appropriate if you intend to run multiple interpreters
924 concurrently in different threads, because it enables support for
925 linking in the thread libraries of your system with the interpreter.
933 /* we're going to embed two interpreters */
934 /* we're going to embed two interpreters */
936 #define SAY_HELLO "-e", "print qq(Hi, I'm $^X\n)"
938 int main(int argc, char **argv, char **env)
940 PerlInterpreter *one_perl, *two_perl;
941 char *one_args[] = { "one_perl", SAY_HELLO };
942 char *two_args[] = { "two_perl", SAY_HELLO };
944 PERL_SYS_INIT3(&argc,&argv,&env);
945 one_perl = perl_alloc();
946 two_perl = perl_alloc();
948 PERL_SET_CONTEXT(one_perl);
949 perl_construct(one_perl);
950 PERL_SET_CONTEXT(two_perl);
951 perl_construct(two_perl);
953 PERL_SET_CONTEXT(one_perl);
954 perl_parse(one_perl, NULL, 3, one_args, (char **)NULL);
955 PERL_SET_CONTEXT(two_perl);
956 perl_parse(two_perl, NULL, 3, two_args, (char **)NULL);
958 PERL_SET_CONTEXT(one_perl);
960 PERL_SET_CONTEXT(two_perl);
963 PERL_SET_CONTEXT(one_perl);
964 perl_destruct(one_perl);
965 PERL_SET_CONTEXT(two_perl);
966 perl_destruct(two_perl);
968 PERL_SET_CONTEXT(one_perl);
970 PERL_SET_CONTEXT(two_perl);
975 Note the calls to PERL_SET_CONTEXT(). These are necessary to initialize
976 the global state that tracks which interpreter is the "current" one on
977 the particular process or thread that may be running it. It should
978 always be used if you have more than one interpreter and are making
979 perl API calls on both interpreters in an interleaved fashion.
981 PERL_SET_CONTEXT(interp) should also be called whenever C<interp> is
982 used by a thread that did not create it (using either perl_alloc(), or
983 the more esoteric perl_clone()).
987 % cc -o multiplicity multiplicity.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
995 =head2 Using Perl modules, which themselves use C libraries, from your C program
997 If you've played with the examples above and tried to embed a script
998 that I<use()>s a Perl module (such as I<Socket>) which itself uses a C or C++ library,
999 this probably happened:
1002 Can't load module Socket, dynamic loading not available in this perl.
1003 (You may need to build a new perl executable which either supports
1004 dynamic loading or has the Socket module statically linked into it.)
1009 Your interpreter doesn't know how to communicate with these extensions
1010 on its own. A little glue will help. Up until now you've been
1011 calling I<perl_parse()>, handing it NULL for the second argument:
1013 perl_parse(my_perl, NULL, argc, my_argv, NULL);
1015 That's where the glue code can be inserted to create the initial contact between
1016 Perl and linked C/C++ routines. Let's take a look some pieces of I<perlmain.c>
1017 to see how Perl does this:
1019 static void xs_init (pTHX);
1021 EXTERN_C void boot_DynaLoader (pTHX_ CV* cv);
1022 EXTERN_C void boot_Socket (pTHX_ CV* cv);
1028 char *file = __FILE__;
1029 /* DynaLoader is a special case */
1030 newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
1031 newXS("Socket::bootstrap", boot_Socket, file);
1034 Simply put: for each extension linked with your Perl executable
1035 (determined during its initial configuration on your
1036 computer or when adding a new extension),
1037 a Perl subroutine is created to incorporate the extension's
1038 routines. Normally, that subroutine is named
1039 I<Module::bootstrap()> and is invoked when you say I<use Module>. In
1040 turn, this hooks into an XSUB, I<boot_Module>, which creates a Perl
1041 counterpart for each of the extension's XSUBs. Don't worry about this
1042 part; leave that to the I<xsubpp> and extension authors. If your
1043 extension is dynamically loaded, DynaLoader creates I<Module::bootstrap()>
1044 for you on the fly. In fact, if you have a working DynaLoader then there
1045 is rarely any need to link in any other extensions statically.
