Cari di Perl 
    Perl User Manual
Daftar Isi
(Sebelumnya) How to submit bug reports on PerlGuts of Perl debugging (Berikutnya)
Internals and C language interface

How to embed perl in your C program

Daftar Isi

NAME

perlembed - how to embed perl in your C program

DESCRIPTION

PREAMBLE

Do you want to:

ROADMAP

  • Compiling your C program

  • Adding a Perl interpreter to your C program

  • Calling a Perl subroutine from your C program

  • Evaluating a Perl statement from your C program

  • Performing Perl pattern matches and substitutions from your C program

  • Fiddling with the Perl stack from your C program

  • Maintaining a persistent interpreter

  • Maintaining multiple interpreter instances

  • Using Perl modules, which themselves use C libraries, from your C program

  • Embedding Perl under Win32

Compiling your C program

If you have trouble compiling the scripts in this documentation,you're not alone. The cardinal rule: COMPILE THE PROGRAMS IN EXACTLYTHE SAME WAY THAT YOUR PERL WAS COMPILED. (Sorry for yelling.)

Also, every C program that uses Perl must link in the perl library.What's that, you ask? Perl is itself written in C; the perl libraryis the collection of compiled C programs that were used to create yourperl executable (/usr/bin/perl or equivalent). (Corollary: youcan't use Perl from your C program unless Perl has been compiled onyour machine, or installed properly--that's why you shouldn't blithelycopy Perl executables from machine to machine without also copying thelib directory.)

When you use Perl from C, your C program will--usually--allocate,"run", and deallocate a PerlInterpreter object, which is defined bythe perl library.

If your copy of Perl is recent enough to contain this documentation(version 5.002 or later), then the perl library (and EXTERN.h andperl.h, which you'll also need) will reside in a directorythat looks like this:

  1. /usr/local/lib/perl5/your_architecture_here/CORE

or perhaps just

  1. /usr/local/lib/perl5/CORE

or maybe something like

  1. /usr/opt/perl5/CORE

Execute this statement for a hint about where to find CORE:

  1. perl -MConfig -e 'print $Config{archlib}'

Here's how you'd compile the example in the next section,Adding a Perl interpreter to your C program, on my Linux box:

  1. % gcc -O2 -Dbool=char -DHAS_BOOL -I/usr/local/include
  2. -I/usr/local/lib/perl5/i586-linux/5.003/CORE
  3. -L/usr/local/lib/perl5/i586-linux/5.003/CORE
  4. -o interp interp.c -lperl -lm

(That's all one line.) On my DEC Alpha running old 5.003_05, the incantation is a bit different:

  1. % cc -O2 -Olimit 2900 -DSTANDARD_C -I/usr/local/include
  2. -I/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE
  3. -L/usr/local/lib/perl5/alpha-dec_osf/5.00305/CORE -L/usr/local/lib
  4. -D__LANGUAGE_C__ -D_NO_PROTO -o interp interp.c -lperl -lm

How can you figure out what to add? Assuming your Perl is post-5.001,execute a perl -V command and pay special attention to the "cc" and"ccflags" information.

You'll have to choose the appropriate compiler (cc, gcc, et al.) foryour machine: perl -MConfig -e 'print $Config{cc}' will tell you whatto use.

You'll also have to choose the appropriate library directory(/usr/local/lib/...) for your machine. If your compiler complainsthat certain functions are undefined, or that it can't locate-lperl, then you need to change the path following the -L. If itcomplains that it can't find EXTERN.h and perl.h, you need tochange the path following the -I.

You may have to add extra libraries as well. Which ones?Perhaps those printed by

  1. perl -MConfig -e 'print $Config{libs}'

Provided your perl binary was properly configured and installed theExtUtils::Embed module will determine all of this information foryou:

  1. % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`

If the ExtUtils::Embed module isn't part of your Perl distribution,you can retrieve it fromhttp://www.perl.com/perl/CPAN/modules/by-module/ExtUtils/(If this documentation came from your Perl distribution, then you'rerunning 5.004 or better and you already have it.)

The ExtUtils::Embed kit on CPAN also contains all source code forthe examples in this document, tests, additional examples and otherinformation you may find useful.

