1 |
=head1 NAME |
2 |
|
3 |
staticperl - perl, libc, 100 modules, all in one 500kb file |
4 |
|
5 |
=head1 SYNOPSIS |
6 |
|
7 |
staticperl help # print the embedded documentation |
8 |
staticperl fetch # fetch and unpack perl sources |
9 |
staticperl configure # fetch and then configure perl |
10 |
staticperl build # configure and then build perl |
11 |
staticperl install # build and then install perl |
12 |
staticperl clean # clean most intermediate files (restart at configure) |
13 |
staticperl distclean # delete everything installed by this script |
14 |
staticperl perl ... # invoke the perlinterpreter |
15 |
staticperl cpan # invoke CPAN shell |
16 |
staticperl instmod path... # install unpacked modules |
17 |
staticperl instcpan modulename... # install modules from CPAN |
18 |
staticperl mkbundle <bundle-args...> # see documentation |
19 |
staticperl mkperl <bundle-args...> # see documentation |
20 |
staticperl mkapp appname <bundle-args...> # see documentation |
21 |
|
22 |
Typical Examples: |
23 |
|
24 |
staticperl install # fetch, configure, build and install perl |
25 |
staticperl cpan # run interactive cpan shell |
26 |
staticperl mkperl -MConfig_heavy.pl # build a perl that supports -V |
27 |
staticperl mkperl -MAnyEvent::Impl::Perl -MAnyEvent::HTTPD -MURI -MURI::http |
28 |
# build a perl with the above modules linked in |
29 |
staticperl mkapp myapp --boot mainprog mymodules |
30 |
# build a binary "myapp" from mainprog and mymodules |
31 |
|
32 |
=head1 DESCRIPTION |
33 |
|
34 |
This script helps you to create single-file perl interpreters |
35 |
or applications, or embedding a perl interpreter in your |
36 |
applications. Single-file means that it is fully self-contained - no |
37 |
separate shared objects, no autoload fragments, no .pm or .pl files are |
38 |
needed. And when linking statically, you can create (or embed) a single |
39 |
file that contains perl interpreter, libc, all the modules you need, all |
40 |
the libraries you need and of course your actual program. |
41 |
|
42 |
With F<uClibc> and F<upx> on x86, you can create a single 500kb binary |
43 |
that contains perl and 100 modules such as POSIX, AnyEvent, EV, IO::AIO, |
44 |
Coro and so on. Or any other choice of modules. |
45 |
|
46 |
To see how this turns out, you can try out smallperl and bigperl, two |
47 |
pre-built static and compressed perl binaries with many and even more |
48 |
modules: just follow the links at L<http://staticperl.schmorp.de/>. |
49 |
|
50 |
The created files do not need write access to the file system (like PAR |
51 |
does). In fact, since this script is in many ways similar to PAR::Packer, |
52 |
here are the differences: |
53 |
|
54 |
=over 4 |
55 |
|
56 |
=item * The generated executables are much smaller than PAR created ones. |
57 |
|
58 |
Shared objects and the perl binary contain a lot of extra info, while |
59 |
the static nature of F<staticperl> allows the linker to remove all |
60 |
functionality and meta-info not required by the final executable. Even |
61 |
extensions statically compiled into perl at build time will only be |
62 |
present in the final executable when needed. |
63 |
|
64 |
In addition, F<staticperl> can strip perl sources much more effectively |
65 |
than PAR. |
66 |
|
67 |
=item * The generated executables start much faster. |
68 |
|
69 |
There is no need to unpack files, or even to parse Zip archives (which is |
70 |
slow and memory-consuming business). |
71 |
|
72 |
=item * The generated executables don't need a writable filesystem. |
73 |
|
74 |
F<staticperl> loads all required files directly from memory. There is no |
75 |
need to unpack files into a temporary directory. |
76 |
|
77 |
=item * More control over included files, more burden. |
78 |
|
79 |
PAR tries to be maintenance and hassle-free - it tries to include more |
80 |
files than necessary to make sure everything works out of the box. It |
81 |
mostly succeeds at this, but he extra files (such as the unicode database) |
82 |
can take substantial amounts of memory and file size. |
83 |
|
84 |
With F<staticperl>, the burden is mostly with the developer - only direct |
85 |
compile-time dependencies and L<AutoLoader> are handled automatically. |
86 |
This means the modules to include often need to be tweaked manually. |
87 |
|
88 |
All this does not preclude more permissive modes to be implemented in |
89 |
the future, but right now, you have to resolve state hidden dependencies |
90 |
manually. |
91 |
|
92 |
=item * PAR works out of the box, F<staticperl> does not. |
93 |
|
94 |
Maintaining your own custom perl build can be a pain in the ass, and while |
95 |
F<staticperl> tries to make this easy, it still requires a custom perl |
96 |
build and possibly fiddling with some modules. PAR is likely to produce |
97 |
results faster. |
98 |
|
99 |
Ok, PAR never has worked for me out of the box, and for some people, |
100 |
F<staticperl> does work out of the box, as they don't count "fiddling with |
101 |
module use lists" against it, but nevertheless, F<staticperl> is certainly |
102 |
a bit more difficult to use. |
103 |
|
104 |
=back |
105 |
|
106 |
=head1 HOW DOES IT WORK? |
107 |
|
108 |
Simple: F<staticperl> downloads, compile and installs a perl version of |
109 |
your choice in F<~/.staticperl>. You can add extra modules either by |
110 |
letting F<staticperl> install them for you automatically, or by using CPAN |
111 |
and doing it interactively. This usually takes 5-10 minutes, depending on |
112 |
the speed of your computer and your internet connection. |
113 |
|
114 |
It is possible to do program development at this stage, too. |
115 |
|
116 |
Afterwards, you create a list of files and modules you want to include, |
117 |
and then either build a new perl binary (that acts just like a normal perl |
118 |
except everything is compiled in), or you create bundle files (basically C |
119 |
sources you can use to embed all files into your project). |
120 |
|
121 |
This step is very fast (a few seconds if PPI is not used for stripping, or |
122 |
the stripped files are in the cache), and can be tweaked and repeated as |
123 |
often as necessary. |
124 |
|
125 |
=head1 THE F<STATICPERL> SCRIPT |
126 |
|
127 |
This module installs a script called F<staticperl> into your perl |
128 |
binary directory. The script is fully self-contained, and can be |
129 |
used without perl (for example, in an uClibc chroot environment). In |
130 |
fact, it can be extracted from the C<App::Staticperl> distribution |
131 |
tarball as F<bin/staticperl>, without any installation. The |
132 |
newest (possibly alpha) version can also be downloaded from |
133 |
L<http://staticperl.schmorp.de/staticperl>. |
134 |
|
135 |
F<staticperl> interprets the first argument as a command to execute, |
136 |
optionally followed by any parameters. |
137 |
|
138 |
There are two command categories: the "phase 1" commands which deal with |
139 |
installing perl and perl modules, and the "phase 2" commands, which deal |
140 |
with creating binaries and bundle files. |
141 |
|
142 |
=head2 PHASE 1 COMMANDS: INSTALLING PERL |
143 |
|
144 |
The most important command is F<install>, which does basically |
145 |
