… | |
… | |
28 | staticperl mkapp myapp --boot mainprog mymodules |
28 | staticperl mkapp myapp --boot mainprog mymodules |
29 | # build a binary "myapp" from mainprog and mymodules |
29 | # build a binary "myapp" from mainprog and mymodules |
30 | |
30 | |
31 | =head1 DESCRIPTION |
31 | =head1 DESCRIPTION |
32 | |
32 | |
33 | This script helps you creating single-file perl interpreters, or embedding |
33 | This script helps you to create single-file perl interpreters |
34 | a perl interpreter in your applications. Single-file means that it is |
34 | or applications, or embedding a perl interpreter in your |
35 | fully self-contained - no separate shared objects, no autoload fragments, |
35 | applications. Single-file means that it is fully self-contained - no |
36 | no .pm or .pl files are needed. And when linking statically, you can |
36 | separate shared objects, no autoload fragments, no .pm or .pl files are |
37 | create (or embed) a single file that contains perl interpreter, libc, all |
37 | needed. And when linking statically, you can create (or embed) a single |
38 | the modules you need and all the libraries you need. |
38 | file that contains perl interpreter, libc, all the modules you need, all |
|
|
39 | the libraries you need and of course your actual program. |
39 | |
40 | |
40 | With F<uClibc> and F<upx> on x86, you can create a single 500kb binary |
41 | With F<uClibc> and F<upx> on x86, you can create a single 500kb binary |
41 | that contains perl and 100 modules such as POSIX, AnyEvent, EV, IO::AIO, |
42 | that contains perl and 100 modules such as POSIX, AnyEvent, EV, IO::AIO, |
42 | Coro and so on. Or any other choice of modules. |
43 | Coro and so on. Or any other choice of modules. |
43 | |
44 | |
… | |
… | |
66 | =item * The generated executables don't need a writable filesystem. |
67 | =item * The generated executables don't need a writable filesystem. |
67 | |
68 | |
68 | F<staticperl> loads all required files directly from memory. There is no |
69 | F<staticperl> loads all required files directly from memory. There is no |
69 | need to unpack files into a temporary directory. |
70 | need to unpack files into a temporary directory. |
70 | |
71 | |
71 | =item * More control over included files. |
72 | =item * More control over included files, more burden. |
72 | |
73 | |
73 | PAR tries to be maintenance and hassle-free - it tries to include more |
74 | PAR tries to be maintenance and hassle-free - it tries to include more |
74 | files than necessary to make sure everything works out of the box. The |
75 | files than necessary to make sure everything works out of the box. It |
75 | extra files (such as the unicode database) can take substantial amounts of |
76 | mostly succeeds at this, but he extra files (such as the unicode database) |
76 | memory and file size. |
77 | can take substantial amounts of memory and file size. |
77 | |
78 | |
78 | With F<staticperl>, the burden is mostly with the developer - only direct |
79 | With F<staticperl>, the burden is mostly with the developer - only direct |
79 | compile-time dependencies and L<AutoLoader> are handled automatically. |
80 | compile-time dependencies and L<AutoLoader> are handled automatically. |
80 | This means the modules to include often need to be tweaked manually. |
81 | This means the modules to include often need to be tweaked manually. |
|
|
82 | |
|
|
83 | All this does not preclude more permissive modes to be implemented in |
|
|
84 | the future, but right now, you have to resolve state hidden dependencies |
|
|
85 | manually. |
81 | |
86 | |
82 | =item * PAR works out of the box, F<staticperl> does not. |
87 | =item * PAR works out of the box, F<staticperl> does not. |
83 | |
88 | |
84 | Maintaining your own custom perl build can be a pain in the ass, and while |
89 | Maintaining your own custom perl build can be a pain in the ass, and while |
85 | F<staticperl> tries to make this easy, it still requires a custom perl |
