Debian Codegraph @

upstream/debian-unstable/man1/perlembed.1

raw
1 .\" -*- mode: troff; coding: utf-8 -*-
2 .\" Automatically generated by Pod::Man 5.0102 (Pod::Simple 3.45)
3 .\"
4 .\" Standard preamble:
5 .\" ========================================================================
6 .de Sp \" Vertical space (when we can't use .PP)
7 .if t .sp .5v
8 .if n .sp
9 ..
10 .de Vb \" Begin verbatim text
11 .ft CW
12 .nf
13 .ne \\$1
14 ..
15 .de Ve \" End verbatim text
16 .ft R
17 .fi
18 ..
19 .\" \*(C` and \*(C' are quotes in nroff, nothing in troff, for use with C<>.
20 .ie n \{\
21 . ds C` ""
22 . ds C' ""
23 'br\}
24 .el\{\
25 . ds C`
26 . ds C'
27 'br\}
28 .\"
29 .\" Escape single quotes in literal strings from groff's Unicode transform.
30 .ie \n(.g .ds Aq \(aq
31 .el .ds Aq '
32 .\"
33 .\" If the F register is >0, we'll generate index entries on stderr for
34 .\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index
35 .\" entries marked with X<> in POD. Of course, you'll have to process the
36 .\" output yourself in some meaningful fashion.
37 .\"
38 .\" Avoid warning from groff about undefined register 'F'.
39 .de IX
40 ..
41 .nr rF 0
42 .if \n(.g .if rF .nr rF 1
43 .if (\n(rF:(\n(.g==0)) \{\
44 . if \nF \{\
45 . de IX
46 . tm Index:\\$1\t\\n%\t"\\$2"
47 ..
48 . if !\nF==2 \{\
49 . nr % 0
50 . nr F 2
51 . \}
52 . \}
53 .\}
54 .rr rF
55 .\" ========================================================================
56 .\"
57 .IX Title "PERLEMBED 1"
58 .TH PERLEMBED 1 2025-11-16 "perl v5.40.1" "Perl Programmers Reference Guide"
59 .\" For nroff, turn off justification. Always turn off hyphenation; it makes
60 .\" way too many mistakes in technical documents.
61 .if n .ad l
62 .nh
63 .SH NAME
64 perlembed \- how to embed perl in your C program
65 .SH DESCRIPTION
66 .IX Header "DESCRIPTION"
67 .SS PREAMBLE
68 .IX Subsection "PREAMBLE"
69 Do you want to:
70 .IP "\fBUse C from Perl?\fR" 5
71 .IX Item "Use C from Perl?"
72 Read perlxstut, perlxs, perlclib, h2xs, perlguts, and perlapi.
73 .IP "\fBUse a Unix program from Perl?\fR" 5
74 .IX Item "Use a Unix program from Perl?"
75 Read about back-quotes and about \f(CW\*(C`system\*(C'\fR and \f(CW\*(C`exec\*(C'\fR in perlfunc.
76 .IP "\fBUse Perl from Perl?\fR" 5
77 .IX Item "Use Perl from Perl?"
78 Read about "do" in perlfunc and "eval" in perlfunc and "require" in perlfunc
79 and "use" in perlfunc.
80 .IP "\fBUse C from C?\fR" 5
81 .IX Item "Use C from C?"
82 Rethink your design.
83 .IP "\fBUse Perl from C?\fR" 5
84 .IX Item "Use Perl from C?"
85 Read on...
86 .SS ROADMAP
87 .IX Subsection "ROADMAP"
88 .IP \(bu 5
89 Compiling your C program
90 .IP \(bu 5
91 Adding a Perl interpreter to your C program
92 .IP \(bu 5
93 Calling a Perl subroutine from your C program
94 .IP \(bu 5
95 Evaluating a Perl statement from your C program
96 .IP \(bu 5
97 Performing Perl pattern matches and substitutions from your C program
98 .IP \(bu 5
99 Fiddling with the Perl stack from your C program
100 .IP \(bu 5
101 Maintaining a persistent interpreter
102 .IP \(bu 5
103 Maintaining multiple interpreter instances
104 .IP \(bu 5
105 Using Perl modules, which themselves use C libraries, from your C program
106 .IP \(bu 5
107 Embedding Perl under Win32
108 .SS "Compiling your C program"
109 .IX Subsection "Compiling your C program"
110 If you have trouble compiling the scripts in this documentation,
111 you're not alone. The cardinal rule: COMPILE THE PROGRAMS IN EXACTLY
112 THE SAME WAY THAT YOUR PERL WAS COMPILED. (Sorry for yelling.)
113 .PP
114 Also, every C program that uses Perl must link in the \fIperl library\fR.
115 What's that, you ask? Perl is itself written in C; the perl library
116 is the collection of compiled C programs that were used to create your
117 perl executable (\fI/usr/bin/perl\fR or equivalent). (Corollary: you
118 can't use Perl from your C program unless Perl has been compiled on
119 your machine, or installed properly\-\-that's why you shouldn't blithely
120 copy Perl executables from machine to machine without also copying the
121 \&\fIlib\fR directory.)
122 .PP
123 When you use Perl from C, your C program will\-\-usually\-\-allocate,
124 "run", and deallocate a \fIPerlInterpreter\fR object, which is defined by
125 the perl library.
126 .PP
127 If your copy of Perl is recent enough to contain this documentation
128 (version 5.002 or later), then the perl library (and \fIEXTERN.h\fR and
129 \&\fIperl.h\fR, which you'll also need) will reside in a directory
130 that looks like this:
131 .PP
132 .Vb 1
133 \& /usr/local/lib/perl5/your_architecture_here/CORE
134 .Ve
135 .PP
136 or perhaps just
137 .PP
138 .Vb 1
139 \& /usr/local/lib/perl5/CORE
140 .Ve
141 .PP
142 or maybe something like
143 .PP
144 .Vb 1
145 \& /usr/opt/perl5/CORE
146 .Ve
147 .PP
148 Execute this statement for a hint about where to find CORE:
149 .PP
150 .Vb 1
151 \& perl \-MConfig \-e \*(Aqprint $Config{archlib}\*(Aq
152 .Ve
153 .PP
154 Here's how you'd compile the example in the next section,
155 "Adding a Perl interpreter to your C program", on my Linux box:
156 .PP
157 .Vb 4
158 \& % gcc \-O2 \-Dbool=char \-DHAS_BOOL \-I/usr/local/include
159 \& \-I/usr/local/lib/perl5/i586\-linux/5.003/CORE
160 \& \-L/usr/local/lib/perl5/i586\-linux/5.003/CORE
161 \& \-o interp interp.c \-lperl \-lm
162 .Ve
163 .PP
164 (That's all one line.) On my DEC Alpha running old 5.003_05, the
165 incantation is a bit different:
166 .PP
167 .Vb 4
168 \& % cc \-O2 \-Olimit 2900 \-I/usr/local/include
169 \& \-I/usr/local/lib/perl5/alpha\-dec_osf/5.00305/CORE
170 \& \-L/usr/local/lib/perl5/alpha\-dec_osf/5.00305/CORE \-L/usr/local/lib
171 \& \-D_\|_LANGUAGE_C_\|_ \-D_NO_PROTO \-o interp interp.c \-lperl \-lm
172 .Ve
173 .PP
174 How can you figure out what to add? Assuming your Perl is post\-5.001,
175 execute a \f(CW\*(C`perl \-V\*(C'\fR command and pay special attention to the "cc" and
176 "ccflags" information.