1048 Once you have this code, slap it into the second argument of I<perl_parse()>:
1051 perl_parse(my_perl, xs_init, argc, my_argv, NULL);
1056 % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
1060 use SomeDynamicallyLoadedModule;
1062 print "Now I can use extensions!\n"'
1064 B<ExtUtils::Embed> can also automate writing the I<xs_init> glue code.
1066 % perl -MExtUtils::Embed -e xsinit -- -o perlxsi.c
1067 % cc -c perlxsi.c `perl -MExtUtils::Embed -e ccopts`
1068 % cc -c interp.c `perl -MExtUtils::Embed -e ccopts`
1069 % cc -o interp perlxsi.o interp.o `perl -MExtUtils::Embed -e ldopts`
1071 Consult L<perlxs>, L<perlguts>, and L<perlapi> for more details.
1073 =head1 Embedding Perl under Win32
1075 In general, all of the source code shown here should work unmodified under
1078 However, there are some caveats about the command-line examples shown.
1079 For starters, backticks won't work under the Win32 native command shell.
1080 The ExtUtils::Embed kit on CPAN ships with a script called
1081 B<genmake>, which generates a simple makefile to build a program from
1082 a single C source file. It can be used like this:
1084 C:\ExtUtils-Embed\eg> perl genmake interp.c
1085 C:\ExtUtils-Embed\eg> nmake
1086 C:\ExtUtils-Embed\eg> interp -e "print qq{I'm embedded in Win32!\n}"
1088 You may wish to use a more robust environment such as the Microsoft
1089 Developer Studio. In this case, run this to generate perlxsi.c:
1091 perl -MExtUtils::Embed -e xsinit
1093 Create a new project and Insert -> Files into Project: perlxsi.c,
1094 perl.lib, and your own source files, e.g. interp.c. Typically you'll
1095 find perl.lib in B<C:\perl\lib\CORE>, if not, you should see the
1096 B<CORE> directory relative to C<perl -V:archlib>. The studio will
1097 also need this path so it knows where to find Perl include files.
1098 This path can be added via the Tools -> Options -> Directories menu.
1099 Finally, select Build -> Build interp.exe and you're ready to go.
1103 If you completely hide the short forms of the Perl public API,
1104 add -DPERL_NO_SHORT_NAMES to the compilation flags. This means that
1105 for example instead of writing
1107 warn("%d bottles of beer on the wall", bottlecount);
1109 you will have to write the explicit full form
1111 Perl_warn(aTHX_ "%d bottles of beer on the wall", bottlecount);
1113 (See L<perlguts/Background and PERL_IMPLICIT_CONTEXT for the explanation
1114 of the C<aTHX_>.> ) Hiding the short forms is very useful for avoiding
1115 all sorts of nasty (C preprocessor or otherwise) conflicts with other
1116 software packages (Perl defines about 2400 APIs with these short names,
1117 take or leave few hundred, so there certainly is room for conflict.)
1121 You can sometimes I<write faster code> in C, but
1122 you can always I<write code faster> in Perl. Because you can use
1123 each from the other, combine them as you wish.
1128 Jon Orwant <F<orwant@media.mit.edu>> and Doug MacEachern
1129 <F<dougm@covalent.net>>, with small contributions from Tim Bunce, Tom
1130 Christiansen, Guy Decoux, Hallvard Furuseth, Dov Grobgeld, and Ilya
1133 Doug MacEachern has an article on embedding in Volume 1, Issue 4 of
1134 The Perl Journal ( http://www.tpj.com/ ). Doug is also the developer of the
1135 most widely-used Perl embedding: the mod_perl system
1136 (perl.apache.org), which embeds Perl in the Apache web server.
1137 Oracle, Binary Evolution, ActiveState, and Ben Sugars's nsapi_perl
1138 have used this model for Oracle, Netscape and Internet Information
1139 Server Perl plugins.
1143 Copyright (C) 1995, 1996, 1997, 1998 Doug MacEachern and Jon Orwant. All
1146 Permission is granted to make and distribute verbatim copies of this
1147 documentation provided the copyright notice and this permission notice are
1148 preserved on all copies.
1150 Permission is granted to copy and distribute modified versions of this
1151 documentation under the conditions for verbatim copying, provided also
1152 that they are marked clearly as modified versions, that the authors'
1153 names and title are unchanged (though subtitles and additional
1154 authors' names may be added), and that the entire resulting derived
1155 work is distributed under the terms of a permission notice identical
1158 Permission is granted to copy and distribute translations of this
1159 documentation into another language, under the above conditions for