90 | F<staticperl> tries to make this easy, it still requires a custom perl |
… | |
… | |
106 | Afterwards, you create a list of files and modules you want to include, |
111 | Afterwards, you create a list of files and modules you want to include, |
107 | and then either build a new perl binary (that acts just like a normal perl |
112 | and then either build a new perl binary (that acts just like a normal perl |
108 | except everything is compiled in), or you create bundle files (basically C |
113 | except everything is compiled in), or you create bundle files (basically C |
109 | sources you can use to embed all files into your project). |
114 | sources you can use to embed all files into your project). |
110 | |
115 | |
111 | This step is very fast (a few seconds if PPI is not used for stripping, |
116 | This step is very fast (a few seconds if PPI is not used for stripping, or |
112 | more seconds otherwise, as PPI is very slow), and can be tweaked and |
117 | the stripped files are in the cache), and can be tweaked and repeated as |
113 | repeated as often as necessary. |
118 | often as necessary. |
114 | |
119 | |
115 | =head1 THE F<STATICPERL> SCRIPT |
120 | =head1 THE F<STATICPERL> SCRIPT |
116 | |
121 | |
117 | This module installs a script called F<staticperl> into your perl |
122 | This module installs a script called F<staticperl> into your perl |
118 | binary directory. The script is fully self-contained, and can be used |
123 | binary directory. The script is fully self-contained, and can be used |
… | |
… | |
281 | |
286 | |
282 | All options that specify modules or files to be added are processed in the |
287 | All options that specify modules or files to be added are processed in the |
283 | order given on the command line (that affects the C<--use> and C<--eval> |
288 | order given on the command line (that affects the C<--use> and C<--eval> |
284 | options at the moment). |
289 | options at the moment). |
285 | |
290 | |
|
|
291 | =head3 PACKAGE SELECTION WORKFLOW |
|
|
292 | |
|
|
293 | F<staticperl mkbundle> has a number of options to control package |
|
|
294 | selection. This section describes how they interact with each other. Also, |
|
|
295 | since I am still a newbie w.r.t. these issues, maybe future versions of |
|
|
296 | F<staticperl> will change this, so watch out :) |
|
|
297 | |
|
|
298 | The idiom "in order" means "in order that they are specified on the |
|
|
299 | commandline". If you use a bundle specification file, then the options |
|
|
300 | will be processed as if they were given in place of the bundle file name. |
|
|
301 | |
|
|
302 | =over 4 |
|
|
303 | |
|
|
304 | =item 1. apply all C<--use>, C<--eval>, C<--add>, C<--addbin> and |
|
|
305 | C<--incglob> options, in order. |
|
|
306 | |
|
|
307 | In addition, C<--use> and C<--eval> dependencies will be added when the |
|
|
308 | options are processed. |
|
|
309 | |
|
|
310 | =item 2. apply all C<--include> and C<--exclude> options, in order. |
|
|
311 | |
|
|
312 | All this step does is potentially reduce the number of files already |
|
|
313 | selected or found in phase 1. |
|
|
314 | |
|
|
315 | =item 3. find all modules (== F<.pm> files), gather their static archives |
|
|
316 | (F<.a>) and AutoLoader splitfiles (F<.ix> and F<.al> files) and find any |
|
|
317 | extra libraries they need for linking (F<extralibs.ld>). |
|
|
318 | |
|
|
319 | This step is required to link against XS extensions and also adds files |
|
|
320 | required for L<AutoLoader> to do it's job. |
|
|
321 | |
|
|
322 | =back |
|
|
323 | |
|
|
324 | After this, all the files selected for bundling will be read and processed |
|
|
325 | (stripped), the bundle files will be written, and optionally a new F<perl> |
|
|
326 | or application binary will be linked. |
|
|
327 | |
286 | =head3 MKBUNDLE OPTIONS |
328 | =head3 MKBUNDLE OPTIONS |
287 | |
329 | |
288 | =over 4 |
330 | =over 4 |
289 | |
331 | |