Adding a Perl interpreter to your C program

In a sense, perl (the C program) is a good example of embedding Perl(the language), so I'll demonstrate embedding with miniperlmain.c,included in the source distribution. Here's a bastardized, non-portableversion of miniperlmain.c containing the essentials of embedding:

  1. #include <EXTERN.h> /* from the Perl distribution */
  2. #include <perl.h> /* from the Perl distribution */
  3. static PerlInterpreter *my_perl; /*** The Perl interpreter ***/
  4. int main(int argc, char **argv, char **env)
  5. {
  6. PERL_SYS_INIT3(&argc,&argv,&env);
  7. my_perl = perl_alloc();
  8. perl_construct(my_perl);
  9. PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
  10. perl_parse(my_perl, NULL, argc, argv, (char **)NULL);
  11. perl_run(my_perl);
  12. perl_destruct(my_perl);
  13. perl_free(my_perl);
  14. PERL_SYS_TERM();
  15. }

Notice that we don't use the env pointer. Normally handed toperl_parse as its final argument, env here is replaced byNULL, which means that the current environment will be used.

The macros PERL_SYS_INIT3() and PERL_SYS_TERM() provide system-specifictune up of the C runtime environment necessary to run Perl interpreters;they should only be called once regardless of how many interpreters youcreate or destroy. Call PERL_SYS_INIT3() before you create your firstinterpreter, and PERL_SYS_TERM() after you free your last interpreter.

Since PERL_SYS_INIT3() may change env, it may be more appropriate toprovide env as an argument to perl_parse().

Also notice that no matter what arguments you pass to perl_parse(),PERL_SYS_INIT3() must be invoked on the C main() argc, argv and env andonly once.

Now compile this program (I'll call it interp.c) into an executable:

  1. % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`

After a successful compilation, you'll be able to use interp justlike perl itself:

  1. % interp
  2. print "Pretty Good Perl \n";
  3. print "10890 - 9801 is ", 10890 - 9801;
  4. <CTRL-D>
  5. Pretty Good Perl
  6. 10890 - 9801 is 1089

or

  1. % interp -e 'printf("%x", 3735928559)'
  2. deadbeef

You can also read and execute Perl statements from a file while in themidst of your C program, by placing the filename in argv[1] beforecalling perl_run.

Calling a Perl subroutine from your C program

To call individual Perl subroutines, you can use any of the call_*functions documented in perlcall.In this example we'll use call_argv.

That's shown below, in a program I'll call showtime.c.

  1. #include <EXTERN.h>
  2. #include <perl.h>
  3. static PerlInterpreter *my_perl;
  4. int main(int argc, char **argv, char **env)
  5. {
  6. char *args[] = { NULL };
  7. PERL_SYS_INIT3(&argc,&argv,&env);
  8. my_perl = perl_alloc();
  9. perl_construct(my_perl);
  10. perl_parse(my_perl, NULL, argc, argv, NULL);
  11. PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
  12. /*** skipping perl_run() ***/
  13. call_argv("showtime", G_DISCARD | G_NOARGS, args);
  14. perl_destruct(my_perl);
  15. perl_free(my_perl);
  16. PERL_SYS_TERM();
  17. }

where showtime is a Perl subroutine that takes no arguments (that's theG_NOARGS) and for which I'll ignore the return value (that's theG_DISCARD). Those flags, and others, are discussed in perlcall.

I'll define the showtime subroutine in a file called showtime.pl:

  1. print "I shan't be printed.";
  2. sub showtime {
  3. print time;
  4. }

Simple enough. Now compile and run:

  1. % cc -o showtime showtime.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
  2. % showtime showtime.pl
  3. 818284590

yielding the number of seconds that elapsed between January 1, 1970(the beginning of the Unix epoch), and the moment I began writing thissentence.

In this particular case we don't have to call perl_run, as we set the PL_exit_flag PERL_EXIT_DESTRUCT_END which executes END blocks inperl_destruct.

If you want to pass arguments to the Perl subroutine, you can addstrings to the NULL-terminated args list passed tocall_argv. For other data types, or to examine return values,you'll need to manipulate the Perl stack. That's demonstrated inFiddling with the Perl stack from your C program.

Evaluating a Perl statement from your C program

Perl provides two API functions to evaluate pieces of Perl code.These are eval_sv in perlapi and eval_pv in perlapi.

Arguably, these are the only routines you'll ever need to executesnippets of Perl code from within your C program. Your code can be aslong as you wish; it can contain multiple statements; it can employuse, require, and do toinclude external Perl files.

eval_pv lets us evaluate individual Perl strings, and thenextract variables for coercion into C types. The following program,string.c, executes three Perl strings, extracting an int fromthe first, a float from the second, and a char * from the third.