177 .PP
178 You'll have to choose the appropriate compiler (\fIcc\fR, \fIgcc\fR, et al.) for
179 your machine: \f(CW\*(C`perl \-MConfig \-e \*(Aqprint $Config{cc}\*(Aq\*(C'\fR will tell you what
180 to use.
181 .PP
182 You'll also have to choose the appropriate library directory
183 (\fI/usr/local/lib/...\fR) for your machine. If your compiler complains
184 that certain functions are undefined, or that it can't locate
185 \&\fI\-lperl\fR, then you need to change the path following the \f(CW\*(C`\-L\*(C'\fR. If it
186 complains that it can't find \fIEXTERN.h\fR and \fIperl.h\fR, you need to
187 change the path following the \f(CW\*(C`\-I\*(C'\fR.
188 .PP
189 You may have to add extra libraries as well. Which ones?
190 Perhaps those printed by
191 .PP
192 .Vb 1
193 \& perl \-MConfig \-e \*(Aqprint $Config{libs}\*(Aq
194 .Ve
195 .PP
196 Provided your perl binary was properly configured and installed the
197 \&\fBExtUtils::Embed\fR module will determine all of this information for
198 you:
199 .PP
200 .Vb 1
201 \& % cc \-o interp interp.c \`perl \-MExtUtils::Embed \-e ccopts \-e ldopts\`
202 .Ve
203 .PP
204 If the \fBExtUtils::Embed\fR module isn't part of your Perl distribution,
205 you can retrieve it from
206 <https://metacpan.org/pod/ExtUtils::Embed>
207 (If this documentation came from your Perl distribution, then you're
208 running 5.004 or better and you already have it.)
209 .PP
210 The \fBExtUtils::Embed\fR kit on CPAN also contains all source code for
211 the examples in this document, tests, additional examples and other
212 information you may find useful.
213 .SS "Adding a Perl interpreter to your C program"
214 .IX Subsection "Adding a Perl interpreter to your C program"
215 In a sense, perl (the C program) is a good example of embedding Perl
216 (the language), so I'll demonstrate embedding with \fIminiperlmain.c\fR,
217 included in the source distribution. Here's a bastardized, non-portable
218 version of \fIminiperlmain.c\fR containing the essentials of embedding:
219 .PP
220 .Vb 2
221 \& #include <EXTERN.h> /* from the Perl distribution */
222 \& #include <perl.h> /* from the Perl distribution */
223 \&
224 \& static PerlInterpreter *my_perl; /*** The Perl interpreter ***/
225 \&
226 \& int main(int argc, char **argv, char **env)
227 \& {
228 \& PERL_SYS_INIT3(&argc,&argv,&env);
229 \& my_perl = perl_alloc();
230 \& perl_construct(my_perl);
231 \& PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
232 \& perl_parse(my_perl, NULL, argc, argv, (char **)NULL);
233 \& perl_run(my_perl);
234 \& perl_destruct(my_perl);
235 \& perl_free(my_perl);
236 \& PERL_SYS_TERM();
237 \& exit(EXIT_SUCCESS);
238 \& }
239 .Ve
240 .PP
241 Notice that we don't use the \f(CW\*(C`env\*(C'\fR pointer. Normally handed to
242 \&\f(CW\*(C`perl_parse\*(C'\fR as its final argument, \f(CW\*(C`env\*(C'\fR here is replaced by
243 \&\f(CW\*(C`NULL\*(C'\fR, which means that the current environment will be used.
244 .PP
245 The macros \fBPERL_SYS_INIT3()\fR and \fBPERL_SYS_TERM()\fR provide system-specific
246 tune up of the C runtime environment necessary to run Perl interpreters;
247 they should only be called once regardless of how many interpreters you
248 create or destroy. Call \fBPERL_SYS_INIT3()\fR before you create your first
249 interpreter, and \fBPERL_SYS_TERM()\fR after you free your last interpreter.
250 .PP
251 Since \fBPERL_SYS_INIT3()\fR may change \f(CW\*(C`env\*(C'\fR, it may be more appropriate to
252 provide \f(CW\*(C`env\*(C'\fR as an argument to \fBperl_parse()\fR.
253 .PP
254 Also notice that no matter what arguments you pass to \fBperl_parse()\fR,
255 \&\fBPERL_SYS_INIT3()\fR must be invoked on the C \fBmain()\fR argc, argv and env and
256 only once.
257 .PP
258 Mind that argv[argc] must be NULL, same as those passed to a main
259 function in C.
260 .PP
261 Now compile this program (I'll call it \fIinterp.c\fR) into an executable:
262 .PP
263 .Vb 1
264 \& % cc \-o interp interp.c \`perl \-MExtUtils::Embed \-e ccopts \-e ldopts\`
265 .Ve
266 .PP
267 After a successful compilation, you'll be able to use \fIinterp\fR just
268 like perl itself:
269 .PP
270 .Vb 6
271 \& % interp
272 \& print "Pretty Good Perl \en";
273 \& print "10890 \- 9801 is ", 10890 \- 9801;
274 \& <CTRL\-D>
275 \& Pretty Good Perl
276 \& 10890 \- 9801 is 1089
277 .Ve
278 .PP
279 or
280 .PP
281 .Vb 2
282 \& % interp \-e \*(Aqprintf("%x", 3735928559)\*(Aq
283 \& deadbeef
284 .Ve
285 .PP
286 You can also read and execute Perl statements from a file while in the
287 midst of your C program, by placing the filename in \fIargv[1]\fR before
288 calling \fIperl_run\fR.
289 .SS "Calling a Perl subroutine from your C program"
290 .IX Subsection "Calling a Perl subroutine from your C program"
291 To call individual Perl subroutines, you can use any of the \fBcall_*\fR
292 functions documented in perlcall.
293 In this example we'll use \f(CW\*(C`call_argv\*(C'\fR.
294 .PP
295 That's shown below, in a program I'll call \fIshowtime.c\fR.
296 .PP
297 .Vb 2
298 \& #include <EXTERN.h>
299 \& #include <perl.h>
300 \&
301 \& static PerlInterpreter *my_perl;
302 \&
303 \& int main(int argc, char **argv, char **env)
304 \& {
305 \& char *args[] = { NULL };
306 \& PERL_SYS_INIT3(&argc,&argv,&env);
307 \& my_perl = perl_alloc();
308 \& perl_construct(my_perl);
309 \&
310 \& perl_parse(my_perl, NULL, argc, argv, NULL);
311 \& PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
312 \&
313 \& /*** skipping perl_run() ***/
314 \&
315 \& call_argv("showtime", G_DISCARD | G_NOARGS, args);
316 \&
317 \& perl_destruct(my_perl);
318 \& perl_free(my_perl);
319 \& PERL_SYS_TERM();
320 \& exit(EXIT_SUCCESS);
321 \& }
322 .Ve
323 .PP
324 where \fIshowtime\fR is a Perl subroutine that takes no arguments (that's the
325 \&\fIG_NOARGS\fR) and for which I'll ignore the return value (that's the
326 \&\fIG_DISCARD\fR). Those flags, and others, are discussed in perlcall.