290 | =item --verbose | -v |
332 | =item --verbose | -v |
… | |
… | |
302 | |
344 | |
303 | The default is C<pod>, which uses the L<Pod::Strip> module to remove all |
345 | The default is C<pod>, which uses the L<Pod::Strip> module to remove all |
304 | pod documentation, which is very fast and reduces file size a lot. |
346 | pod documentation, which is very fast and reduces file size a lot. |
305 | |
347 | |
306 | The C<ppi> method uses L<PPI> to parse and condense the perl sources. This |
348 | The C<ppi> method uses L<PPI> to parse and condense the perl sources. This |
307 | saves a lot more than just L<Pod::Strip>, and is generally safer, but |
349 | saves a lot more than just L<Pod::Strip>, and is generally safer, |
308 | is also a lot slower, so is best used for production builds. Note that |
350 | but is also a lot slower (some files take almost a minute to strip - |
309 | this method doesn't optimise for raw file size, but for best compression |
351 | F<staticperl> maintains a cache of stripped files to speed up subsequent |
310 | (that means that the uncompressed file size is a bit larger, but the files |
352 | runs for this reason). Note that this method doesn't optimise for raw file |
311 | compress better, e.g. with F<upx>). |
353 | size, but for best compression (that means that the uncompressed file size |
|
|
354 | is a bit larger, but the files compress better, e.g. with F<upx>). |
312 | |
355 | |
313 | Last not least, if you need accurate line numbers in error messages, |
356 | Last not least, if you need accurate line numbers in error messages, |
314 | or in the unlikely case where C<pod> is too slow, or some module gets |
357 | or in the unlikely case where C<pod> is too slow, or some module gets |
315 | mistreated, you can specify C<none> to not mangle included perl sources in |
358 | mistreated, you can specify C<none> to not mangle included perl sources in |
316 | any way. |
359 | any way. |
… | |
… | |
410 | (using a C<require>) before anything else when the new perl is |
453 | (using a C<require>) before anything else when the new perl is |
411 | initialised. This can be used to modify C<@INC> or anything else before |
454 | initialised. This can be used to modify C<@INC> or anything else before |
412 | the perl interpreter executes scripts given on the command line (or via |
455 | the perl interpreter executes scripts given on the command line (or via |
413 | C<-e>). This works even in an embedded interpreter. |
456 | C<-e>). This works even in an embedded interpreter. |
414 | |
457 | |
415 | =item --add "file" | --add "file alias" |
458 | =item --incglob pattern |
|
|
459 | |
|
|
460 | This goes through all library directories and tries to match any F<.pm> |
|
|
461 | and F<.pl> files against the extended glob pattern (see below). If a file |
|
|
462 | matches, it is added. This switch will automatically detect L<AutoLoader> |
|
|
463 | files and the required link libraries for XS modules, but it will I<not> |
|
|
464 | scan the file for dependencies (at the moment). |
|
|
465 | |
|
|
466 | This is mainly useful to include "everything": |
|
|
467 | |
|
|
468 | --incglob '*' |
|
|
469 | |
|
|
470 | Or to include perl libraries, or trees of those, such as the unicode |
|
|
471 | database files needed by many other modules: |
|
|
472 | |
|
|
473 | --incglob '/unicore/**.pl' |
|
|
474 | |
|
|
475 | =item --add file | --add "file alias" |
416 | |
476 | |
417 | Adds the given (perl) file into the bundle (and optionally call it |
477 | Adds the given (perl) file into the bundle (and optionally call it |
418 | "alias"). This is useful to include any custom files into the bundle. |
478 | "alias"). This is useful to include any custom files into the bundle. |
419 | |
479 | |
420 | Example: embed the file F<httpd> as F<httpd.pm> when creating the bundle. |
480 | Example: embed the file F<httpd> as F<httpd.pm> when creating the bundle. |
… | |
… | |
426 | # specification file |