everything. The default is to download and install perl 5.12.3 and a few |
146 |
modules required by F<staticperl> itself, but all this can (and should) be |
147 |
changed - see L<CONFIGURATION>, below. |
148 |
|
149 |
The command |
150 |
|
151 |
staticperl install |
152 |
|
153 |
is normally all you need: It installs the perl interpreter in |
154 |
F<~/.staticperl/perl>. It downloads, configures, builds and installs the |
155 |
perl interpreter if required. |
156 |
|
157 |
Most of the following F<staticperl> subcommands simply run one or more |
158 |
steps of this sequence. |
159 |
|
160 |
If it fails, then most commonly because the compiler options I selected |
161 |
are not supported by your compiler - either edit the F<staticperl> script |
162 |
yourself or create F<~/.staticperl> shell script where your set working |
163 |
C<PERL_CCFLAGS> etc. variables. |
164 |
|
165 |
To force recompilation or reinstallation, you need to run F<staticperl |
166 |
distclean> first. |
167 |
|
168 |
=over 4 |
169 |
|
170 |
=item F<staticperl version> |
171 |
|
172 |
Prints some info about the version of the F<staticperl> script you are using. |
173 |
|
174 |
=item F<staticperl fetch> |
175 |
|
176 |
Runs only the download and unpack phase, unless this has already happened. |
177 |
|
178 |
=item F<staticperl configure> |
179 |
|
180 |
Configures the unpacked perl sources, potentially after downloading them first. |
181 |
|
182 |
=item F<staticperl build> |
183 |
|
184 |
Builds the configured perl sources, potentially after automatically |
185 |
configuring them. |
186 |
|
187 |
=item F<staticperl install> |
188 |
|
189 |
Wipes the perl installation directory (usually F<~/.staticperl/perl>) and |
190 |
installs the perl distribution, potentially after building it first. |
191 |
|
192 |
=item F<staticperl perl> [args...] |
193 |
|
194 |
Invokes the compiled perl interpreter with the given args. Basically the |
195 |
same as starting perl directly (usually via F<~/.staticperl/bin/perl>), |
196 |
but beats typing the path sometimes. |
197 |
|
198 |
Example: check that the Gtk2 module is installed and loadable. |
199 |
|
200 |
staticperl perl -MGtk2 -e0 |
201 |
|
202 |
=item F<staticperl cpan> [args...] |
203 |
|
204 |
Starts an interactive CPAN shell that you can use to install further |
205 |
modules. Installs the perl first if necessary, but apart from that, |
206 |
no magic is involved: you could just as well run it manually via |
207 |
F<~/.staticperl/perl/bin/cpan>. |
208 |
|
209 |
Any additional arguments are simply passed to the F<cpan> command. |
210 |
|
211 |
=item F<staticperl instcpan> module... |
212 |
|
213 |
Tries to install all the modules given and their dependencies, using CPAN. |
214 |
|
215 |
Example: |
216 |
|
217 |
staticperl instcpan EV AnyEvent::HTTPD Coro |
218 |
|
219 |
=item F<staticperl instsrc> directory... |
220 |
|
221 |
In the unlikely case that you have unpacked perl modules around and want |
222 |
to install from these instead of from CPAN, you can do this using this |
223 |
command by specifying all the directories with modules in them that you |
224 |
want to have built. |
225 |
|
226 |
=item F<staticperl clean> |
227 |
|
228 |
Deletes the perl source directory (and potentially cleans up other |
229 |
intermediate files). This can be used to clean up files only needed for |
230 |
building perl, without removing the installed perl interpreter. |
231 |
|
232 |
At the moment, it doesn't delete downloaded tarballs. |
233 |
|
234 |
The exact semantics of this command will probably change. |
235 |
|
236 |
=item F<staticperl distclean> |
237 |
|
238 |
This wipes your complete F<~/.staticperl> directory. Be careful with this, |
239 |
it nukes your perl download, perl sources, perl distribution and any |
240 |
installed modules. It is useful if you wish to start over "from scratch" |
241 |
or when you want to uninstall F<staticperl>. |
242 |
|
243 |
=back |
244 |
|
245 |
=head2 PHASE 2 COMMANDS: BUILDING PERL BUNDLES |
246 |
|
247 |
Building (linking) a new F<perl> binary is handled by a separate |
248 |
script. To make it easy to use F<staticperl> from a F<chroot>, the script |
249 |
is embedded into F<staticperl>, which will write it out and call for you |
250 |
with any arguments you pass: |
251 |
|
252 |
staticperl mkbundle mkbundle-args... |
253 |
|
254 |
In the oh so unlikely case of something not working here, you |
255 |
can run the script manually as well (by default it is written to |
256 |
F<~/.staticperl/mkbundle>). |
257 |
|
258 |
F<mkbundle> is a more conventional command and expect the argument |
259 |
syntax commonly used on UNIX clones. For example, this command builds |
260 |
a new F<perl> binary and includes F<Config.pm> (for F<perl -V>), |
261 |
F<AnyEvent::HTTPD>, F<URI> and a custom F<httpd> script (from F<eg/httpd> |
262 |
in this distribution): |
263 |
|
264 |
# first make sure we have perl and the required modules |
265 |
staticperl instcpan AnyEvent::HTTPD |
266 |
|
267 |
# now build the perl |
268 |
staticperl mkperl -MConfig_heavy.pl -MAnyEvent::Impl::Perl \ |
269 |
-MAnyEvent::HTTPD -MURI::http \ |
270 |
--add 'eg/httpd httpd.pm' |
271 |
|
272 |
# finally, invoke it |
273 |
./perl -Mhttpd |
274 |
|
275 |
As you can see, things are not quite as trivial: the L<Config> module has |
276 |
a hidden dependency which is not even a perl module (F<Config_heavy.pl>), |
277 |
L<AnyEvent> needs at least one event loop backend that we have to |
278 |
specify manually (here L<AnyEvent::Impl::Perl>), and the F<URI> module |
279 |
(required by L<AnyEvent::HTTPD>) implements various URI schemes as extra |
280 |
modules - since L<AnyEvent::HTTPD> only needs C<http> URIs, we only need |
281 |
to include that module. I found out about these dependencies by carefully |
282 |
watching any error messages about missing modules... |
283 |
|
284 |
Instead of building a new perl binary, you can also build a standalone |
285 |
application: |
286 |
|
287 |
# build the app |
288 |
staticperl mkapp app --boot eg/httpd \ |
289 |
-MAnyEvent::Impl::Perl -MAnyEvent::HTTPD -MURI::http |
290 |
|
291 |
# run it |
292 |
./app |
293 |
|
294 |
Here are the three phase 2 commands: |
295 |
|
296 |
=over 4 |
297 |
|
298 |
=item F<staticperl mkbundle> args... |
299 |
|
300 |
The "default" bundle command - it interprets the given bundle options and |
301 |
writes out F<bundle.h>, F<bundle.c>, F<bundle.ccopts> and F<bundle.ldopts> |
302 |
files, useful for embedding. |
303 |
|
304 |
=item F<staticperl mkperl> args... |
305 |
|
306 |
Creates a bundle just like F<staticperl mkbundle> (in fact, it's the same |
307 |
as invoking F<staticperl mkbundle --perl> args...), but then compiles and |
308 |
links a new perl interpreter that embeds the created bundle, then deletes |
309 |
all intermediate files. |
310 |
|
311 |
=item F<staticperl mkapp> filename args... |
312 |
|
313 |
Does the same as F<staticperl mkbundle> (in fact, it's the same as |
314 |
invoking F<staticperl mkbundle --app> filename args...), but then compiles |
315 |
and links a new standalone application that simply initialises the perl |
316 |
interpreter. |
317 |
|
318 |
The difference to F<staticperl mkperl> is that the standalone application |