327 .PP
328 I'll define the \fIshowtime\fR subroutine in a file called \fIshowtime.pl\fR:
329 .PP
330 .Vb 1
331 \& print "I shan\*(Aqt be printed.";
332 \&
333 \& sub showtime {
334 \& print time;
335 \& }
336 .Ve
337 .PP
338 Simple enough. Now compile and run:
339 .PP
340 .Vb 4
341 \& % cc \-o showtime showtime.c \e
342 \& \`perl \-MExtUtils::Embed \-e ccopts \-e ldopts\`
343 \& % showtime showtime.pl
344 \& 818284590
345 .Ve
346 .PP
347 yielding the number of seconds that elapsed between January 1, 1970
348 (the beginning of the Unix epoch), and the moment I began writing this
349 sentence.
350 .PP
351 In this particular case we don't have to call \fIperl_run\fR, as we set
352 the PL_exit_flag PERL_EXIT_DESTRUCT_END which executes END blocks in
353 perl_destruct.
354 .PP
355 If you want to pass arguments to the Perl subroutine, you can add
356 strings to the \f(CW\*(C`NULL\*(C'\fR\-terminated \f(CW\*(C`args\*(C'\fR list passed to
357 \&\fIcall_argv\fR. For other data types, or to examine return values,
358 you'll need to manipulate the Perl stack. That's demonstrated in
359 "Fiddling with the Perl stack from your C program".
360 .SS "Evaluating a Perl statement from your C program"
361 .IX Subsection "Evaluating a Perl statement from your C program"
362 Perl provides two API functions to evaluate pieces of Perl code.
363 These are "eval_sv" in perlapi and "eval_pv" in perlapi.
364 .PP
365 Arguably, these are the only routines you'll ever need to execute
366 snippets of Perl code from within your C program. Your code can be as
367 long as you wish; it can contain multiple statements; it can employ
368 "use" in perlfunc, "require" in perlfunc, and "do" in perlfunc to
369 include external Perl files.
370 .PP
371 \&\fIeval_pv\fR lets us evaluate individual Perl strings, and then
372 extract variables for coercion into C types. The following program,
373 \&\fIstring.c\fR, executes three Perl strings, extracting an \f(CW\*(C`int\*(C'\fR from
374 the first, a \f(CW\*(C`float\*(C'\fR from the second, and a \f(CW\*(C`char *\*(C'\fR from the third.
375 .PP
376 .Vb 2
377 \& #include <EXTERN.h>
378 \& #include <perl.h>
379 \&
380 \& static PerlInterpreter *my_perl;
381 \&
382 \& main (int argc, char **argv, char **env)
383 \& {
384 \& char *embedding[] = { "", "\-e", "0", NULL };
385 \&
386 \& PERL_SYS_INIT3(&argc,&argv,&env);
387 \& my_perl = perl_alloc();
388 \& perl_construct( my_perl );
389 \&
390 \& perl_parse(my_perl, NULL, 3, embedding, NULL);
391 \& PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
392 \& perl_run(my_perl);
393 \&
394 \& /** Treat $x as an integer **/
395 \& eval_pv("$x = 3; $x **= 2", TRUE);
396 \& printf("x = %d\en", SvIV(get_sv("x", 0)));
397 \&
398 \& /** Treat $x as a float **/
399 \& eval_pv("$x = 3.14; $x **= 2", TRUE);
400 \& printf("x = %f\en", SvNV(get_sv("x", 0)));
401 \&
402 \& /** Treat $x as a string **/
403 \& eval_pv(
404 \& "$x = \*(AqrekcaH lreP rehtonA tsuJ\*(Aq; $x = reverse($x);", TRUE);
405 \& printf("x = %s\en", SvPV_nolen(get_sv("x", 0)));
406 \&
407 \& perl_destruct(my_perl);
408 \& perl_free(my_perl);
409 \& PERL_SYS_TERM();
410 \& }
411 .Ve
412 .PP
413 All of those strange functions with \fIsv\fR in their names help convert Perl
414 scalars to C types. They're described in perlguts and perlapi.
415 .PP
416 If you compile and run \fIstring.c\fR, you'll see the results of using
417 \&\fR\f(BISvIV()\fR\fI\fR to create an \f(CW\*(C`int\*(C'\fR, \fI\fR\f(BISvNV()\fR\fI\fR to create a \f(CW\*(C`float\*(C'\fR, and
418 \&\fI\fR\f(BISvPV()\fR\fI\fR to create a string:
419 .PP
420 .Vb 3
421 \& a = 9
422 \& a = 9.859600
423 \& a = Just Another Perl Hacker
424 .Ve
425 .PP
426 In the example above, we've created a global variable to temporarily
427 store the computed value of our eval'ed expression. It is also
428 possible and in most cases a better strategy to fetch the return value
429 from \fR\f(BIeval_pv()\fR\fI\fR instead. Example:
430 .PP
431 .Vb 4
432 \& ...
433 \& SV *val = eval_pv("reverse \*(AqrekcaH lreP rehtonA tsuJ\*(Aq", TRUE);
434 \& printf("%s\en", SvPV_nolen(val));
435 \& ...
436 .Ve
437 .PP
438 This way, we avoid namespace pollution by not creating global
439 variables and we've simplified our code as well.
440 .SS "Performing Perl pattern matches and substitutions from your C program"
441 .IX Subsection "Performing Perl pattern matches and substitutions from your C program"
442 The \fR\f(BIeval_sv()\fR\fI\fR function lets us evaluate strings of Perl code, so we can
443 define some functions that use it to "specialize" in matches and
444 substitutions: \fI\fR\f(BImatch()\fR\fI\fR, \fI\fR\f(BIsubstitute()\fR\fI\fR, and \fI\fR\f(BImatches()\fR\fI\fR.
445 .PP
446 .Vb 1
447 \& I32 match(SV *string, char *pattern);
448 .Ve
449 .PP
450 Given a string and a pattern (e.g., \f(CW\*(C`m/clasp/\*(C'\fR or \f(CW\*(C`/\eb\ew*\eb/\*(C'\fR, which
451 in your C program might appear as "/\e\eb\e\ew*\e\eb/"), \fBmatch()\fR
452 returns 1 if the string matches the pattern and 0 otherwise.
453 .PP
454 .Vb 1
455 \& int substitute(SV **string, char *pattern);
456 .Ve
457 .PP
458 Given a pointer to an \f(CW\*(C`SV\*(C'\fR and an \f(CW\*(C`=~\*(C'\fR operation (e.g.,
459 \&\f(CW\*(C`s/bob/robert/g\*(C'\fR or \f(CW\*(C`tr[A\-Z][a\-z]\*(C'\fR), \fBsubstitute()\fR modifies the string
460 within the \f(CW\*(C`SV\*(C'\fR as according to the operation, returning the number of
461 substitutions made.