486 | # specification file |
427 | add file1 myfiles/file1 |
487 | add file1 myfiles/file1 |
428 | add file2 myfiles/file2 |
488 | add file2 myfiles/file2 |
429 | add file3 myfiles/file3 |
489 | add file3 myfiles/file3 |
430 | |
490 | |
431 | =item --binadd "file" | --add "file alias" |
491 | =item --binadd file | --add "file alias" |
432 | |
492 | |
433 | Just like C<--add>, except that it treats the file as binary and adds it |
493 | Just like C<--add>, except that it treats the file as binary and adds it |
434 | without any processing. |
494 | without any processing. |
435 | |
495 | |
436 | You should probably add a C</> prefix to avoid clashing with embedded |
496 | You should probably add a C</> prefix to avoid clashing with embedded |
437 | perl files (whose paths do not start with C</>), and/or use a special |
497 | perl files (whose paths do not start with C</>), and/or use a special |
438 | directory, such as C</res/name>. |
498 | directory, such as C</res/name>. |
439 | |
499 | |
440 | You can later get a copy of these files by calling C<staticperl::find |
500 | You can later get a copy of these files by calling C<staticperl::find |
441 | "alias">. |
501 | "alias">. |
|
|
502 | |
|
|
503 | =item --include pattern | -i pattern | --exclude pattern | -x pattern |
|
|
504 | |
|
|
505 | These two options define an include/exclude filter that is used after all |
|
|
506 | files selected by the other options have been found. Each include/exclude |
|
|
507 | is applied to all files found so far - an include makes sure that the |
|
|
508 | given files will be part of the resulting file set, an exclude will |
|
|
509 | exclude files. The patterns are "extended glob patterns" (see below). |
|
|
510 | |
|
|
511 | For example, to include everything, except C<Devel> modules, but still |
|
|
512 | include F<Devel::PPPort>, you could use this: |
|
|
513 | |
|
|
514 | --incglob '*' -i '/Devel/PPPort.pm' -x '/Devel/**' |
442 | |
515 | |
443 | =item --static |
516 | =item --static |
444 | |
517 | |
445 | When C<--perl> is also given, link statically instead of dynamically. The |
518 | When C<--perl> is also given, link statically instead of dynamically. The |
446 | default is to link the new perl interpreter fully dynamic (that means all |
519 | default is to link the new perl interpreter fully dynamic (that means all |
… | |
… | |
451 | systems based on GNU libc don't really support it in a usable fashion |
524 | systems based on GNU libc don't really support it in a usable fashion |
452 | either. Try uClibc if you want to create fully statically linked |
525 | either. Try uClibc if you want to create fully statically linked |
453 | executables, or try the C<--staticlibs> option to link only some libraries |
526 | executables, or try the C<--staticlibs> option to link only some libraries |
454 | statically. |
527 | statically. |
455 | |
528 | |
|
|
529 | =item --staticlib libname |
|
|
530 | |
|
|
531 | When not linking fully statically, this option allows you to link specific |
|
|
532 | libraries statically. What it does is simply replace all occurances of |
|
|
533 | C<-llibname> with the GCC-specific C<-Wl,-Bstatic -llibname -Wl,-Bdynamic> |
|
|
534 | option. |
|
|
535 | |
|
|
536 | This will have no effect unless the library is actually linked against, |
|
|
537 | specifically, C<--staticlib> will not link against the named library |
|
|
538 | unless it would be linked against anyway. |
|
|
539 | |
|
|
540 | Example: link libcrypt statically into the binary. |
|
|
541 | |
|
|
542 | staticperl mkperl -MIO::AIO --staticlib crypt |
|
|
543 | |
|
|
544 | # ldopts might nwo contain: |
|
|
545 | # -lm -Wl,-Bstatic -lcrypt -Wl,-Bdynamic -lpthread |
|
|
546 | |
456 | =item any other argument |
547 | =item any other argument |
457 | |
548 | |
458 | Any other argument is interpreted as a bundle specification file, which |
549 | Any other argument is interpreted as a bundle specification file, which |