  1. #include <EXTERN.h>
  2. #include <perl.h>
  3. static PerlInterpreter *my_perl;
  4. main (int argc, char **argv, char **env)
  5. {
  6. char *embedding[] = { "", "-e", "0" };
  7. PERL_SYS_INIT3(&argc,&argv,&env);
  8. my_perl = perl_alloc();
  9. perl_construct( my_perl );
  10. perl_parse(my_perl, NULL, 3, embedding, NULL);
  11. PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
  12. perl_run(my_perl);
  13. /** Treat $a as an integer **/
  14. eval_pv("$a = 3; $a **= 2", TRUE);
  15. printf("a = %d\n", SvIV(get_sv("a", 0)));
  16. /** Treat $a as a float **/
  17. eval_pv("$a = 3.14; $a **= 2", TRUE);
  18. printf("a = %f\n", SvNV(get_sv("a", 0)));
  19. /** Treat $a as a string **/
  20. eval_pv("$a = 'rekcaH lreP rehtonA tsuJ'; $a = reverse($a);", TRUE);
  21. printf("a = %s\n", SvPV_nolen(get_sv("a", 0)));
  22. perl_destruct(my_perl);
  23. perl_free(my_perl);
  24. PERL_SYS_TERM();
  25. }

All of those strange functions with sv in their names help convert Perl scalars to C types. They're described in perlguts and perlapi.

If you compile and run string.c, you'll see the results of usingSvIV() to create an int, SvNV() to create a float, andSvPV() to create a string:

  1. a = 9
  2. a = 9.859600
  3. a = Just Another Perl Hacker

In the example above, we've created a global variable to temporarilystore the computed value of our eval'ed expression. It is alsopossible and in most cases a better strategy to fetch the return valuefrom eval_pv() instead. Example:

  1. ...
  2. SV *val = eval_pv("reverse 'rekcaH lreP rehtonA tsuJ'", TRUE);
  3. printf("%s\n", SvPV_nolen(val));
  4. ...

This way, we avoid namespace pollution by not creating globalvariables and we've simplified our code as well.

Performing Perl pattern matches and substitutions from your C program

The eval_sv() function lets us evaluate strings of Perl code, so we candefine some functions that use it to "specialize" in matches andsubstitutions: match(), substitute(), and matches().

  1. I32 match(SV *string, char *pattern);

Given a string and a pattern (e.g., m/clasp/ or /\b\w*\b/, whichin your C program might appear as "/\b\w*\b/"), match()returns 1 if the string matches the pattern and 0 otherwise.

  1. int substitute(SV **string, char *pattern);

Given a pointer to an SV and an =~ operation (e.g.,s/bob/robert/g or tr[A-Z][a-z]), substitute() modifies the stringwithin the SV as according to the operation, returning the number of substitutionsmade.

  1. int matches(SV *string, char *pattern, AV **matches);

Given an SV, a pattern, and a pointer to an empty AV,matches() evaluates $string =~ $pattern in a list context, andfills in matches with the array elements, returning the number of matches found.

Here's a sample program, match.c, that uses all three (long lines havebeen wrapped here):