462 .PP
463 .Vb 1
464 \& SSize_t matches(SV *string, char *pattern, AV **matches);
465 .Ve
466 .PP
467 Given an \f(CW\*(C`SV\*(C'\fR, a pattern, and a pointer to an empty \f(CW\*(C`AV\*(C'\fR,
468 \&\fBmatches()\fR evaluates \f(CW\*(C`$string =~ $pattern\*(C'\fR in a list context, and
469 fills in \fImatches\fR with the array elements, returning the number of matches
470 found.
471 .PP
472 Here's a sample program, \fImatch.c\fR, that uses all three (long lines have
473 been wrapped here):
474 .PP
475 .Vb 2
476 \& #include <EXTERN.h>
477 \& #include <perl.h>
478 \&
479 \& static PerlInterpreter *my_perl;
480 \&
481 \& /** my_eval_sv(code, error_check)
482 \& ** kinda like eval_sv(),
483 \& ** but we pop the return value off the stack
484 \& **/
485 \& SV* my_eval_sv(SV *sv, I32 croak_on_error)
486 \& {
487 \& dSP;
488 \& SV* retval;
489 \&
490 \&
491 \& PUSHMARK(SP);
492 \& eval_sv(sv, G_SCALAR);
493 \&
494 \& SPAGAIN;
495 \& retval = POPs;
496 \& PUTBACK;
497 \&
498 \& if (croak_on_error && SvTRUE(ERRSV))
499 \& croak_sv(ERRSV);
500 \&
501 \& return retval;
502 \& }
503 \&
504 \& /** match(string, pattern)
505 \& **
506 \& ** Used for matches in a scalar context.
507 \& **
508 \& ** Returns 1 if the match was successful; 0 otherwise.
509 \& **/
510 \&
511 \& I32 match(SV *string, char *pattern)
512 \& {
513 \& SV *command = newSV(0), *retval;
514 \&
515 \& sv_setpvf(command, "my $string = \*(Aq%s\*(Aq; $string =~ %s",
516 \& SvPV_nolen(string), pattern);
517 \&
518 \& retval = my_eval_sv(command, TRUE);
519 \& SvREFCNT_dec(command);
520 \&
521 \& return SvIV(retval);
522 \& }
523 \&
524 \& /** substitute(string, pattern)
525 \& **
526 \& ** Used for =~ operations that
527 \& ** modify their left\-hand side (s/// and tr///)
528 \& **
529 \& ** Returns the number of successful matches, and
530 \& ** modifies the input string if there were any.
531 \& **/
532 \&
533 \& I32 substitute(SV **string, char *pattern)
534 \& {
535 \& SV *command = newSV(0), *retval;
536 \&
537 \& sv_setpvf(command, "$string = \*(Aq%s\*(Aq; ($string =~ %s)",
538 \& SvPV_nolen(*string), pattern);
539 \&
540 \& retval = my_eval_sv(command, TRUE);
541 \& SvREFCNT_dec(command);
542 \&
543 \& *string = get_sv("string", 0);
544 \& return SvIV(retval);
545 \& }
546 \&
547 \& /** matches(string, pattern, matches)
548 \& **
549 \& ** Used for matches in a list context.
550 \& **
551 \& ** Returns the number of matches,
552 \& ** and fills in **matches with the matching substrings
553 \& **/
554 \&
555 \& SSize_t matches(SV *string, char *pattern, AV **match_list)
556 \& {
557 \& SV *command = newSV(0);
558 \& SSize_t num_matches;
559 \&
560 \& sv_setpvf(command, "my $string = \*(Aq%s\*(Aq; @array = ($string =~ %s)",
561 \& SvPV_nolen(string), pattern);
562 \&
563 \& my_eval_sv(command, TRUE);
564 \& SvREFCNT_dec(command);
565 \&
566 \& *match_list = get_av("array", 0);
567 \& num_matches = av_top_index(*match_list) + 1;
568 \&
569 \& return num_matches;
570 \& }
571 \&
572 \& main (int argc, char **argv, char **env)
573 \& {
574 \& char *embedding[] = { "", "\-e", "0", NULL };
575 \& AV *match_list;
576 \& I32 num_matches, i;
577 \& SV *text;
578 \&
579 \& PERL_SYS_INIT3(&argc,&argv,&env);
580 \& my_perl = perl_alloc();
581 \& perl_construct(my_perl);
582 \& perl_parse(my_perl, NULL, 3, embedding, NULL);
583 \& PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
584 \&
585 \& text = newSV(0);
586 \& sv_setpv(text, "When he is at a convenience store and the "
587 \& "bill comes to some amount like 76 cents, Maynard is "
588 \& "aware that there is something he *should* do, something "
589 \& "that will enable him to get back a quarter, but he has "
590 \& "no idea *what*. He fumbles through his red squeezey "
591 \& "changepurse and gives the boy three extra pennies with "
592 \& "his dollar, hoping that he might luck into the correct "
593 \& "amount. The boy gives him back two of his own pennies "
594 \& "and then the big shiny quarter that is his prize. "
595 \& "\-RICHH");
596 \&
597 \& if (match(text, "m/quarter/")) /** Does text contain \*(Aqquarter\*(Aq? **/
598 \& printf("match: Text contains the word \*(Aqquarter\*(Aq.\en\en");
599 \& else
600 \& printf("match: Text doesn\*(Aqt contain the word \*(Aqquarter\*(Aq.\en\en");
601 \&
602 \& if (match(text, "m/eighth/")) /** Does text contain \*(Aqeighth\*(Aq? **/
603 \& printf("match: Text contains the word \*(Aqeighth\*(Aq.\en\en");
604 \& else
605 \& printf("match: Text doesn\*(Aqt contain the word \*(Aqeighth\*(Aq.\en\en");
606 \&
607 \& /** Match all occurrences of /wi../ **/
608 \& num_matches = matches(text, "m/(wi..)/g", &match_list);
609 \& printf("matches: m/(wi..)/g found %d matches...\en", num_matches);
610 \&
611 \& for (i = 0; i < num_matches; i++)
612 \& printf("match: %s\en",
613 \& SvPV_nolen(*av_fetch(match_list, i, FALSE)));
614 \& printf("\en");
615 \&
616 \& /** Remove all vowels from text **/
617 \& num_matches = substitute(&text, "s/[aeiou]//gi");
618 \& if (num_matches) {
619 \& printf("substitute: s/[aeiou]//gi...%lu substitutions made.\en",
620 \& (unsigned long)num_matches);
621 \& printf("Now text is: %s\en\en", SvPV_nolen(text));
622 \& }
623 \&
624 \& /** Attempt a substitution **/
625 \& if (!substitute(&text, "s/Perl/C/")) {
626 \& printf("substitute: s/Perl/C...No substitution made.\en\en");
627 \& }
628 \&
629 \& SvREFCNT_dec(text);
630 \& PL_perl_destruct_level = 1;
631 \& perl_destruct(my_perl);
632 \& perl_free(my_perl);
633 \& PERL_SYS_TERM();
634 \& }
635 .Ve
636 .PP
637 which produces the output (again, long lines have been wrapped here)
638 .PP
639 .Vb 1
640 \& match: Text contains the word \*(Aqquarter\*(Aq.
641 \&
642 \& match: Text doesn\*(Aqt contain the word \*(Aqeighth\*(Aq.
643 \&
644 \& matches: m/(wi..)/g found 2 matches...