459 | supports most long options (without extra quoting), one option per line. |
550 | supports most long options (without extra quoting), one option per line. |
|
|
551 | |
|
|
552 | =back |
|
|
553 | |
|
|
554 | =head3 EXTENDED GLOB PATTERNS |
|
|
555 | |
|
|
556 | Some options of F<staticperl mkbundle> expect an I<extended glob |
|
|
557 | pattern>. This is neither a normal shell glob nor a regex, but something |
|
|
558 | in between. The idea has been copied from rsync, and there are the current |
|
|
559 | matching rules: |
|
|
560 | |
|
|
561 | =over 4 |
|
|
562 | |
|
|
563 | =item Patterns starting with F</> will be a anchored at the root of the library tree. |
|
|
564 | |
|
|
565 | That is, F</unicore> will match the F<unicore> directory in C<@INC>, but |
|
|
566 | nothing inside, and neither any other file or directory called F<unicore> |
|
|
567 | anywhere else in the hierarchy. |
|
|
568 | |
|
|
569 | =item Patterns not starting with F</> will be anchored at the end of the path. |
|
|
570 | |
|
|
571 | That is, F<idna.pl> will match any file called F<idna.pl> anywhere in the |
|
|
572 | hierarchy, but not any directories of the same name. |
|
|
573 | |
|
|
574 | =item A F<*> matches any single component. |
|
|
575 | |
|
|
576 | That is, F</unicore/*.pl> would match all F<.pl> files directly inside |
|
|
577 | C</unicore>, not any deeper level F<.pl> files. Or in other words, F<*> |
|
|
578 | will not match slashes. |
|
|
579 | |
|
|
580 | =item A F<**> matches anything. |
|
|
581 | |
|
|
582 | That is, F</unicore/**.pl> would match all F<.pl> files under F</unicore>, |
|
|
583 | no matter how deeply nested they are inside subdirectories. |
|
|
584 | |
|
|
585 | =item A F<?> matches a single character within a component. |
|
|
586 | |
|
|
587 | That is, F</Encode/??.pm> matches F</Encode/JP.pm>, but not the |
|
|
588 | hypothetical F</Encode/J/.pm>, as F<?> does not match F</>. |
460 | |
589 | |
461 | =back |
590 | =back |
462 | |
591 | |
463 | =head2 F<STATICPERL> CONFIGURATION AND HOOKS |
592 | =head2 F<STATICPERL> CONFIGURATION AND HOOKS |
464 | |
593 | |
… | |
… | |
782 | After you have compiled and set up your buildroot target, you can copy |
911 | After you have compiled and set up your buildroot target, you can copy |
783 | F<staticperl> from the C<App::Staticperl> distribution or from your |
912 | F<staticperl> from the C<App::Staticperl> distribution or from your |
784 | perl f<bin> directory (if you installed it) into the F<output/target> |
913 | perl f<bin> directory (if you installed it) into the F<output/target> |
785 | filesystem, chroot inside and run it. |
914 | filesystem, chroot inside and run it. |
786 | |
915 | |
|
|
916 | =head1 RECIPES / SPECIFIC MODULES |
|
|
917 | |
|
|
918 | This section contains some common(?) recipes and information about |
|
|
919 | problems with some common modules or perl constructs that require extra |
|
|
920 | files to be included. |
|
|
921 | |
|
|
922 | =head2 MODULES |
|
|
923 | |
|
|
924 | =over 4 |
|
|
925 | |
|
|
926 | =item utf8 |
|
|
927 | |
|
|
928 | Some functionality in the utf8 module, such as swash handling (used |
|
|
929 | for unicode character ranges in regexes) is implemented in the |
|
|
930 | C<"utf8_heavy.pl"> library: |
|
|
931 | |
|
|
932 | -M'"utf8_heavy.pl"' |
|
|
933 | |
|
|
934 | Many Unicode properties in turn are defined in separate modules, |
|
|
935 | such as C<"unicore/Heavy.pl"> and more specific data tables such as |
|
|
936 | C<"unicore/To/Digit.pl"> or C<"unicore/lib/Perl/Word.pl">. These tables |
|
|
937 | are big (7MB uncompressed, although F<staticperl> contains special |
|
|
938 | handling for those files), so including them on demand by your application |
|
|
939 | only might pay off. |
|
|
940 | |
|
|
941 | To simply include the whole unicode database, use: |
|
|
942 | |
|
|
943 | --incglob '/unicore/*.pl' |
|
|
944 | |
|
|
945 | =item AnyEvent |
|
|
946 | |
|
|