  1. #include <EXTERN.h>
  2. #include <perl.h>
  3. static PerlInterpreter *my_perl;
  4. /** my_eval_sv(code, error_check)
  5. ** kinda like eval_sv(),
  6. ** but we pop the return value off the stack
  7. **/
  8. SV* my_eval_sv(SV *sv, I32 croak_on_error)
  9. {
  10. dSP;
  11. SV* retval;
  12. PUSHMARK(SP);
  13. eval_sv(sv, G_SCALAR);
  14. SPAGAIN;
  15. retval = POPs;
  16. PUTBACK;
  17. if (croak_on_error && SvTRUE(ERRSV))
  18. croak(SvPVx_nolen(ERRSV));
  19. return retval;
  20. }
  21. /** match(string, pattern)
  22. **
  23. ** Used for matches in a scalar context.
  24. **
  25. ** Returns 1 if the match was successful; 0 otherwise.
  26. **/
  27. I32 match(SV *string, char *pattern)
  28. {
  29. SV *command = newSV(0), *retval;
  30. sv_setpvf(command, "my $string = '%s' $string =~ %s",
  31. SvPV_nolen(string), pattern);
  32. retval = my_eval_sv(command, TRUE);
  33. SvREFCNT_dec(command);
  34. return SvIV(retval);
  35. }
  36. /** substitute(string, pattern)
  37. **
  38. ** Used for =~ operations that modify their left-hand side (s/// and tr///)
  39. **
  40. ** Returns the number of successful matches, and
  41. ** modifies the input string if there were any.
  42. **/
  43. I32 substitute(SV **string, char *pattern)
  44. {
  45. SV *command = newSV(0), *retval;
  46. sv_setpvf(command, "$string = '%s' ($string =~ %s)",
  47. SvPV_nolen(*string), pattern);
  48. retval = my_eval_sv(command, TRUE);
  49. SvREFCNT_dec(command);
  50. *string = get_sv("string", 0);
  51. return SvIV(retval);
  52. }
  53. /** matches(string, pattern, matches)
  54. **
  55. ** Used for matches in a list context.
  56. **
  57. ** Returns the number of matches,
  58. ** and fills in **matches with the matching substrings
  59. **/
  60. I32 matches(SV *string, char *pattern, AV **match_list)
  61. {
  62. SV *command = newSV(0);
  63. I32 num_matches;
  64. sv_setpvf(command, "my $string = '%s' @array = ($string =~ %s)",
  65. SvPV_nolen(string), pattern);
  66. my_eval_sv(command, TRUE);
  67. SvREFCNT_dec(command);
  68. *match_list = get_av("array", 0);
  69. num_matches = av_len(*match_list) + 1;
  70. return num_matches;
  71. }
  72. main (int argc, char **argv, char **env)
  73. {
  74. char *embedding[] = { "", "-e", "0" };
  75. AV *match_list;
  76. I32 num_matches, i;
  77. SV *text;
  78. PERL_SYS_INIT3(&argc,&argv,&env);
  79. my_perl = perl_alloc();
  80. perl_construct(my_perl);
  81. perl_parse(my_perl, NULL, 3, embedding, NULL);
  82. PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
  83. text = newSV(0);
  84. sv_setpv(text, "When he is at a convenience store and the "
  85. "bill comes to some amount like 76 cents, Maynard is "
  86. "aware that there is something he *should* do, something "
  87. "that will enable him to get back a quarter, but he has "
  88. "no idea *what*. He fumbles through his red squeezey "
  89. "changepurse and gives the boy three extra pennies with "
  90. "his dollar, hoping that he might luck into the correct "
  91. "amount. The boy gives him back two of his own pennies "
  92. "and then the big shiny quarter that is his prize. "
  93. "-RICHH");
  94. if (match(text, "m/quarter/")) /** Does text contain 'quarter'? **/
  95. printf("match: Text contains the word 'quarter'.\n\n");
  96. else
  97. printf("match: Text doesn't contain the word 'quarter'.\n\n");
  98. if (match(text, "m/eighth/")) /** Does text contain 'eighth'? **/
  99. printf("match: Text contains the word 'eighth'.\n\n");
  100. else
  101. printf("match: Text doesn't contain the word 'eighth'.\n\n");
  102. /** Match all occurrences of /wi../ **/
  103. num_matches = matches(text, "m/(wi..)/g", &match_list);
  104. printf("matches: m/(wi..)/g found %d matches...\n", num_matches);
  105. for (i = 0; i < num_matches; i++)
  106. printf("match: %s\n", SvPV_nolen(*av_fetch(match_list, i, FALSE)));
  107. printf("\n");
  108. /** Remove all vowels from text **/
  109. num_matches = substitute(&text, "s/[aeiou]//gi");
  110. if (num_matches) {
  111. printf("substitute: s/[aeiou]//gi...%d substitutions made.\n",
  112. num_matches);
  113. printf("Now text is: %s\n\n", SvPV_nolen(text));
  114. }
  115. /** Attempt a substitution **/
  116. if (!substitute(&text, "s/Perl/C/")) {
  117. printf("substitute: s/Perl/C...No substitution made.\n\n");
  118. }
  119. SvREFCNT_dec(text);
  120. PL_perl_destruct_level = 1;
  121. perl_destruct(my_perl);
  122. perl_free(my_perl);
  123. PERL_SYS_TERM();
  124. }

which produces the output (again, long lines have been wrapped here)

  1. match: Text contains the word 'quarter'.
  2. match: Text doesn't contain the word 'eighth'.
  3. matches: m/(wi..)/g found 2 matches...
  4. match: will
  5. match: with
  6. substitute: s/[aeiou]//gi...139 substitutions made.
  7. Now text is: Whn h s t cnvnnc str nd th bll cms t sm mnt lk 76 cnts,
  8. Mynrd s wr tht thr s smthng h *shld* d, smthng tht wll nbl hm t gt bck
  9. qrtr, bt h hs n d *wht*. H fmbls thrgh hs rd sqzy chngprs nd gvs th by
  10. thr xtr pnns wth hs dllr, hpng tht h mght lck nt th crrct mnt. Th by gvs
  11. hm bck tw f hs wn pnns nd thn th bg shny qrtr tht s hs prz. -RCHH
  12. substitute: s/Perl/C...No substitution made.

Fiddling with the Perl stack from your C program

When trying to explain stacks, most computer science textbooks mumblesomething about spring-loaded columns of cafeteria plates: the lastthing you pushed on the stack is the first thing you pop off. That'lldo for our purposes: your C program will push some arguments onto "the Perlstack", shut its eyes while some magic happens, and then pop theresults--the return value of your Perl subroutine--off the stack.