645 \& match: will
646 \& match: with
647 \&
648 \& substitute: s/[aeiou]//gi...139 substitutions made.
649 \& Now text is: Whn h s t cnvnnc str nd th bll cms t sm mnt lk 76 cnts,
650 \& Mynrd s wr tht thr s smthng h *shld* d, smthng tht wll nbl hm t gt
651 \& bck qrtr, bt h hs n d *wht*. H fmbls thrgh hs rd sqzy chngprs nd
652 \& gvs th by thr xtr pnns wth hs dllr, hpng tht h mght lck nt th crrct
653 \& mnt. Th by gvs hm bck tw f hs wn pnns nd thn th bg shny qrtr tht s
654 \& hs prz. \-RCHH
655 \&
656 \& substitute: s/Perl/C...No substitution made.
657 .Ve
658 .SS "Fiddling with the Perl stack from your C program"
659 .IX Subsection "Fiddling with the Perl stack from your C program"
660 When trying to explain stacks, most computer science textbooks mumble
661 something about spring-loaded columns of cafeteria plates: the last
662 thing you pushed on the stack is the first thing you pop off. That'll
663 do for our purposes: your C program will push some arguments onto "the Perl
664 stack", shut its eyes while some magic happens, and then pop the
665 results\-\-the return value of your Perl subroutine\-\-off the stack.
666 .PP
667 First you'll need to know how to convert between C types and Perl
668 types, with \fBnewSViv()\fR and \fBsv_setnv()\fR and \fBnewAV()\fR and all their
669 friends. They're described in perlguts and perlapi.
670 .PP
671 Then you'll need to know how to manipulate the Perl stack. That's
672 described in perlcall.
673 .PP
674 Once you've understood those, embedding Perl in C is easy.
675 .PP
676 Because C has no builtin function for integer exponentiation, let's
677 make Perl's ** operator available to it (this is less useful than it
678 sounds, because Perl implements ** with C's \fR\f(BIpow()\fR\fI\fR function). First
679 I'll create a stub exponentiation function in \fIpower.pl\fR:
680 .PP
681 .Vb 4
682 \& sub expo {
683 \& my ($x, $y) = @_;
684 \& return $x ** $y;
685 \& }
686 .Ve
687 .PP
688 Now I'll create a C program, \fIpower.c\fR, with a function
689 \&\fR\f(BIPerlPower()\fR\fI\fR that contains all the perlguts necessary to push the
690 two arguments into \fI\fR\f(BIexpo()\fR\fI\fR and to pop the return value out. Take a
691 deep breath...
692 .PP
693 .Vb 2
694 \& #include <EXTERN.h>
695 \& #include <perl.h>
696 \&
697 \& static PerlInterpreter *my_perl;
698 \&
699 \& static void
700 \& PerlPower(int a, int b)
701 \& {
702 \& dSP; /* initialize stack pointer */
703 \& ENTER; /* everything created after here */
704 \& SAVETMPS; /* ...is a temporary variable. */
705 \& PUSHMARK(SP); /* remember the stack pointer */
706 \& XPUSHs(sv_2mortal(newSViv(a))); /* push the base onto the stack */
707 \& XPUSHs(sv_2mortal(newSViv(b))); /* push the exponent onto stack */
708 \& PUTBACK; /* make local stack pointer global */
709 \& call_pv("expo", G_SCALAR); /* call the function */
710 \& SPAGAIN; /* refresh stack pointer */
711 \& /* pop the return value from stack */
712 \& printf ("%d to the %dth power is %d.\en", a, b, POPi);
713 \& PUTBACK;
714 \& FREETMPS; /* free that return value */
715 \& LEAVE; /* ...and the XPUSHed "mortal" args.*/
716 \& }
717 \&
718 \& int main (int argc, char **argv, char **env)
719 \& {
720 \& char *my_argv[] = { "", "power.pl", NULL };
721 \&
722 \& PERL_SYS_INIT3(&argc,&argv,&env);
723 \& my_perl = perl_alloc();
724 \& perl_construct( my_perl );
725 \&
726 \& perl_parse(my_perl, NULL, 2, my_argv, (char **)NULL);
727 \& PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
728 \& perl_run(my_perl);
729 \&
730 \& PerlPower(3, 4); /*** Compute 3 ** 4 ***/
731 \&
732 \& perl_destruct(my_perl);
733 \& perl_free(my_perl);
734 \& PERL_SYS_TERM();
735 \& exit(EXIT_SUCCESS);
736 \& }
737 .Ve
738 .PP
739 Compile and run:
740 .PP
741 .Vb 1
742 \& % cc \-o power power.c \`perl \-MExtUtils::Embed \-e ccopts \-e ldopts\`
743 \&
744 \& % power
745 \& 3 to the 4th power is 81.
746 .Ve
747 .SS "Maintaining a persistent interpreter"
748 .IX Subsection "Maintaining a persistent interpreter"
749 When developing interactive and/or potentially long-running
750 applications, it's a good idea to maintain a persistent interpreter
751 rather than allocating and constructing a new interpreter multiple
752 times. The major reason is speed: since Perl will only be loaded into
753 memory once.
754 .PP
755 However, you have to be more cautious with namespace and variable
756 scoping when using a persistent interpreter. In previous examples
757 we've been using global variables in the default package \f(CW\*(C`main\*(C'\fR. We
758 knew exactly what code would be run, and assumed we could avoid
759 variable collisions and outrageous symbol table growth.
760 .PP
761 Let's say your application is a server that will occasionally run Perl
762 code from some arbitrary file. Your server has no way of knowing what
763 code it's going to run. Very dangerous.
764 .PP
765 If the file is pulled in by \f(CWperl_parse()\fR, compiled into a newly
766 constructed interpreter, and subsequently cleaned out with
767 \&\f(CWperl_destruct()\fR afterwards, you're shielded from most namespace
768 troubles.
769 .PP
770 One way to avoid namespace collisions in this scenario is to translate
771 the filename into a guaranteed-unique package name, and then compile
772 the code into that package using "eval" in perlfunc. In the example
773 below, each file will only be compiled once. Or, the application
774 might choose to clean out the symbol table associated with the file
775 after it's no longer needed. Using "call_argv" in perlapi, We'll
776 call the subroutine \f(CW\*(C`Embed::Persistent::eval_file\*(C'\fR which lives in the
777 file \f(CW\*(C`persistent.pl\*(C'\fR and pass the filename and boolean cleanup/cache
778 flag as arguments.
779 .PP
780 Note that the process will continue to grow for each file that it
781 uses. In addition, there might be \f(CW\*(C`AUTOLOAD\*(C'\fRed subroutines and other
782 conditions that cause Perl's symbol table to grow. You might want to
783 add some logic that keeps track of the process size, or restarts
784 itself after a certain number of requests, to ensure that memory
785 consumption is minimized. You'll also want to scope your variables
786 with "my" in perlfunc whenever possible.