319 |
does not act like a perl interpreter would - in fact, by default it would |
320 |
just do nothing and exit immediately, so you should specify some code to |
321 |
be executed via the F<--boot> option. |
322 |
|
323 |
=back |
324 |
|
325 |
=head3 OPTION PROCESSING |
326 |
|
327 |
All options can be given as arguments on the command line (typically |
328 |
using long (e.g. C<--verbose>) or short option (e.g. C<-v>) style). Since |
329 |
specifying a lot of options can make the command line very long and |
330 |
unwieldy, you can put all long options into a "bundle specification file" |
331 |
(one option per line, with or without C<--> prefix) and specify this |
332 |
bundle file instead. |
333 |
|
334 |
For example, the command given earlier to link a new F<perl> could also |
335 |
look like this: |
336 |
|
337 |
staticperl mkperl httpd.bundle |
338 |
|
339 |
With all options stored in the F<httpd.bundle> file (one option per line, |
340 |
everything after the option is an argument): |
341 |
|
342 |
use "Config_heavy.pl" |
343 |
use AnyEvent::Impl::Perl |
344 |
use AnyEvent::HTTPD |
345 |
use URI::http |
346 |
add eg/httpd httpd.pm |
347 |
|
348 |
All options that specify modules or files to be added are processed in the |
349 |
order given on the command line. |
350 |
|
351 |
=head3 BUNDLE CREATION WORKFLOW / STATICPELR MKBUNDLE OPTIONS |
352 |
|
353 |
F<staticperl mkbundle> works by first assembling a list of candidate |
354 |
files and modules to include, then filtering them by include/exclude |
355 |
patterns. The remaining modules (together with their direct dependencies, |
356 |
such as link libraries and L<AutoLoader> files) are then converted into |
357 |
bundle files suitable for embedding. F<staticperl mkbundle> can then |
358 |
optionally build a new perl interpreter or a standalone application. |
359 |
|
360 |
=over 4 |
361 |
|
362 |
=item Step 0: Generic argument processing. |
363 |
|
364 |
The following options influence F<staticperl mkbundle> itself. |
365 |
|
366 |
=over 4 |
367 |
|
368 |
=item C<--verbose> | C<-v> |
369 |
|
370 |
Increases the verbosity level by one (the default is C<1>). |
371 |
|
372 |
=item C<--quiet> | C<-q> |
373 |
|
374 |
Decreases the verbosity level by one. |
375 |
|
376 |
=item any other argument |
377 |
|
378 |
Any other argument is interpreted as a bundle specification file, which |
379 |
supports all options (without extra quoting), one option per line, in the |
380 |
format C<option> or C<option argument>. They will effectively be expanded |
381 |
and processed as if they were directly written on the command line, in |
382 |
place of the file name. |
383 |
|
384 |
=back |
385 |
|
386 |
=item Step 1: gather candidate files and modules |
387 |
|
388 |
In this step, modules, perl libraries (F<.pl> files) and other files are |
389 |
selected for inclusion in the bundle. The relevant options are executed |
390 |
in order (this makes a difference mostly for C<--eval>, which can rely on |
391 |
earlier C<--use> options to have been executed). |
392 |
|
393 |
=over 4 |
394 |
|
395 |
=item C<--use> F<module> | C<-M>F<module> |
396 |
|
397 |
Include the named module or perl library and trace direct |
398 |
dependencies. This is done by loading the module in a subprocess and |
399 |
tracing which other modules and files it actually loads. |
400 |
|
401 |
Example: include AnyEvent and AnyEvent::Impl::Perl. |
402 |
|
403 |
staticperl mkbundle --use AnyEvent --use AnyEvent::Impl::Perl |
404 |
|
405 |
Sometimes you want to load old-style "perl libraries" (F<.pl> files), or |
406 |
maybe other weirdly named files. To support this, the C<--use> option |
407 |
actually tries to do what you mean, depending on the string you specify: |
408 |
|
409 |
=over 4 |
410 |
|
411 |
=item a possibly valid module name, e.g. F<common::sense>, F<Carp>, |
412 |
F<Coro::Mysql>. |
413 |
|
414 |
If the string contains no quotes, no F</> and no F<.>, then C<--use> |
415 |
assumes that it is a normal module name. It will create a new package and |
416 |
evaluate a C<use module> in it, i.e. it will load the package and do a |
417 |
default import. |
418 |
|
419 |
The import step is done because many modules trigger more dependencies |
420 |
when something is imported than without. |
421 |
|
422 |
=item anything that contains F</> or F<.> characters, |
423 |
e.g. F<utf8_heavy.pl>, F<Module/private/data.pl>. |
424 |
|
425 |
The string will be quoted and passed to require, as if you used C<require |
426 |
$module>. Nothing will be imported. |
427 |
|
428 |
=item "path" or 'path', e.g. C<"utf8_heavy.pl">. |
429 |
|
430 |
If you enclose the name into single or double quotes, then the quotes will |
431 |
be removed and the resulting string will be passed to require. This syntax |
432 |
is form compatibility with older versions of staticperl and should not be |
433 |
used anymore. |
434 |
|
435 |
=back |
436 |
|
437 |
Example: C<use> AnyEvent::Socket, once using C<use> (importing the |
438 |
symbols), and once via C<require>, not importing any symbols. The first |
439 |
form is preferred as many modules load some extra dependencies when asked |
440 |
to export symbols. |
441 |
|
442 |
staticperl mkbundle -MAnyEvent::Socket # use + import |
443 |
staticperl mkbundle -MAnyEvent/Socket.pm # require only |
444 |
|
445 |
Example: include the required files for F<perl -V> to work in all its |
446 |
glory (F<Config.pm> is included automatically by the dependency tracker). |
447 |
|
448 |
# shell command |
449 |
staticperl mkbundle -MConfig_heavy.pl |
450 |
|
451 |
# bundle specification file |
452 |
use Config_heavy.pl |
453 |
|
454 |
The C<-M>module syntax is included as a convenience that might be easier |
455 |
to remember than C<--use> - it's the same switch as perl itself uses |
456 |
to load modules. Or maybe it confuses people. Time will tell. Or maybe |
457 |
not. Sigh. |
458 |
|
459 |
=item C<--eval> "perl code" | C<-e> "perl code" |
460 |
|
461 |
Sometimes it is easier (or necessary) to specify dependencies using perl |
462 |
code, or maybe one of the modules you use need a special use statement. In |
463 |
that case, you can use C<--eval> to execute some perl snippet or set some |
464 |
variables or whatever you need. All files C<require>'d or C<use>'d while |
465 |
executing the snippet are included in the final bundle. |
466 |
|
467 |
Keep in mind that F<mkbundle> will not import any symbols from the modules |
468 |
named by the C<--use> option, so do not expect the symbols from modules |
469 |
you C<--use>'d earlier on the command line to be available. |
470 |
|
471 |
Example: force L<AnyEvent> to detect a backend and therefore include it |
472 |
in the final bundle. |
473 |
|
474 |
staticperl mkbundle --eval 'use AnyEvent; AnyEvent::detect' |
475 |
|
476 |
# or like this |
477 |
staticperl mkbundle -MAnyEvent --eval 'AnyEvent::detect' |
478 |
|
479 |
Example: use a separate "bootstrap" script that C<use>'s lots of modules |
480 |
and also include this in the final bundle, to be executed automatically |
481 |
when the interpreter is initialised. |
482 |
|
483 |
staticperl mkbundle --eval 'do "bootstrap"' --boot bootstrap |
484 |
|
485 |
=item C<--boot> F<filename> |
486 |