First you'll need to know how to convert between C types and Perltypes, with newSViv() and sv_setnv() and newAV() and all theirfriends. They're described in perlguts and perlapi.

Then you'll need to know how to manipulate the Perl stack. That'sdescribed in perlcall.

Once you've understood those, embedding Perl in C is easy.

Because C has no builtin function for integer exponentiation, let'smake Perl's ** operator available to it (this is less useful than itsounds, because Perl implements ** with C's pow() function). FirstI'll create a stub exponentiation function in power.pl:

  1. sub expo {
  2. my ($a, $b) = @_;
  3. return $a ** $b;
  4. }

Now I'll create a C program, power.c, with a functionPerlPower() that contains all the perlguts necessary to push thetwo arguments into expo() and to pop the return value out. Take adeep breath...

  1. #include <EXTERN.h>
  2. #include <perl.h>
  3. static PerlInterpreter *my_perl;
  4. static void
  5. PerlPower(int a, int b)
  6. {
  7. dSP; /* initialize stack pointer */
  8. ENTER; /* everything created after here */
  9. SAVETMPS; /* ...is a temporary variable. */
  10. PUSHMARK(SP); /* remember the stack pointer */
  11. XPUSHs(sv_2mortal(newSViv(a))); /* push the base onto the stack */
  12. XPUSHs(sv_2mortal(newSViv(b))); /* push the exponent onto stack */
  13. PUTBACK; /* make local stack pointer global */
  14. call_pv("expo", G_SCALAR); /* call the function */
  15. SPAGAIN; /* refresh stack pointer */
  16. /* pop the return value from stack */
  17. printf ("%d to the %dth power is %d.\n", a, b, POPi);
  18. PUTBACK;
  19. FREETMPS; /* free that return value */
  20. LEAVE; /* ...and the XPUSHed "mortal" args.*/
  21. }
  22. int main (int argc, char **argv, char **env)
  23. {
  24. char *my_argv[] = { "", "power.pl" };
  25. PERL_SYS_INIT3(&argc,&argv,&env);
  26. my_perl = perl_alloc();
  27. perl_construct( my_perl );
  28. perl_parse(my_perl, NULL, 2, my_argv, (char **)NULL);
  29. PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
  30. perl_run(my_perl);
  31. PerlPower(3, 4); /*** Compute 3 ** 4 ***/
  32. perl_destruct(my_perl);
  33. perl_free(my_perl);
  34. PERL_SYS_TERM();
  35. }

Compile and run:

  1. % cc -o power power.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
  2. % power
  3. 3 to the 4th power is 81.

Maintaining a persistent interpreter

When developing interactive and/or potentially long-runningapplications, it's a good idea to maintain a persistent interpreterrather than allocating and constructing a new interpreter multipletimes. The major reason is speed: since Perl will only be loaded intomemory once.

However, you have to be more cautious with namespace and variablescoping when using a persistent interpreter. In previous exampleswe've been using global variables in the default package main. Weknew exactly what code would be run, and assumed we could avoidvariable collisions and outrageous symbol table growth.

Let's say your application is a server that will occasionally run Perlcode from some arbitrary file. Your server has no way of knowing whatcode it's going to run. Very dangerous.

If the file is pulled in by perl_parse(), compiled into a newlyconstructed interpreter, and subsequently cleaned out withperl_destruct() afterwards, you're shielded from most namespacetroubles.

One way to avoid namespace collisions in this scenario is to translatethe filename into a guaranteed-unique package name, and then compilethe code into that package using eval. In the examplebelow, each file will only be compiled once. Or, the applicationmight choose to clean out the symbol table associated with the fileafter it's no longer needed. Using call_argv in perlapi, We'llcall the subroutine Embed::Persistent::eval_file which lives in thefile persistent.pl and pass the filename and boolean cleanup/cacheflag as arguments.

Note that the process will continue to grow for each file that ituses. In addition, there might be AUTOLOADed subroutines and otherconditions that cause Perl's symbol table to grow. You might want toadd some logic that keeps track of the process size, or restartsitself after a certain number of requests, to ensure that memoryconsumption is minimized. You'll also want to scope your variableswith my whenever possible.