787 .PP
788 .Vb 2
789 \& package Embed::Persistent;
790 \& #persistent.pl
791 \&
792 \& use strict;
793 \& our %Cache;
794 \& use Symbol qw(delete_package);
795 \&
796 \& sub valid_package_name {
797 \& my($string) = @_;
798 \& $string =~ s/([^A\-Za\-z0\-9\e/])/sprintf("_%2x",unpack("C",$1))/eg;
799 \& # second pass only for words starting with a digit
800 \& $string =~ s|/(\ed)|sprintf("/_%2x",unpack("C",$1))|eg;
801 \&
802 \& # Dress it up as a real package name
803 \& $string =~ s|/|::|g;
804 \& return "Embed" . $string;
805 \& }
806 \&
807 \& sub eval_file {
808 \& my($filename, $delete) = @_;
809 \& my $package = valid_package_name($filename);
810 \& my $mtime = \-M $filename;
811 \& if(defined $Cache{$package}{mtime}
812 \& &&
813 \& $Cache{$package}{mtime} <= $mtime)
814 \& {
815 \& # we have compiled this subroutine already,
816 \& # it has not been updated on disk, nothing left to do
817 \& print STDERR "already compiled $package\->handler\en";
818 \& }
819 \& else {
820 \& local *FH;
821 \& open FH, $filename or die "open \*(Aq$filename\*(Aq $!";
822 \& local($/) = undef;
823 \& my $sub = <FH>;
824 \& close FH;
825 \&
826 \& #wrap the code into a subroutine inside our unique package
827 \& my $eval = qq{package $package; sub handler { $sub; }};
828 \& {
829 \& # hide our variables within this block
830 \& my($filename,$mtime,$package,$sub);
831 \& eval $eval;
832 \& }
833 \& die $@ if $@;
834 \&
835 \& #cache it unless we\*(Aqre cleaning out each time
836 \& $Cache{$package}{mtime} = $mtime unless $delete;
837 \& }
838 \&
839 \& eval {$package\->handler;};
840 \& die $@ if $@;
841 \&
842 \& delete_package($package) if $delete;
843 \&
844 \& #take a look if you want
845 \& #print Devel::Symdump\->rnew($package)\->as_string, $/;
846 \& }
847 \&
848 \& 1;
849 \&
850 \& _\|_END_\|_
851 \&
852 \& /* persistent.c */
853 \& #include <EXTERN.h>
854 \& #include <perl.h>
855 \&
856 \& /* 1 = clean out filename\*(Aqs symbol table after each request,
857 \& 0 = don\*(Aqt
858 \& */
859 \& #ifndef DO_CLEAN
860 \& #define DO_CLEAN 0
861 \& #endif
862 \&
863 \& #define BUFFER_SIZE 1024
864 \&
865 \& static PerlInterpreter *my_perl = NULL;
866 \&
867 \& int
868 \& main(int argc, char **argv, char **env)
869 \& {
870 \& char *embedding[] = { "", "persistent.pl", NULL };
871 \& char *args[] = { "", DO_CLEAN, NULL };
872 \& char filename[BUFFER_SIZE];
873 \& int failing, exitstatus;
874 \&
875 \& PERL_SYS_INIT3(&argc,&argv,&env);
876 \& if((my_perl = perl_alloc()) == NULL) {
877 \& fprintf(stderr, "no memory!");
878 \& exit(EXIT_FAILURE);
879 \& }
880 \& perl_construct(my_perl);
881 \&
882 \& PL_origalen = 1; /* don\*(Aqt let $0 assignment update the
883 \& proctitle or embedding[0] */
884 \& failing = perl_parse(my_perl, NULL, 2, embedding, NULL);
885 \& PL_exit_flags |= PERL_EXIT_DESTRUCT_END;
886 \& if(!failing)
887 \& failing = perl_run(my_perl);
888 \& if(!failing) {
889 \& while(printf("Enter file name: ") &&
890 \& fgets(filename, BUFFER_SIZE, stdin)) {
891 \&
892 \& filename[strlen(filename)\-1] = \*(Aq\e0\*(Aq; /* strip \en */
893 \& /* call the subroutine,
894 \& passing it the filename as an argument */
895 \& args[0] = filename;
896 \& call_argv("Embed::Persistent::eval_file",
897 \& G_DISCARD | G_EVAL, args);
898 \&
899 \& /* check $@ */
900 \& if(SvTRUE(ERRSV))
901 \& fprintf(stderr, "eval error: %s\en", SvPV_nolen(ERRSV));
902 \& }
903 \& }
904 \&
905 \& PL_perl_destruct_level = 0;
906 \& exitstatus = perl_destruct(my_perl);
907 \& perl_free(my_perl);
908 \& PERL_SYS_TERM();
909 \& exit(exitstatus);
910 \& }
911 .Ve
912 .PP
913 Now compile:
914 .PP
915 .Vb 2
916 \& % cc \-o persistent persistent.c \e
917 \& \`perl \-MExtUtils::Embed \-e ccopts \-e ldopts\`
918 .Ve
919 .PP
920 Here's an example script file:
921 .PP
922 .Vb 3
923 \& #test.pl
924 \& my $string = "hello";
925 \& foo($string);
926 \&
927 \& sub foo {
928 \& print "foo says: @_\en";
929 \& }
930 .Ve
931 .PP
932 Now run:
933 .PP
934 .Vb 7
935 \& % persistent
936 \& Enter file name: test.pl
937 \& foo says: hello
938 \& Enter file name: test.pl
939 \& already compiled Embed::test_2epl\->handler
940 \& foo says: hello
941 \& Enter file name: ^C
942 .Ve
943 .SS "Execution of END blocks"
944 .IX Subsection "Execution of END blocks"
945 Traditionally END blocks have been executed at the end of the perl_run.
946 This causes problems for applications that never call perl_run. Since
947 perl 5.7.2 you can specify \f(CW\*(C`PL_exit_flags |= PERL_EXIT_DESTRUCT_END\*(C'\fR
948 to get the new behaviour. This also enables the running of END blocks if
949 the perl_parse fails and \f(CW\*(C`perl_destruct\*(C'\fR will return the exit value.
950 .ie n .SS "$0 assignments"
951 .el .SS "\f(CW$0\fP assignments"
952 .IX Subsection "$0 assignments"
953 When a perl script assigns a value to \f(CW$0\fR then the perl runtime will
954 try to make this value show up as the program name reported by "ps" by
955 updating the memory pointed to by the argv passed to \fBperl_parse()\fR and
956 also calling API functions like \fBsetproctitle()\fR where available. This
957 behaviour might not be appropriate when embedding perl and can be
958 disabled by assigning the value \f(CW1\fR to the variable \f(CW\*(C`PL_origalen\*(C'\fR
959 before \fBperl_parse()\fR is called.
960 .PP
961 The \fIpersistent.c\fR example above is for instance likely to segfault
962 when \f(CW$0\fR is assigned to if the \f(CW\*(C`PL_origalen = 1;\*(C'\fR assignment is
963 removed. This because perl will try to write to the read only memory
964 of the \f(CW\*(C`embedding[]\*(C'\fR strings.
965 .SS "Maintaining multiple interpreter instances"
966 .IX Subsection "Maintaining multiple interpreter instances"
967 Some rare applications will need to create more than one interpreter
968 during a session. Such an application might sporadically decide to
969 release any resources associated with the interpreter.