|
487 |
Include the given file in the bundle and arrange for it to be |
488 |
executed (using C<require>) before the main program when the new perl |
489 |
is initialised. This can be used to modify C<@INC> or do similar |
490 |
modifications before the perl interpreter executes scripts given on the |
491 |
command line (or via C<-e>). This works even in an embedded interpreter - |
492 |
the file will be executed during interpreter initialisation in that case. |
493 |
|
494 |
=item C<--incglob> pattern |
495 |
|
496 |
This goes through all standard library directories and tries to match any |
497 |
F<.pm> and F<.pl> files against the extended glob pattern (see below). If |
498 |
a file matches, it is added. The pattern is matched against the full path |
499 |
of the file (sans the library directory prefix), e.g. F<Sys/Syslog.pm>. |
500 |
|
501 |
This is very useful to include "everything": |
502 |
|
503 |
--incglob '*' |
504 |
|
505 |
It is also useful for including perl libraries, or trees of those, such as |
506 |
the unicode database files needed by some perl built-ins, the regex engine |
507 |
and other modules. |
508 |
|
509 |
--incglob '/unicore/**.pl' |
510 |
|
511 |
=item C<--add> F<file> | C<--add> "F<file> alias" |
512 |
|
513 |
Adds the given (perl) file into the bundle (and optionally call it |
514 |
"alias"). The F<file> is either an absolute path or a path relative to the |
515 |
current directory. If an alias is specified, then this is the name it will |
516 |
use for C<@INC> searches, otherwise the path F<file> will be used as the |
517 |
internal name. |
518 |
|
519 |
This switch is used to include extra files into the bundle. |
520 |
|
521 |
Example: embed the file F<httpd> in the current directory as F<httpd.pm> |
522 |
when creating the bundle. |
523 |
|
524 |
staticperl mkperl --add "httpd httpd.pm" |
525 |
|
526 |
# can be accessed via "use httpd" |
527 |
|
528 |
Example: add a file F<initcode> from the current directory. |
529 |
|
530 |
staticperl mkperl --add 'initcode &initcode' |
531 |
|
532 |
# can be accessed via "do '&initcode'" |
533 |
|
534 |
Example: add local files as extra modules in the bundle. |
535 |
|
536 |
# specification file |
537 |
add file1 myfiles/file1.pm |
538 |
add file2 myfiles/file2.pm |
539 |
add file3 myfiles/file3.pl |
540 |
|
541 |
# then later, in perl, use |
542 |
use myfiles::file1; |
543 |
require myfiles::file2; |
544 |
my $res = do "myfiles/file3.pl"; |
545 |
|
546 |
=item C<--binadd> F<file> | C<--add> "F<file> alias" |
547 |
|
548 |
Just like C<--add>, except that it treats the file as binary and adds it |
549 |
without any postprocessing (perl files might get stripped to reduce their |
550 |
size). |
551 |
|
552 |
If you specify an alias you should probably add a C<&> prefix to avoid |
553 |
clashing with embedded perl files (whose paths never start with C<&>), |
554 |
and/or use a special directory prefix, such as C<&res/name>. |
555 |
|
556 |
You can later get a copy of these files by calling C<staticperl::find |
557 |
"alias">. |
558 |
|
559 |
An alternative way to embed binary files is to convert them to perl and |
560 |
use C<do> to get the contents - this method is a bit cumbersome, but works |
561 |
both inside and outside of a staticperl bundle: |
562 |
|
563 |
# a "binary" file, call it "bindata.pl" |
564 |
<<'SOME_MARKER' |
565 |
binary data NOT containing SOME_MARKER |
566 |
SOME_MARKER |
567 |
|
568 |
# load the binary |
569 |
chomp (my $data = do "bindata.pl"); |
570 |
|
571 |
=back |
572 |
|
573 |
=item Step 2: filter all files using C<--include> and C<--exclude> options. |
574 |
|
575 |
After all candidate files and modules are added, they are I<filtered> |
576 |
by a combination of C<--include> and C<--exclude> patterns (there is an |
577 |
implicit C<--include *> at the end, so if no filters are specified, all |
578 |
files are included). |
579 |
|
580 |
All that this step does is potentially reduce the number of files that are |
581 |
to be included - no new files are added during this step. |
582 |
|
583 |
=over 4 |
584 |
|
585 |
=item C<--include> pattern | C<-i> pattern | C<--exclude> pattern | C<-x> pattern |
586 |
|
587 |
These specify an include or exclude pattern to be applied to the candidate |
588 |
file list. An include makes sure that the given files will be part of the |
589 |
resulting file set, an exclude will exclude remaining files. The patterns |
590 |
are "extended glob patterns" (see below). |
591 |
|
592 |
The patterns are applied "in order" - files included via earlier |
593 |
C<--include> specifications cannot be removed by any following |
594 |
C<--exclude>, and likewise, and file excluded by an earlier C<--exclude> |
595 |
cannot be added by any following C<--include>. |
596 |
|
597 |
For example, to include everything except C<Devel> modules, but still |
598 |
include F<Devel::PPPort>, you could use this: |
599 |
|
600 |
--incglob '*' -i '/Devel/PPPort.pm' -x '/Devel/**' |
601 |
|
602 |
=back |
603 |
|
604 |
=item Step 3: add any extra or "hidden" dependencies. |
605 |
|
606 |
F<staticperl> currently knows about three extra types of depdendencies |
607 |
that are added automatically. Only one (F<.packlist> files) is currently |
608 |
optional and can be influenced, the others are always included: |
609 |
|
610 |
=over 4 |
611 |
|
612 |
=item C<--usepacklists> |
613 |
|
614 |
Read F<.packlist> files for each distribution that happens to match a |
615 |
module name you specified. Sounds weird, and it is, so expect semantics to |
616 |
change somehow in the future. |
617 |
|
618 |
The idea is that most CPAN distributions have a F<.pm> file that matches |
619 |
the name of the distribution (which is rather reasonable after all). |
620 |
|
621 |
If this switch is enabled, then if any of the F<.pm> files that have been |
622 |
selected match an install distribution, then all F<.pm>, F<.pl>, F<.al> |
623 |
and F<.ix> files installed by this distribution are also included. |
624 |
|
625 |
For example, using this switch, when the L<URI> module is specified, then |
626 |
all L<URI> submodules that have been installed via the CPAN distribution |
627 |
are included as well, so you don't have to manually specify them. |
628 |
|
629 |
=item L<AutoLoader> splitfiles |
630 |
|
631 |
Some modules use L<AutoLoader> - less commonly (hopefully) used functions |
632 |
are split into separate F<.al> files, and an index (F<.ix>) file contains |
633 |
the prototypes. |
634 |
|
635 |
Both F<.ix> and F<.al> files will be detected automatically and added to |
636 |
the bundle. |
637 |
|
638 |
=item link libraries (F<.a> files) |
639 |
|
640 |
Modules using XS (or any other non-perl language extension compiled at |
641 |
installation time) will have a static archive (typically F<.a>). These |
642 |
will automatically be added to the linker options in F<bundle.ldopts>. |
643 |
|
644 |
Should F<staticperl> find a dynamic link library (typically F<.so>) it |
645 |
will warn about it - obviously this shouldn't happen unless you use |
646 |
F<staticperl> on the wrong perl, or one (probably wrongly) configured to |
647 |
use dynamic loading. |
648 |
|
649 |
=item extra libraries (F<extralibs.ld>) |
650 |
|