947 | AnyEvent needs a backend implementation that it will load in a delayed |
|
|
948 | fashion. The L<AnyEvent::Impl::Perl> backend is the default choice |
|
|
949 | for AnyEvent if it can't find anything else, and is usually a safe |
|
|
950 | fallback. If you plan to use e.g. L<EV> (L<POE>...), then you need to |
|
|
951 | include the L<AnyEvent::Impl::EV> (L<AnyEvent::Impl::POE>...) backend as |
|
|
952 | well. |
|
|
953 | |
|
|
954 | If you want to handle IRIs or IDNs (L<AnyEvent::Util> punycode and idn |
|
|
955 | functions), you also need to include C<"AnyEvent/Util/idna.pl"> and |
|
|
956 | C<"AnyEvent/Util/uts46data.pl">. |
|
|
957 | |
|
|
958 | =item Carp |
|
|
959 | |
|
|
960 | Carp had (in older versions of perl) a dependency on L<Carp::Heavy>. As of |
|
|
961 | perl 5.12.2 (maybe earlier), this dependency no longer exists. |
|
|
962 | |
|
|
963 | =item Config |
|
|
964 | |
|
|
965 | The F<perl -V> switch (as well as many modules) needs L<Config>, which in |
|
|
966 | turn might need L<"Config_heavy.pl">. Including the latter gives you |
|
|
967 | both. |
|
|
968 | |
|
|
969 | =item Term::ReadLine::Perl |
|
|
970 | |
|
|
971 | Also needs L<Term::ReadLine::readline>. |
|
|
972 | |
|
|
973 | =item URI |
|
|
974 | |
|
|
975 | URI implements schemes as separate modules - the generic URL scheme is |
|
|
976 | implemented in L<URI::_generic>, HTTP is implemented in L<URI::http>. If |
|
|
977 | you need to use any of these schemes, you should include these manually. |
|
|
978 | |
|
|
979 | =back |
|
|
980 | |
|
|
981 | =head2 RECIPES |
|
|
982 | |
|
|
983 | =over 4 |
|
|
984 | |
|
|
985 | =item Linking everything in |
|
|
986 | |
|
|
987 | To link just about everything installed in the perl library into a new |
|
|
988 | perl, try this: |
|
|
989 | |
|
|
990 | staticperl mkperl --strip ppi --incglob '*' |
|
|
991 | |
|
|
992 | =item Getting rid of netdb function |
|
|
993 | |
|
|
994 | The perl core has lots of netdb functions (C<getnetbyname>, C<getgrent> |
|
|
995 | and so on) that few applications use. You can avoid compiling them in by |
|
|
996 | putting the following fragment into a C<preconfigure> hook: |
|
|
997 | |
|
|
998 | preconfigure() { |
|
|
999 | for sym in \ |
|
|
1000 | d_getgrnam_r d_endgrent d_endgrent_r d_endhent \ |
|
|
1001 | d_endhostent_r d_endnent d_endnetent_r d_endpent \ |
|
|
1002 | d_endprotoent_r d_endpwent d_endpwent_r d_endsent \ |
|
|
1003 | d_endservent_r d_getgrent d_getgrent_r d_getgrgid_r \ |
|
|
1004 | d_getgrnam_r d_gethbyaddr d_gethent d_getsbyport \ |
|
|
1005 | d_gethostbyaddr_r d_gethostbyname_r d_gethostent_r \ |
|
|
1006 | d_getlogin_r d_getnbyaddr d_getnbyname d_getnent \ |
|
|
1007 | d_getnetbyaddr_r d_getnetbyname_r d_getnetent_r \ |
|
|
1008 | d_getpent d_getpbyname d_getpbynumber d_getprotobyname_r \ |
|
|
1009 | d_getprotobynumber_r d_getprotoent_r d_getpwent \ |
|
|
1010 | d_getpwent_r d_getpwnam_r d_getpwuid_r d_getsent \ |
|
|
1011 | d_getservbyname_r d_getservbyport_r d_getservent_r \ |
|
|
1012 | d_getspnam_r d_getsbyname |
|
|
1013 | # d_gethbyname |
|
|
1014 | do |
|
|
1015 | PERL_CONFIGURE="$PERL_CONFIGURE -U$sym" |
|
|
1016 | done |
|
|
1017 | } |
|
|
1018 | |
|
|
1019 | This mostly gains space when linking staticaly, as the functions will |
|
|
1020 | liekly not be linked in. The gain for dynamically-linked binaries is |
|
|
1021 | smaller. |
|
|
1022 | |
|
|
1023 | Also, this leaves C<gethostbyname> in - not only is it actually used |
|
|
1024 | often, the L<Socket> module also exposes it, so leaving it out usually |
|
|
1025 | gains little. Why Socket exposes a C function that is in the core already |
|
|
1026 | is anybody's guess. |
|
|
1027 | |
|
|
1028 | =back |
|
|
1029 | |
787 | =head1 AUTHOR |
1030 | =head1 AUTHOR |
788 | |
1031 | |
789 | Marc Lehmann <schmorp@schmorp.de> |
1032 | Marc Lehmann <schmorp@schmorp.de> |
790 | http://software.schmorp.de/pkg/staticperl.html |
1033 | http://software.schmorp.de/pkg/staticperl.html |