  1. package Embed::Persistent;
  2. #persistent.pl
  3. use strict;
  4. our %Cache;
  5. use Symbol qw(delete_package);
  6. sub valid_package_name {
  7. my($string) = @_;
  8. $string =~ s/([^A-Za-z0-9\/])/sprintf("_%2x",unpack("C",$1))/eg;
  9. # second pass only for words starting with a digit
  10. $string =~ s|/(\d)|sprintf("/_%2x",unpack("C",$1))|eg;
  11. # Dress it up as a real package name
  12. $string =~ s|/|::|g;
  13. return "Embed" . $string;
  14. }
  15. sub eval_file {
  16. my($filename, $delete) = @_;
  17. my $package = valid_package_name($filename);
  18. my $mtime = -M $filename;
  19. if(defined $Cache{$package}{mtime}
  20. &&
  21. $Cache{$package}{mtime} <= $mtime)
  22. {
  23. # we have compiled this subroutine already,
  24. # it has not been updated on disk, nothing left to do
  25. print STDERR "already compiled $package->handler\n";
  26. }
  27. else {
  28. local *FH;
  29. open FH, $filename or die "open '$filename' $!";
  30. local($/) = undef;
  31. my $sub = <FH>;
  32. close FH;
  33. #wrap the code into a subroutine inside our unique package
  34. my $eval = qq{package $package; sub handler { $sub; }};
  35. {
  36. # hide our variables within this block
  37. my($filename,$mtime,$package,$sub);
  38. eval $eval;
  39. }
  40. die $@ if $@;
  41. #cache it unless we're cleaning out each time
  42. $Cache{$package}{mtime} = $mtime unless $delete;
  43. }
  44. eval {$package->handler;};
  45. die $@ if $@;
  46. delete_package($package) if $delete;
  47. #take a look if you want
  48. #print Devel::Symdump->rnew($package)->as_string, $/;
  49. }
  50. 1;
  51. __END__
  52. /* persistent.c */
  53. #include <EXTERN.h>
  54. #include <perl.h>
  55. /* 1 = clean out filename's symbol table after each request, 0 = don't */
  56. #ifndef DO_CLEAN
  57. #define DO_CLEAN 0
  58. #endif
  59. #define BUFFER_SIZE 1024
  60. static PerlInterpreter *my_perl = NULL;
  61. int
  62. main(int argc, char **argv, char **env)
  63. {
  64. char *embedding[] = { "", "persistent.pl" };
  65. char *args[] = { "", DO_CLEAN, NULL };
  66. char filename[BUFFER_SIZE];
  67. int exitstatus = 0;
  68. PERL_SYS_INIT3(&argc,&argv,&env);
  69. if((my_perl = perl_alloc()) == NULL) {
  70. fprintf(stderr, "no memory!");
  71. exit(1);
  72. }
  73. perl_construct(my_perl);
  74. PL_origalen = 1; /* don't let $0 assignment update the proctitle or embedding[0] */
  75. exitstatus = perl_parse(my_perl, NULL, 2, embedding, NULL);
  76. PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
  77. if(!exitstatus) {
  78. exitstatus = perl_run(my_perl);
  79. while(printf("Enter file name: ") &&
  80. fgets(filename, BUFFER_SIZE, stdin)) {
  81. filename[strlen(filename)-1] = '\0'; /* strip \n */
  82. /* call the subroutine, passing it the filename as an argument */
  83. args[0] = filename;
  84. call_argv("Embed::Persistent::eval_file",
  85. G_DISCARD | G_EVAL, args);
  86. /* check $@ */
  87. if(SvTRUE(ERRSV))
  88. fprintf(stderr, "eval error: %s\n", SvPV_nolen(ERRSV));
  89. }
  90. }
  91. PL_perl_destruct_level = 0;
  92. perl_destruct(my_perl);
  93. perl_free(my_perl);
  94. PERL_SYS_TERM();
  95. exit(exitstatus);
  96. }

Now compile:

  1. % cc -o persistent persistent.c `perl -MExtUtils::Embed -e ccopts -e ldopts`

Here's an example script file:

  1. #test.pl
  2. my $string = "hello";
  3. foo($string);
  4. sub foo {
  5. print "foo says: @_\n";
  6. }

Now run:

  1. % persistent
  2. Enter file name: test.pl
  3. foo says: hello
  4. Enter file name: test.pl
  5. already compiled Embed::test_2epl->handler
  6. foo says: hello
  7. Enter file name: ^C

Execution of END blocks

Traditionally END blocks have been executed at the end of the perl_run.This causes problems for applications that never call perl_run. Sinceperl 5.7.2 you can specify PL_exit_flags |= PERL_EXIT_DESTRUCT_ENDto get the new behaviour. This also enables the running of END blocks ifthe perl_parse fails and perl_destruct will return the exit value.

$0 assignments

When a perl script assigns a value to $0 then the perl runtime willtry to make this value show up as the program name reported by "ps" byupdating the memory pointed to by the argv passed to perl_parse() andalso calling API functions like setproctitle() where available. Thisbehaviour might not be appropriate when embedding perl and can bedisabled by assigning the value 1 to the variable PL_origalenbefore perl_parse() is called.