651 |
Some modules need linking against external libraries - these are found in |
652 |
F<extralibs.ld> and added to F<bundle.ldopts>. |
653 |
|
654 |
=back |
655 |
|
656 |
=item Step 4: write bundle files and optionally link a program |
657 |
|
658 |
At this point, the select files will be read, processed (stripped) and |
659 |
finally the bundle files get written to disk, and F<staticperl mkbundle> |
660 |
is normally finished. Optionally, it can go a step further and either link |
661 |
a new F<perl> binary with all selected modules and files inside, or build |
662 |
a standalone application. |
663 |
|
664 |
Both the contents of the bundle files and any extra linking is controlled |
665 |
by these options: |
666 |
|
667 |
=over 4 |
668 |
|
669 |
=item C<--strip> C<none>|C<pod>|C<ppi> |
670 |
|
671 |
Specify the stripping method applied to reduce the file of the perl |
672 |
sources included. |
673 |
|
674 |
The default is C<pod>, which uses the L<Pod::Strip> module to remove all |
675 |
pod documentation, which is very fast and reduces file size a lot. |
676 |
|
677 |
The C<ppi> method uses L<PPI> to parse and condense the perl sources. This |
678 |
saves a lot more than just L<Pod::Strip>, and is generally safer, |
679 |
but is also a lot slower (some files take almost a minute to strip - |
680 |
F<staticperl> maintains a cache of stripped files to speed up subsequent |
681 |
runs for this reason). Note that this method doesn't optimise for raw file |
682 |
size, but for best compression (that means that the uncompressed file size |
683 |
is a bit larger, but the files compress better, e.g. with F<upx>). |
684 |
|
685 |
Last not least, if you need accurate line numbers in error messages, |
686 |
or in the unlikely case where C<pod> is too slow, or some module gets |
687 |
mistreated, you can specify C<none> to not mangle included perl sources in |
688 |
any way. |
689 |
|
690 |
=item C<--perl> |
691 |
|
692 |
After writing out the bundle files, try to link a new perl interpreter. It |
693 |
will be called F<perl> and will be left in the current working |
694 |
directory. The bundle files will be removed. |
695 |
|
696 |
This switch is automatically used when F<staticperl> is invoked with the |
697 |
C<mkperl> command instead of C<mkbundle>. |
698 |
|
699 |
Example: build a new F<./perl> binary with only L<common::sense> inside - |
700 |
it will be even smaller than the standard perl interpreter as none of the |
701 |
modules of the base distribution (such as L<Fcntl>) will be included. |
702 |
|
703 |
staticperl mkperl -Mcommon::sense |
704 |
|
705 |
=item C<--app> F<name> |
706 |
|
707 |
After writing out the bundle files, try to link a new standalone |
708 |
program. It will be called C<name>, and the bundle files get removed after |
709 |
linking it. |
710 |
|
711 |
This switch is automatically used when F<staticperl> is invoked with the |
712 |
C<mkapp> command instead of C<mkbundle>. |
713 |
|
714 |
The difference to the (mutually exclusive) C<--perl> option is that the |
715 |
binary created by this option will not try to act as a perl interpreter - |
716 |
instead it will simply initialise the perl interpreter, clean it up and |
717 |
exit. |
718 |
|
719 |
This means that, by default, it will do nothing but burn a few CPU cycles |
720 |
- for it to do something useful you I<must> add some boot code, e.g. with |
721 |
the C<--boot> option. |
722 |
|
723 |
Example: create a standalone perl binary called F<./myexe> that will |
724 |
execute F<appfile> when it is started. |
725 |
|
726 |
staticperl mkbundle --app myexe --boot appfile |
727 |
|
728 |
=item C<--ignore-env> |
729 |
|
730 |
Generates extra code to unset some environment variables before |
731 |
initialising/running perl. Perl supports a lot of environment variables |
732 |
that might alter execution in ways that might be undesirablre for |
733 |
standalone applications, and this option removes those known to cause |
734 |
trouble. |
735 |
|
736 |
Specifically, these are removed: |
737 |
|
738 |
C<PERL_HASH_SEED_DEBUG> and C<PERL_DEBUG_MSTATS> can cause underaible |
739 |
output, C<PERL5OPT>, C<PERL_DESTRUCT_LEVEL>, C<PERL_HASH_SEED> and |
740 |
C<PERL_SIGNALS> can alter execution significantly, and C<PERL_UNICODE>, |
741 |
C<PERLIO_DEBUG> and C<PERLIO> can affect input and output. |
742 |
|
743 |
The variables C<PERL_LIB> and C<PERL5_LIB> are always ignored because the |
744 |
startup code used by F<staticperl> overrides C<@INC> in all cases. |
745 |
|
746 |
This option will not make your program more secure (unless you are |
747 |
running with elevated privileges), but it will reduce the surprise effect |
748 |
when a user has these environment variables set and doesn't expect your |
749 |
standalone program to act like a perl interpreter. |
750 |
|
751 |
=item C<--static> |
752 |
|
753 |
Add C<-static> to F<bundle.ldopts>, which means a fully static (if |
754 |
supported by the OS) executable will be created. This is not immensely |
755 |
useful when just creating the bundle files, but is most useful when |
756 |
linking a binary with the C<--perl> or C<--app> options. |
757 |
|
758 |
The default is to link the new binary dynamically (that means all perl |
759 |
modules are linked statically, but all external libraries are still |
760 |
referenced dynamically). |
761 |
|
762 |
Keep in mind that Solaris doesn't support static linking at all, and |
763 |
systems based on GNU libc don't really support it in a very usable |
764 |
fashion either. Try uClibc if you want to create fully statically linked |
765 |
executables, or try the C<--staticlib> option to link only some libraries |
766 |
statically. |
767 |
|
768 |
=item C<--staticlib> libname |
769 |
|
770 |
When not linking fully statically, this option allows you to link specific |
771 |
libraries statically. What it does is simply replace all occurrences of |
772 |
C<-llibname> with the GCC-specific C<-Wl,-Bstatic -llibname -Wl,-Bdynamic> |
773 |
option. |
774 |
|
775 |
This will have no effect unless the library is actually linked against, |
776 |
specifically, C<--staticlib> will not link against the named library |
777 |
unless it would be linked against anyway. |
778 |
|
779 |
Example: link libcrypt statically into the final binary. |
780 |
|
781 |
staticperl mkperl -MIO::AIO --staticlib crypt |
782 |
|
783 |
# ldopts might now contain: |
784 |
# -lm -Wl,-Bstatic -lcrypt -Wl,-Bdynamic -lpthread |
785 |
|
786 |
=back |
787 |
|
788 |
=back |
789 |
|
790 |
=head3 EXTENDED GLOB PATTERNS |
791 |
|
792 |
Some options of F<staticperl mkbundle> expect an I<extended glob |
793 |
pattern>. This is neither a normal shell glob nor a regex, but something |
794 |
in between. The idea has been copied from rsync, and there are the current |
795 |
matching rules: |
796 |
|
797 |
=over 4 |
798 |
|
799 |
=item Patterns starting with F</> will be a anchored at the root of the library tree. |
800 |
|
801 |
That is, F</unicore> will match the F<unicore> directory in C<@INC>, but |
802 |
nothing inside, and neither any other file or directory called F<unicore> |
803 |
anywhere else in the hierarchy. |
804 |
|
805 |
=item Patterns not starting with F</> will be anchored at the end of the path. |