970 .PP
971 The program must take care to ensure that this takes place \fIbefore\fR
972 the next interpreter is constructed. By default, when perl is not
973 built with any special options, the global variable
974 \&\f(CW\*(C`PL_perl_destruct_level\*(C'\fR is set to \f(CW0\fR, since extra cleaning isn't
975 usually needed when a program only ever creates a single interpreter
976 in its entire lifetime.
977 .PP
978 Setting \f(CW\*(C`PL_perl_destruct_level\*(C'\fR to \f(CW1\fR makes everything squeaky clean:
979 .PP
980 .Vb 10
981 \& while(1) {
982 \& ...
983 \& /* reset global variables here with PL_perl_destruct_level = 1 */
984 \& PL_perl_destruct_level = 1;
985 \& perl_construct(my_perl);
986 \& ...
987 \& /* clean and reset _everything_ during perl_destruct */
988 \& PL_perl_destruct_level = 1;
989 \& perl_destruct(my_perl);
990 \& perl_free(my_perl);
991 \& ...
992 \& /* let\*(Aqs go do it again! */
993 \& }
994 .Ve
995 .PP
996 When \fR\f(BIperl_destruct()\fR\fI\fR is called, the interpreter's syntax parse tree
997 and symbol tables are cleaned up, and global variables are reset. The
998 second assignment to \f(CW\*(C`PL_perl_destruct_level\*(C'\fR is needed because
999 perl_construct resets it to \f(CW0\fR.
1000 .PP
1001 Now suppose we have more than one interpreter instance running at the
1002 same time. This is feasible, but only if you used the Configure option
1003 \&\f(CW\*(C`\-Dusemultiplicity\*(C'\fR or the options \f(CW\*(C`\-Dusethreads \-Duseithreads\*(C'\fR when
1004 building perl. By default, enabling one of these Configure options
1005 sets the per-interpreter global variable \f(CW\*(C`PL_perl_destruct_level\*(C'\fR to
1006 \&\f(CW1\fR, so that thorough cleaning is automatic and interpreter variables
1007 are initialized correctly. Even if you don't intend to run two or
1008 more interpreters at the same time, but to run them sequentially, like
1009 in the above example, it is recommended to build perl with the
1010 \&\f(CW\*(C`\-Dusemultiplicity\*(C'\fR option otherwise some interpreter variables may
1011 not be initialized correctly between consecutive runs and your
1012 application may crash.
1013 .PP
1014 See also "Dealing with embedded perls and threads" in perlclib.
1015 .PP
1016 Using \f(CW\*(C`\-Dusethreads \-Duseithreads\*(C'\fR rather than \f(CW\*(C`\-Dusemultiplicity\*(C'\fR
1017 is more appropriate if you intend to run multiple interpreters
1018 concurrently in different threads, because it enables support for
1019 linking in the thread libraries of your system with the interpreter.
1020 .PP
1021 Let's give it a try:
1022 .PP
1023 .Vb 2
1024 \& #include <EXTERN.h>
1025 \& #include <perl.h>
1026 \&
1027 \& /* we\*(Aqre going to embed two interpreters */
1028 \&
1029 \& #define SAY_HELLO "\-e", "print qq(Hi, I\*(Aqm $^X\en)"
1030 \&
1031 \& int main(int argc, char **argv, char **env)
1032 \& {
1033 \& PerlInterpreter *one_perl, *two_perl;
1034 \& char *one_args[] = { "one_perl", SAY_HELLO, NULL };
1035 \& char *two_args[] = { "two_perl", SAY_HELLO, NULL };
1036 \&
1037 \& PERL_SYS_INIT3(&argc,&argv,&env);
1038 \& one_perl = perl_alloc();
1039 \& two_perl = perl_alloc();
1040 \&
1041 \& PERL_SET_CONTEXT(one_perl);
1042 \& perl_construct(one_perl);
1043 \& PERL_SET_CONTEXT(two_perl);
1044 \& perl_construct(two_perl);
1045 \&
1046 \& PERL_SET_CONTEXT(one_perl);
1047 \& perl_parse(one_perl, NULL, 3, one_args, (char **)NULL);
1048 \& PERL_SET_CONTEXT(two_perl);
1049 \& perl_parse(two_perl, NULL, 3, two_args, (char **)NULL);
1050 \&
1051 \& PERL_SET_CONTEXT(one_perl);
1052 \& perl_run(one_perl);
1053 \& PERL_SET_CONTEXT(two_perl);
1054 \& perl_run(two_perl);
1055 \&
1056 \& PERL_SET_CONTEXT(one_perl);
1057 \& perl_destruct(one_perl);
1058 \& PERL_SET_CONTEXT(two_perl);
1059 \& perl_destruct(two_perl);
1060 \&
1061 \& PERL_SET_CONTEXT(one_perl);
1062 \& perl_free(one_perl);
1063 \& PERL_SET_CONTEXT(two_perl);
1064 \& perl_free(two_perl);
1065 \& PERL_SYS_TERM();
1066 \& exit(EXIT_SUCCESS);
1067 \& }
1068 .Ve
1069 .PP
1070 Note the calls to \fBPERL_SET_CONTEXT()\fR. These are necessary to initialize
1071 the global state that tracks which interpreter is the "current" one on
1072 the particular process or thread that may be running it. It should
1073 always be used if you have more than one interpreter and are making
1074 perl API calls on both interpreters in an interleaved fashion.
1075 .PP
1076 PERL_SET_CONTEXT(interp) should also be called whenever \f(CW\*(C`interp\*(C'\fR is
1077 used by a thread that did not create it (using either \fBperl_alloc()\fR, or
1078 the more esoteric \fBperl_clone()\fR).
1079 .PP
1080 Compile as usual:
1081 .PP
1082 .Vb 2
1083 \& % cc \-o multiplicity multiplicity.c \e
1084 \& \`perl \-MExtUtils::Embed \-e ccopts \-e ldopts\`
1085 .Ve
1086 .PP
1087 Run it, Run it:
1088 .PP
1089 .Vb 3
1090 \& % multiplicity
1091 \& Hi, I\*(Aqm one_perl
1092 \& Hi, I\*(Aqm two_perl
1093 .Ve
1094 .SS "Using Perl modules, which themselves use C libraries, from your C program"
1095 .IX Subsection "Using Perl modules, which themselves use C libraries, from your C program"
1096 If you've played with the examples above and tried to embed a script
1097 that \fR\f(BIuse()\fR\fI\fRs a Perl module (such as \fISocket\fR) which itself uses a C or C++
1098 library, this probably happened:
1099 .PP
1100 .Vb 3
1101 \& Can\*(Aqt load module Socket, dynamic loading not available in this perl.
1102 \& (You may need to build a new perl executable which either supports
1103 \& dynamic loading or has the Socket module statically linked into it.)