The persistent.c example above is for instance likely to segfaultwhen $0 is assigned to if the PL_origalen = 1; assignment isremoved. This because perl will try to write to the read only memoryof the embedding[] strings.

Maintaining multiple interpreter instances

Some rare applications will need to create more than one interpreterduring a session. Such an application might sporadically decide torelease any resources associated with the interpreter.

The program must take care to ensure that this takes place beforethe next interpreter is constructed. By default, when perl is notbuilt with any special options, the global variablePL_perl_destruct_level is set to 0, since extra cleaning isn'tusually needed when a program only ever creates a single interpreterin its entire lifetime.

Setting PL_perl_destruct_level to 1 makes everything squeaky clean:

  1. while(1) {
  2. ...
  3. /* reset global variables here with PL_perl_destruct_level = 1 */
  4. PL_perl_destruct_level = 1;
  5. perl_construct(my_perl);
  6. ...
  7. /* clean and reset _everything_ during perl_destruct */
  8. PL_perl_destruct_level = 1;
  9. perl_destruct(my_perl);
  10. perl_free(my_perl);
  11. ...
  12. /* let's go do it again! */
  13. }

When perl_destruct() is called, the interpreter's syntax parse treeand symbol tables are cleaned up, and global variables are reset. Thesecond assignment to PL_perl_destruct_level is needed becauseperl_construct resets it to 0.

Now suppose we have more than one interpreter instance running at thesame time. This is feasible, but only if you used the Configure option-Dusemultiplicity or the options -Dusethreads -Duseithreads whenbuilding perl. By default, enabling one of these Configure optionssets the per-interpreter global variable PL_perl_destruct_level to1, so that thorough cleaning is automatic and interpreter variablesare initialized correctly. Even if you don't intend to run two ormore interpreters at the same time, but to run them sequentially, likein the above example, it is recommended to build perl with the-Dusemultiplicity option otherwise some interpreter variables maynot be initialized correctly between consecutive runs and yourapplication may crash.

See also Thread-aware system interfaces in perlxs.

Using -Dusethreads -Duseithreads rather than -Dusemultiplicityis more appropriate if you intend to run multiple interpretersconcurrently in different threads, because it enables support forlinking in the thread libraries of your system with the interpreter.

Let's give it a try:

  1. #include <EXTERN.h>
  2. #include <perl.h>
  3. /* we're going to embed two interpreters */
  4. #define SAY_HELLO "-e", "print qq(Hi, I'm $^X\n)"
  5. int main(int argc, char **argv, char **env)
  6. {
  7. PerlInterpreter *one_perl, *two_perl;
  8. char *one_args[] = { "one_perl", SAY_HELLO };
  9. char *two_args[] = { "two_perl", SAY_HELLO };
  10. PERL_SYS_INIT3(&argc,&argv,&env);
  11. one_perl = perl_alloc();
  12. two_perl = perl_alloc();
  13. PERL_SET_CONTEXT(one_perl);
  14. perl_construct(one_perl);
  15. PERL_SET_CONTEXT(two_perl);
  16. perl_construct(two_perl);
  17. PERL_SET_CONTEXT(one_perl);
  18. perl_parse(one_perl, NULL, 3, one_args, (char **)NULL);
  19. PERL_SET_CONTEXT(two_perl);
  20. perl_parse(two_perl, NULL, 3, two_args, (char **)NULL);
  21. PERL_SET_CONTEXT(one_perl);
  22. perl_run(one_perl);
  23. PERL_SET_CONTEXT(two_perl);
  24. perl_run(two_perl);
  25. PERL_SET_CONTEXT(one_perl);
  26. perl_destruct(one_perl);
  27. PERL_SET_CONTEXT(two_perl);
  28. perl_destruct(two_perl);
  29. PERL_SET_CONTEXT(one_perl);
  30. perl_free(one_perl);
  31. PERL_SET_CONTEXT(two_perl);
  32. perl_free(two_perl);
  33. PERL_SYS_TERM();
  34. }

Note the calls to PERL_SET_CONTEXT(). These are necessary to initializethe global state that tracks which interpreter is the "current" one onthe particular process or thread that may be running it. It shouldalways be used if you have more than one interpreter and are makingperl API calls on both interpreters in an interleaved fashion.

PERL_SET_CONTEXT(interp) should also be called whenever interp isused by a thread that did not create it (using either perl_alloc(), orthe more esoteric perl_clone()).

Compile as usual:

  1. % cc -o multiplicity multiplicity.c `perl -MExtUtils::Embed -e ccopts -e ldopts`

Run it, Run it:

  1. % multiplicity
  2. Hi, I'm one_perl
  3. Hi, I'm two_perl

Using Perl modules, which themselves use C libraries, from your C program

If you've played with the examples above and tried to embed a scriptthat use()s a Perl module (such as Socket) which itself uses a C or C++ library,this probably happened:

  1. Can't load module Socket, dynamic loading not available in this perl.
  2. (You may need to build a new perl executable which either supports
  3. dynamic loading or has the Socket module statically linked into it.)