806 |
|
807 |
That is, F<idna.pl> will match any file called F<idna.pl> anywhere in the |
808 |
hierarchy, but not any directories of the same name. |
809 |
|
810 |
=item A F<*> matches anything within a single path component. |
811 |
|
812 |
That is, F</unicore/*.pl> would match all F<.pl> files directly inside |
813 |
C</unicore>, not any deeper level F<.pl> files. Or in other words, F<*> |
814 |
will not match slashes. |
815 |
|
816 |
=item A F<**> matches anything. |
817 |
|
818 |
That is, F</unicore/**.pl> would match all F<.pl> files under F</unicore>, |
819 |
no matter how deeply nested they are inside subdirectories. |
820 |
|
821 |
=item A F<?> matches a single character within a component. |
822 |
|
823 |
That is, F</Encode/??.pm> matches F</Encode/JP.pm>, but not the |
824 |
hypothetical F</Encode/J/.pm>, as F<?> does not match F</>. |
825 |
|
826 |
=back |
827 |
|
828 |
=head2 F<STATICPERL> CONFIGURATION AND HOOKS |
829 |
|
830 |
During (each) startup, F<staticperl> tries to source some shell files to |
831 |
allow you to fine-tune/override configuration settings. |
832 |
|
833 |
In them you can override shell variables, or define shell functions |
834 |
("hooks") to be called at specific phases during installation. For |
835 |
example, you could define a C<postinstall> hook to install additional |
836 |
modules from CPAN each time you start from scratch. |
837 |
|
838 |
If the env variable C<$STATICPERLRC> is set, then F<staticperl> will try |
839 |
to source the file named with it only. Otherwise, it tries the following |
840 |
shell files in order: |
841 |
|
842 |
/etc/staticperlrc |
843 |
~/.staticperlrc |
844 |
$STATICPERL/rc |
845 |
|
846 |
Note that the last file is erased during F<staticperl distclean>, so |
847 |
generally should not be used. |
848 |
|
849 |
=head3 CONFIGURATION VARIABLES |
850 |
|
851 |
=head4 Variables you I<should> override |
852 |
|
853 |
=over 4 |
854 |
|
855 |
=item C<EMAIL> |
856 |
|
857 |
The e-mail address of the person who built this binary. Has no good |
858 |
default, so should be specified by you. |
859 |
|
860 |
=item C<CPAN> |
861 |
|
862 |
The URL of the CPAN mirror to use (e.g. L<http://mirror.netcologne.de/cpan/>). |
863 |
|
864 |
=item C<EXTRA_MODULES> |
865 |
|
866 |
Additional modules installed during F<staticperl install>. Here you can |
867 |
set which modules you want have to installed from CPAN. |
868 |
|
869 |
Example: I really really need EV, AnyEvent, Coro and AnyEvent::AIO. |
870 |
|
871 |
EXTRA_MODULES="EV AnyEvent Coro AnyEvent::AIO" |
872 |
|
873 |
Note that you can also use a C<postinstall> hook to achieve this, and |
874 |
more. |
875 |
|
876 |
=back |
877 |
|
878 |
=head4 Variables you might I<want> to override |
879 |
|
880 |
=over 4 |
881 |
|
882 |
=item C<STATICPERL> |
883 |
|
884 |
The directory where staticperl stores all its files |
885 |
(default: F<~/.staticperl>). |
886 |
|
887 |
=item C<PERL_MM_USE_DEFAULT>, C<EV_EXTRA_DEFS>, ... |
888 |
|
889 |
Usually set to C<1> to make modules "less inquisitive" during their |
890 |
installation, you can set any environment variable you want - some modules |
891 |
(such as L<Coro> or L<EV>) use environment variables for further tweaking. |
892 |
|
893 |
=item C<PERL_VERSION> |
894 |
|
895 |
The perl version to install - default is currently C<5.12.3>, but C<5.8.9> |
896 |
is also a good choice (5.8.9 is much smaller than 5.12.3, while 5.10.1 is |
897 |
about as big as 5.12.3). |
898 |
|
899 |
=item C<PERL_PREFIX> |
900 |
|
901 |
The prefix where perl gets installed (default: F<$STATICPERL/perl>), |
902 |
i.e. where the F<bin> and F<lib> subdirectories will end up. |
903 |
|
904 |
=item C<PERL_CONFIGURE> |
905 |
|
906 |
Additional Configure options - these are simply passed to the perl |
907 |
Configure script. For example, if you wanted to enable dynamic loading, |
908 |
you could pass C<-Dusedl>. To enable ithreads (Why would you want that |
909 |
insanity? Don't! Use L<forks> instead!) you would pass C<-Duseithreads> |
910 |
and so on. |
911 |
|
912 |
More commonly, you would either activate 64 bit integer support |
913 |
(C<-Duse64bitint>), or disable large files support (-Uuselargefiles), to |
914 |
reduce filesize further. |
915 |
|
916 |
=item C<PERL_CC>, C<PERL_CCFLAGS>, C<PERL_OPTIMIZE>, C<PERL_LDFLAGS>, C<PERL_LIBS> |
917 |
|
918 |
These flags are passed to perl's F<Configure> script, and are generally |
919 |
optimised for small size (at the cost of performance). Since they also |
920 |
contain subtle workarounds around various build issues, changing these |
921 |
usually requires understanding their default values - best look at |
922 |
the top of the F<staticperl> script for more info on these, and use a |
923 |
F<~/.staticperlrc> to override them. |
924 |
|
925 |
Most of the variables override (or modify) the corresponding F<Configure> |
926 |
variable, except C<PERL_CCFLAGS>, which gets appended. |
927 |
|
928 |
=back |
929 |
|
930 |
=head4 Variables you probably I<do not want> to override |
931 |
|
932 |
=over 4 |
933 |
|
934 |
=item C<MAKE> |
935 |
|
936 |
The make command to use - default is C<make>. |
937 |
|
938 |
=item C<MKBUNDLE> |
939 |
|
940 |
Where F<staticperl> writes the C<mkbundle> command to |
941 |
(default: F<$STATICPERL/mkbundle>). |
942 |
|
943 |
=item C<STATICPERL_MODULES> |
944 |
|
945 |
Additional modules needed by C<mkbundle> - should therefore not be changed |
946 |
unless you know what you are doing. |
947 |
|
948 |
=back |
949 |
|
950 |
=head3 OVERRIDABLE HOOKS |
951 |
|
952 |
In addition to environment variables, it is possible to provide some |
953 |
shell functions that are called at specific times. To provide your own |
954 |
commands, just define the corresponding function. |
955 |
|
956 |
The actual order in which hooks are invoked during a full install |
957 |
from scratch is C<preconfigure>, C<patchconfig>, C<postconfigure>, |
958 |
C<postbuild>, C<postinstall>. |
959 |
|
960 |
Example: install extra modules from CPAN and from some directories |
961 |
at F<staticperl install> time. |
962 |
|
963 |
postinstall() { |
964 |
rm -rf lib/threads* # weg mit Schaden |
965 |
instcpan IO::AIO EV |
966 |
instsrc ~/src/AnyEvent |
967 |
instsrc ~/src/XML-Sablotron-1.0100001 |
968 |
instcpan Anyevent::AIO AnyEvent::HTTPD |
969 |
} |
970 |
|
971 |
=over 4 |
972 |
|
973 |
=item preconfigure |
974 |
|
975 |
Called just before running F<./Configure> in the perl source |
976 |
directory. Current working directory is the perl source directory. |
977 |
|
978 |
This can be used to set any C<PERL_xxx> variables, which might be costly |
979 |
to compute. |
980 |
|
981 |
=item patchconfig |
982 |
|
983 |
Called after running F<./Configure> in the perl source directory to create |
984 |
F<./config.sh>, but before running F<./Configure -S> to actually apply the |
985 |
config. Current working directory is the perl source directory. |
986 |
|
987 |
Can be used to tailor/patch F<config.sh> or do any other modifications. |
988 |
|
989 |
=item postconfigure |
990 |
|
991 |
Called after configuring, but before building perl. Current working |
992 |
directory is the perl source directory. |
993 |
|
994 |
=item postbuild |
995 |