1104 .Ve
1105 .PP
1106 What's wrong?
1107 .PP
1108 Your interpreter doesn't know how to communicate with these extensions
1109 on its own. A little glue will help. Up until now you've been
1110 calling \fR\f(BIperl_parse()\fR\fI\fR, handing it NULL for the second argument:
1111 .PP
1112 .Vb 1
1113 \& perl_parse(my_perl, NULL, argc, my_argv, NULL);
1114 .Ve
1115 .PP
1116 That's where the glue code can be inserted to create the initial contact
1117 between Perl and linked C/C++ routines. Let's take a look some pieces of
1118 \&\fIperlmain.c\fR to see how Perl does this:
1119 .PP
1120 .Vb 1
1121 \& static void xs_init (pTHX);
1122 \&
1123 \& EXTERN_C void boot_DynaLoader (pTHX_ CV* cv);
1124 \& EXTERN_C void boot_Socket (pTHX_ CV* cv);
1125 \&
1126 \&
1127 \& EXTERN_C void
1128 \& xs_init(pTHX)
1129 \& {
1130 \& char *file = _\|_FILE_\|_;
1131 \& /* DynaLoader is a special case */
1132 \& newXS("DynaLoader::boot_DynaLoader", boot_DynaLoader, file);
1133 \& newXS("Socket::bootstrap", boot_Socket, file);
1134 \& }
1135 .Ve
1136 .PP
1137 Simply put: for each extension linked with your Perl executable
1138 (determined during its initial configuration on your
1139 computer or when adding a new extension),
1140 a Perl subroutine is created to incorporate the extension's
1141 routines. Normally, that subroutine is named
1142 \&\fR\f(BIModule::bootstrap()\fR\fI\fR and is invoked when you say \fIuse Module\fR. In
1143 turn, this hooks into an XSUB, \fIboot_Module\fR, which creates a Perl
1144 counterpart for each of the extension's XSUBs. Don't worry about this
1145 part; leave that to the \fIxsubpp\fR and extension authors. If your
1146 extension is dynamically loaded, DynaLoader creates \fI\fR\f(BIModule::bootstrap()\fR\fI\fR
1147 for you on the fly. In fact, if you have a working DynaLoader then there
1148 is rarely any need to link in any other extensions statically.
1149 .PP
1150 Once you have this code, slap it into the second argument of \fR\f(BIperl_parse()\fR\fI\fR:
1151 .PP
1152 .Vb 1
1153 \& perl_parse(my_perl, xs_init, argc, my_argv, NULL);
1154 .Ve
1155 .PP
1156 Then compile:
1157 .PP
1158 .Vb 1
1159 \& % cc \-o interp interp.c \`perl \-MExtUtils::Embed \-e ccopts \-e ldopts\`
1160 \&
1161 \& % interp
1162 \& use Socket;
1163 \& use SomeDynamicallyLoadedModule;
1164 \&
1165 \& print "Now I can use extensions!\en"\*(Aq
1166 .Ve
1167 .PP
1168 \&\fBExtUtils::Embed\fR can also automate writing the \fIxs_init\fR glue code.
1169 .PP
1170 .Vb 4
1171 \& % perl \-MExtUtils::Embed \-e xsinit \-\- \-o perlxsi.c
1172 \& % cc \-c perlxsi.c \`perl \-MExtUtils::Embed \-e ccopts\`
1173 \& % cc \-c interp.c \`perl \-MExtUtils::Embed \-e ccopts\`
1174 \& % cc \-o interp perlxsi.o interp.o \`perl \-MExtUtils::Embed \-e ldopts\`
1175 .Ve
1176 .PP
1177 Consult perlxs, perlclib, perlguts, and perlapi for more details.
1178 .SS "Using embedded Perl with POSIX locales"
1179 .IX Subsection "Using embedded Perl with POSIX locales"
1180 (See perllocale for information about these.)
1181 When a Perl interpreter normally starts up, it tells the system it wants
1182 to use the system's default locale. This is often, but not necessarily,
1183 the "C" or "POSIX" locale. Absent a \f(CW"use\ locale"\fR within the perl
1184 code, this mostly has no effect (but see "Not within the
1185 scope of "use locale"" in perllocale). Also, there is not a problem if the
1186 locale you want to use in your embedded perl is the same as the system
1187 default. However, this doesn't work if you have set up and want to use
1188 a locale that isn't the system default one. Starting in Perl v5.20, you
1189 can tell the embedded Perl interpreter that the locale is already
1190 properly set up, and to skip doing its own normal initialization. It
1191 skips if the environment variable \f(CW\*(C`PERL_SKIP_LOCALE_INIT\*(C'\fR is set (even
1192 if set to 0 or \f(CW""\fR). A perl that has this capability will define the
1193 C pre-processor symbol \f(CW\*(C`HAS_SKIP_LOCALE_INIT\*(C'\fR. This allows code that
1194 has to work with multiple Perl versions to do some sort of work-around
1195 when confronted with an earlier Perl.
1196 .PP
1197 If your program is using the POSIX 2008 multi-thread locale
1198 functionality, you should switch into the global locale and set that up
1199 properly before starting the Perl interpreter. It will then properly
1200 switch back to using the thread-safe functions.
1201 .SH "Hiding Perl_"
1202 .IX Header "Hiding Perl_"
1203 If you completely hide the short forms of the Perl public API,
1204 add \-DPERL_NO_SHORT_NAMES to the compilation flags. This means that
1205 for example instead of writing
1206 .PP
1207 .Vb 1
1208 \& warn("%d bottles of beer on the wall", bottlecount);
1209 .Ve
1210 .PP
1211 you will have to write the explicit full form
1212 .PP
1213 .Vb 1
1214 \& Perl_warn(aTHX_ "%d bottles of beer on the wall", bottlecount);
1215 .Ve
1216 .PP
1217 (See "Background and MULTIPLICITY" in perlguts for the explanation
1218 of the \f(CW\*(C`aTHX_\*(C'\fR. ) Hiding the short forms is very useful for avoiding
1219 all sorts of nasty (C preprocessor or otherwise) conflicts with other
1220 software packages (Perl defines about 2400 APIs with these short names,
1221 take or leave few hundred, so there certainly is room for conflict.)
1222 .SH MORAL
1223 .IX Header "MORAL"
1224 You can sometimes \fIwrite faster code\fR in C, but
1225 you can always \fIwrite code faster\fR in Perl. Because you can use
1226 each from the other, combine them as you wish.
1227 .SH AUTHOR
1228 .IX Header "AUTHOR"
1229 Jon Orwant <\[email protected]\fR> and Doug MacEachern
1230 <\[email protected]\fR>, with small contributions from Tim Bunce, Tom
1231 Christiansen, Guy Decoux, Hallvard Furuseth, Dov Grobgeld, and Ilya
1232 Zakharevich.
1233 .PP
1234 Doug MacEachern has an article on embedding in Volume 1,
1235 Issue 4 <https://www.foo.be/docs/tpj/issues/vol1_4/ewtoc.html> of
1236 The Perl Journal ( <http://www.tpj.com/> ). Doug is also the developer of the
1237 most widely-used Perl embedding: the mod_perl system
1238 (perl.apache.org), which embeds Perl in the Apache web server.
1239 Oracle, Binary Evolution, ActiveState, and Ben Sugars's nsapi_perl
1240 have used this model for Oracle, Netscape and Internet Information
1241 Server Perl plugins.
1242 .SH COPYRIGHT
1243 .IX Header "COPYRIGHT"
1244 Copyright (C) 1995, 1996, 1997, 1998 Doug MacEachern and Jon Orwant. All
1245 Rights Reserved.
1246 .PP
1247 This document may be distributed under the same terms as Perl itself.