What's wrong?

Your interpreter doesn't know how to communicate with these extensionson its own. A little glue will help. Up until now you've beencalling perl_parse(), handing it NULL for the second argument:

  1. perl_parse(my_perl, NULL, argc, my_argv, NULL);

That's where the glue code can be inserted to create the initial contact betweenPerl and linked C/C++ routines. Let's take a look some pieces of perlmain.cto see how Perl does this:

  1. static void xs_init (pTHX);
  2. EXTERN_C void boot_DynaLoader (pTHX_ CV* cv);
  3. EXTERN_C void boot_Socket (pTHX_ CV* cv);
  4. EXTERN_C void
  5. xs_init(pTHX)
  6. {
  7. char *file = __FILE__;
  8. /* DynaLoader is a special case */
  9. newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
  10. newXS("Socket::bootstrap", boot_Socket, file);
  11. }

Simply put: for each extension linked with your Perl executable(determined during its initial configuration on yourcomputer or when adding a new extension),a Perl subroutine is created to incorporate the extension'sroutines. Normally, that subroutine is namedModule::bootstrap() and is invoked when you say use Module. Inturn, this hooks into an XSUB, boot_Module, which creates a Perlcounterpart for each of the extension's XSUBs. Don't worry about thispart; leave that to the xsubpp and extension authors. If yourextension is dynamically loaded, DynaLoader creates Module::bootstrap()for you on the fly. In fact, if you have a working DynaLoader then thereis rarely any need to link in any other extensions statically.

Once you have this code, slap it into the second argument of perl_parse():

  1. perl_parse(my_perl, xs_init, argc, my_argv, NULL);

Then compile:

  1. % cc -o interp interp.c `perl -MExtUtils::Embed -e ccopts -e ldopts`
  2. % interp
  3. use Socket;
  4. use SomeDynamicallyLoadedModule;
  5. print "Now I can use extensions!\n"'

ExtUtils::Embed can also automate writing the xs_init glue code.

  1. % perl -MExtUtils::Embed -e xsinit -- -o perlxsi.c
  2. % cc -c perlxsi.c `perl -MExtUtils::Embed -e ccopts`
  3. % cc -c interp.c `perl -MExtUtils::Embed -e ccopts`
  4. % cc -o interp perlxsi.o interp.o `perl -MExtUtils::Embed -e ldopts`

Consult perlxs, perlguts, and perlapi for more details.

Hiding Perl_

If you completely hide the short forms of the Perl public API,add -DPERL_NO_SHORT_NAMES to the compilation flags. This means thatfor example instead of writing

  1. warn("%d bottles of beer on the wall", bottlecount);

you will have to write the explicit full form

  1. Perl_warn(aTHX_ "%d bottles of beer on the wall", bottlecount);

(See Background and PERL_IMPLICIT_CONTEXT in perlguts for the explanationof the aTHX_. ) Hiding the short forms is very useful for avoidingall sorts of nasty (C preprocessor or otherwise) conflicts with othersoftware packages (Perl defines about 2400 APIs with these short names,take or leave few hundred, so there certainly is room for conflict.)

MORAL

You can sometimes write faster code in C, butyou can always write code faster in Perl. Because you can useeach from the other, combine them as you wish.

AUTHOR

Jon Orwant <[email protected]> and Doug MacEachern<[email protected]>, with small contributions from Tim Bunce, TomChristiansen, Guy Decoux, Hallvard Furuseth, Dov Grobgeld, and IlyaZakharevich.

Doug MacEachern has an article on embedding in Volume 1, Issue 4 ofThe Perl Journal ( http://www.tpj.com/ ). Doug is also the developer of themost widely-used Perl embedding: the mod_perl system(perl.apache.org), which embeds Perl in the Apache web server.Oracle, Binary Evolution, ActiveState, and Ben Sugars's nsapi_perlhave used this model for Oracle, Netscape and Internet InformationServer Perl plugins.

COPYRIGHT

Copyright (C) 1995, 1996, 1997, 1998 Doug MacEachern and Jon Orwant. AllRights Reserved.

This document may be distributed under the same terms as Perl itself.

 
Source : perldoc.perl.org - Official documentation for the Perl programming language
Site maintained by Jon Allen (JJ)     See the project page for more details
Documentation maintained by the Perl 5 Porters
(Sebelumnya) How to submit bug reports on PerlGuts of Perl debugging (Berikutnya)