|
996 |
Called after building, but before installing perl. Current working |
997 |
directory is the perl source directory. |
998 |
|
999 |
I have no clue what this could be used for - tell me. |
1000 |
|
1001 |
=item postinstall |
1002 |
|
1003 |
Called after perl and any extra modules have been installed in C<$PREFIX>, |
1004 |
but before setting the "installation O.K." flag. |
1005 |
|
1006 |
The current working directory is C<$PREFIX>, but maybe you should not rely |
1007 |
on that. |
1008 |
|
1009 |
This hook is most useful to customise the installation, by deleting files, |
1010 |
or installing extra modules using the C<instcpan> or C<instsrc> functions. |
1011 |
|
1012 |
The script must return with a zero exit status, or the installation will |
1013 |
fail. |
1014 |
|
1015 |
=back |
1016 |
|
1017 |
=head1 ANATOMY OF A BUNDLE |
1018 |
|
1019 |
When not building a new perl binary, C<mkbundle> will leave a number of |
1020 |
files in the current working directory, which can be used to embed a perl |
1021 |
interpreter in your program. |
1022 |
|
1023 |
Intimate knowledge of L<perlembed> and preferably some experience with |
1024 |
embedding perl is highly recommended. |
1025 |
|
1026 |
C<mkperl> (or the C<--perl> option) basically does this to link the new |
1027 |
interpreter (it also adds a main program to F<bundle.>): |
1028 |
|
1029 |
$Config{cc} $(cat bundle.ccopts) -o perl bundle.c $(cat bundle.ldopts) |
1030 |
|
1031 |
=over 4 |
1032 |
|
1033 |
=item bundle.h |
1034 |
|
1035 |
A header file that contains the prototypes of the few symbols "exported" |
1036 |
by bundle.c, and also exposes the perl headers to the application. |
1037 |
|
1038 |
=over 4 |
1039 |
|
1040 |
=item staticperl_init (xs_init = 0) |
1041 |
|
1042 |
Initialises the perl interpreter. You can use the normal perl functions |
1043 |
after calling this function, for example, to define extra functions or |
1044 |
to load a .pm file that contains some initialisation code, or the main |
1045 |
program function: |
1046 |
|
1047 |
XS (xsfunction) |
1048 |
{ |
1049 |
dXSARGS; |
1050 |
|
1051 |
// now we have items, ST(i) etc. |
1052 |
} |
1053 |
|
1054 |
static void |
1055 |
run_myapp(void) |
1056 |
{ |
1057 |
staticperl_init (0); |
1058 |
newXSproto ("myapp::xsfunction", xsfunction, __FILE__, "$$;$"); |
1059 |
eval_pv ("require myapp::main", 1); // executes "myapp/main.pm" |
1060 |
} |
1061 |
|
1062 |
When your bootcode already wants to access some XS functions at |
1063 |
compiletime, then you need to supply an C<xs_init> function pointer that |
1064 |
is called as soon as perl is initialised enough to define XS functions, |
1065 |
but before the preamble code is executed: |
1066 |
|
1067 |
static void |
1068 |
xs_init (pTHX) |
1069 |
{ |
1070 |
newXSproto ("myapp::xsfunction", xsfunction, __FILE__, "$$;$"); |
1071 |
} |
1072 |
|
1073 |
static void |
1074 |
run_myapp(void) |
1075 |
{ |
1076 |
staticperl_init (xs_init); |
1077 |
} |
1078 |
|
1079 |
=item staticperl_cleanup () |
1080 |
|
1081 |
In the unlikely case that you want to destroy the perl interpreter, here |
1082 |
is the corresponding function. |
1083 |
|
1084 |
=item staticperl_xs_init (pTHX) |
1085 |
|
1086 |
Sometimes you need direct control over C<perl_parse> and C<perl_run>, in |
1087 |
which case you do not want to use C<staticperl_init> but call them on your |
1088 |
own. |
1089 |
|
1090 |
Then you need this function - either pass it directly as the C<xs_init> |
1091 |
function to C<perl_parse>, or call it as one of the first things from your |
1092 |
own C<xs_init> function. |
1093 |
|
1094 |
=item PerlInterpreter *staticperl |
1095 |
|
1096 |
The perl interpreter pointer used by staticperl. Not normally so useful, |
1097 |
but there it is. |
1098 |
|
1099 |
=back |
1100 |
|
1101 |
=item bundle.ccopts |
1102 |
|
1103 |
Contains the compiler options required to compile at least F<bundle.c> and |
1104 |
any file that includes F<bundle.h> - you should probably use it in your |
1105 |
C<CFLAGS>. |
1106 |
|
1107 |
=item bundle.ldopts |
1108 |
|
1109 |
The linker options needed to link the final program. |
1110 |
|
1111 |
=back |
1112 |
|
1113 |
=head1 RUNTIME FUNCTIONALITY |
1114 |
|
1115 |
Binaries created with C<mkbundle>/C<mkperl> contain extra functions, which |
1116 |
are required to access the bundled perl sources, but might be useful for |
1117 |
other purposes. |
1118 |
|
1119 |
In addition, for the embedded loading of perl files to work, F<staticperl> |
1120 |
overrides the C<@INC> array. |
1121 |
|
1122 |
=over 4 |
1123 |
|
1124 |
=item $file = staticperl::find $path |
1125 |
|
1126 |
Returns the data associated with the given C<$path> |
1127 |
(e.g. C<Digest/MD5.pm>, C<auto/POSIX/autosplit.ix>), which is basically |
1128 |
the UNIX path relative to the perl library directory. |
1129 |
|
1130 |
Returns C<undef> if the file isn't embedded. |
1131 |
|
1132 |
=item @paths = staticperl::list |
1133 |
|
1134 |
Returns the list of all paths embedded in this binary. |
1135 |
|
1136 |
=back |
1137 |
|
1138 |
=head1 FULLY STATIC BINARIES - UCLIBC AND BUILDROOT |
1139 |
|
1140 |
To make truly static (Linux-) libraries, you might want to have a look at |
1141 |
buildroot (L<http://buildroot.uclibc.org/>). |
1142 |
|
1143 |
Buildroot is primarily meant to set up a cross-compile environment (which |
1144 |
is not so useful as perl doesn't quite like cross compiles), but it can also compile |
1145 |
a chroot environment where you can use F<staticperl>. |
1146 |
|
1147 |
To do so, download buildroot, and enable "Build options => development |
1148 |
files in target filesystem" and optionally "Build options => gcc |
1149 |
optimization level (optimize for size)". At the time of writing, I had |
1150 |
good experiences with GCC 4.4.x but not GCC 4.5. |
1151 |
|
1152 |
To minimise code size, I used C<-pipe -ffunction-sections -fdata-sections |
1153 |
-finline-limit=8 -fno-builtin-strlen -mtune=i386>. The C<-mtune=i386> |
1154 |
doesn't decrease codesize much, but it makes the file much more |
1155 |
compressible. |
1156 |
|
1157 |
If you don't need Coro or threads, you can go with "linuxthreads.old" (or |
1158 |
no thread support). For Coro, it is highly recommended to switch to a |
1159 |
uClibc newer than 0.9.31 (at the time of this writing, I used the 20101201 |
1160 |
snapshot) and enable NPTL, otherwise Coro needs to be configured with the |
1161 |
ultra-slow pthreads backend to work around linuxthreads bugs (it also uses |
1162 |
twice the address space needed for stacks). |
1163 |
|
1164 |
If you use C<linuxthreads.old>, then you should also be aware that |
1165 |
uClibc shares C<errno> between all threads when statically linking. See |
1166 |
L<http://lists.uclibc.org/pipermail/uclibc/2010-June/044157.html> for a |
1167 |
workaround (And L<https://bugs.uclibc.org/2089> for discussion). |
1168 |
|
1169 |
C<ccache> support is also recommended, especially if you want |
1170 |
to play around with buildroot options. Enabling the C<miniperl> |
1171 |
package will probably enable all options required for a successful |
1172 |
perl build. F<staticperl> itself additionally needs either C<wget> |
1173 |
(recommended, for CPAN) or C<curl>. |
1174 |
|
1175 |
As for shells, busybox should provide all that is needed, but the default |
1176 |
busybox configuration doesn't include F<comm> which is needed by perl - |
1177 |
either make a custom busybox config, or compile coreutils. |
1178 |
|
1179 |
For the latter route, you might find that bash has some bugs that keep |
1180 |
it from working properly in a chroot - either use dash (and link it to |
1181 |
F</bin/sh> inside the chroot) or link busybox to F</bin/sh>, using it's |
1182 |
built-in ash shell. |
1183 |
|
1184 |
Finally, you need F</dev/null> inside the chroot for many scripts to work |
1185 |
- F<cp /dev/null output/target/dev> or bind-mounting your F</dev> will |
1186 |
both provide this. |
1187 |
|
1188 |
After you have compiled and set up your buildroot target, you can copy |
1189 |
F<staticperl> from the C<App::Staticperl> distribution or from your |
1190 |
perl f<bin> directory (if you installed it) into the F<output/target> |
1191 |
filesystem, chroot inside and run it. |
1192 |
|
1193 |
=head1 RECIPES / SPECIFIC MODULES |
1194 |
|
1195 |
This section contains some common(?) recipes and information about |
1196 |
problems with some common modules or perl constructs that require extra |
1197 |
files to be included. |
1198 |
|
1199 |
=head2 MODULES |
1200 |
|
1201 |
=over 4 |
1202 |
|
1203 |
=item utf8 |
1204 |
|
1205 |
Some functionality in the utf8 module, such as swash handling (used |
1206 |
for unicode character ranges in regexes) is implemented in the |
1207 |
C<"utf8_heavy.pl"> library: |
1208 |
|
1209 |
-Mutf8_heavy.pl |
1210 |
|
1211 |
Many Unicode properties in turn are defined in separate modules, |
1212 |
such as C<"unicore/Heavy.pl"> and more specific data tables such as |
1213 |
C<"unicore/To/Digit.pl"> or C<"unicore/lib/Perl/Word.pl">. These tables |
1214 |
are big (7MB uncompressed, although F<staticperl> contains special |
1215 |
handling for those files), so including them on demand by your application |
1216 |
only might pay off. |
1217 |
|
1218 |
To simply include the whole unicode database, use: |
1219 |
|
1220 |
--incglob '/unicore/**.pl' |
1221 |
|
1222 |
=item AnyEvent |
1223 |
|
1224 |
AnyEvent needs a backend implementation that it will load in a delayed |
1225 |
fashion. The L<AnyEvent::Impl::Perl> backend is the default choice |
1226 |
for AnyEvent if it can't find anything else, and is usually a safe |
1227 |
fallback. If you plan to use e.g. L<EV> (L<POE>...), then you need to |
1228 |
include the L<AnyEvent::Impl::EV> (L<AnyEvent::Impl::POE>...) backend as |
1229 |
well. |
1230 |
|
1231 |
If you want to handle IRIs or IDNs (L<AnyEvent::Util> punycode and idn |
1232 |
functions), you also need to include C<"AnyEvent/Util/idna.pl"> and |
1233 |
C<"AnyEvent/Util/uts46data.pl">. |
1234 |
|
1235 |
Or you can use C<--usepacklists> and specify C<-MAnyEvent> to include |
1236 |
everything. |
1237 |
|
1238 |
=item Cairo |
1239 |
|
1240 |
See Glib, same problem, same solution. |
1241 |
|
1242 |
=item Carp |
1243 |
|
1244 |
Carp had (in older versions of perl) a dependency on L<Carp::Heavy>. As of |
1245 |
perl 5.12.2 (maybe earlier), this dependency no longer exists. |
1246 |
|
1247 |
=item Config |
1248 |
|
1249 |
The F<perl -V> switch (as well as many modules) needs L<Config>, which in |
1250 |
turn might need L<"Config_heavy.pl">. Including the latter gives you |
1251 |
both. |
1252 |
|
1253 |
=item Glib |
1254 |
|
1255 |
Glib literally requires Glib to be installed already to build - it tries |
1256 |
to fake this by running Glib out of the build directory before being |
1257 |
built. F<staticperl> tries to work around this by forcing C<MAN1PODS> and |
1258 |
C<MAN3PODS> to be empty via the C<PERL_MM_OPT> environment variable. |
1259 |
|
1260 |
=item Gtk2 |
1261 |
|
1262 |
See Pango, same problems, same solution. |
1263 |
|
1264 |
=item Pango |
1265 |
|
1266 |
In addition to the C<MAN3PODS> problem in Glib, Pango also routes around |
1267 |
L<ExtUtils::MakeMaker> by compiling its files on its own. F<staticperl> |
1268 |
tries to patch L<ExtUtils::MM_Unix> to route around Pango. |
1269 |
|
1270 |
=item Term::ReadLine::Perl |
1271 |
|
1272 |
Also needs L<Term::ReadLine::readline>, or C<--usepacklists>. |
1273 |
|
1274 |
=item URI |
1275 |
|
1276 |
URI implements schemes as separate modules - the generic URL scheme is |
1277 |
implemented in L<URI::_generic>, HTTP is implemented in L<URI::http>. If |
1278 |
you need to use any of these schemes, you should include these manually, |
1279 |
or use C<--usepacklists>. |
1280 |
|
1281 |
=back |
1282 |
|
1283 |
=head2 RECIPES |
1284 |
|
1285 |
=over 4 |
1286 |
|
1287 |
=item Just link everything in |
1288 |
|
1289 |
To link just about everything installed in the perl library into a new |
1290 |
perl, try this (the first time this runs it will take a long time, as a |
1291 |
lot of files need to be parsed): |
1292 |
|
1293 |
staticperl mkperl -v --strip ppi --incglob '*' |
1294 |
|
1295 |
If you don't mind the extra megabytes, this can be a very effective way of |
1296 |
creating bundles without having to worry about forgetting any modules. |
1297 |
|
1298 |
You get even more useful variants of this method by first selecting |
1299 |
everything, and then excluding stuff you are reasonable sure not to need - |
1300 |
L<bigperl|http://staticperl.schmorp.de/bigperl.html> uses this approach. |
1301 |
|
1302 |
=item Getting rid of netdb functions |
1303 |
|
1304 |
The perl core has lots of netdb functions (C<getnetbyname>, C<getgrent> |
1305 |
and so on) that few applications use. You can avoid compiling them in by |
1306 |
putting the following fragment into a C<preconfigure> hook: |
1307 |
|
1308 |
preconfigure() { |
1309 |
for sym in \ |
1310 |
d_getgrnam_r d_endgrent d_endgrent_r d_endhent \ |
1311 |
d_endhostent_r d_endnent d_endnetent_r d_endpent \ |
1312 |
d_endprotoent_r d_endpwent d_endpwent_r d_endsent \ |
1313 |
d_endservent_r d_getgrent d_getgrent_r d_getgrgid_r \ |
1314 |
d_getgrnam_r d_gethbyaddr d_gethent d_getsbyport \ |
1315 |
d_gethostbyaddr_r d_gethostbyname_r d_gethostent_r \ |
1316 |
d_getlogin_r d_getnbyaddr d_getnbyname d_getnent \ |
1317 |
d_getnetbyaddr_r d_getnetbyname_r d_getnetent_r \ |
1318 |
d_getpent d_getpbyname d_getpbynumber d_getprotobyname_r \ |
1319 |
d_getprotobynumber_r d_getprotoent_r d_getpwent \ |
1320 |
d_getpwent_r d_getpwnam_r d_getpwuid_r d_getsent \ |
1321 |
d_getservbyname_r d_getservbyport_r d_getservent_r \ |
1322 |
d_getspnam_r d_getsbyname |
1323 |
# d_gethbyname |
1324 |
do |
1325 |
PERL_CONFIGURE="$PERL_CONFIGURE -U$sym" |
1326 |
done |
1327 |
} |
1328 |
|
1329 |
This mostly gains space when linking statically, as the functions will |
1330 |
likely not be linked in. The gain for dynamically-linked binaries is |
1331 |
smaller. |
1332 |
|
1333 |
Also, this leaves C<gethostbyname> in - not only is it actually used |
1334 |
often, the L<Socket> module also exposes it, so leaving it out usually |
1335 |
gains little. Why Socket exposes a C function that is in the core already |
1336 |
is anybody's guess. |
1337 |
|
1338 |
=back |
1339 |
|
1340 |
=head1 AUTHOR |
1341 |
|
1342 |
Marc Lehmann <schmorp@schmorp.de> |
1343 |
http://software.schmorp.de/pkg/staticperl.html |