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 cpan # invoke CPAN shell |
15 |
staticperl instmod path... # install unpacked modules |
16 |
staticperl instcpan modulename... # install modules from CPAN |
17 |
staticperl mkbundle <bundle-args...> # see documentation |
18 |
staticperl mkperl <bundle-args...> # see documentation |
19 |
staticperl mkapp appname <bundle-args...> # see documentation |
20 |
|
21 |
Typical Examples: |
22 |
|
23 |
staticperl install # fetch, configure, build and install perl |
24 |
staticperl cpan # run interactive cpan shell |
25 |
staticperl mkperl -M '"Config_heavy.pl"' # build a perl that supports -V |
26 |
staticperl mkperl -MAnyEvent::Impl::Perl -MAnyEvent::HTTPD -MURI -MURI::http |
27 |
# build a perl with the above modules linked in |
28 |
staticperl mkapp myapp --boot mainprog mymodules |
29 |
# build a binary "myapp" from mainprog and mymodules |
30 |
|
31 |
=head1 DESCRIPTION |
32 |
|
33 |
This script helps you to create single-file perl interpreters |
34 |
or applications, or embedding a perl interpreter in your |
35 |
applications. Single-file means that it is fully self-contained - no |
36 |
separate shared objects, no autoload fragments, no .pm or .pl files are |
37 |
needed. And when linking statically, you can create (or embed) a single |
38 |
file that contains perl interpreter, libc, all the modules you need, all |
39 |
the libraries you need and of course your actual program. |
40 |
|
41 |
With F<uClibc> and F<upx> on x86, you can create a single 500kb binary |
42 |
that contains perl and 100 modules such as POSIX, AnyEvent, EV, IO::AIO, |
43 |
Coro and so on. Or any other choice of modules. |
44 |
|
45 |
The created files do not need write access to the file system (like PAR |
46 |
does). In fact, since this script is in many ways similar to PAR::Packer, |
47 |
here are the differences: |
48 |
|
49 |
=over 4 |
50 |
|
51 |
=item * The generated executables are much smaller than PAR created ones. |
52 |
|
53 |
Shared objects and the perl binary contain a lot of extra info, while |
54 |
the static nature of F<staticperl> allows the linker to remove all |
55 |
functionality and meta-info not required by the final executable. Even |
56 |
extensions statically compiled into perl at build time will only be |
57 |
present in the final executable when needed. |
58 |
|
59 |
In addition, F<staticperl> can strip perl sources much more effectively |
60 |
than PAR. |
61 |
|
62 |
=item * The generated executables start much faster. |
63 |
|
64 |
There is no need to unpack files, or even to parse Zip archives (which is |
65 |
slow and memory-consuming business). |
66 |
|
67 |
=item * The generated executables don't need a writable filesystem. |
68 |
|
69 |
F<staticperl> loads all required files directly from memory. There is no |
70 |
need to unpack files into a temporary directory. |
71 |
|
72 |
=item * More control over included files, more burden. |
73 |
|
74 |
PAR tries to be maintenance and hassle-free - it tries to include more |
75 |
files than necessary to make sure everything works out of the box. It |
76 |
mostly succeeds at this, but he extra files (such as the unicode database) |
77 |
can take substantial amounts of memory and file size. |
78 |
|
79 |
With F<staticperl>, the burden is mostly with the developer - only direct |
80 |
compile-time dependencies and L<AutoLoader> are handled automatically. |
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. |
86 |
|
87 |
=item * PAR works out of the box, F<staticperl> does not. |
88 |
|
89 |
Maintaining your own custom perl build can be a pain in the ass, and while |
90 |
F<staticperl> tries to make this easy, it still requires a custom perl |
91 |
build and possibly fiddling with some modules. PAR is likely to produce |
92 |
results faster. |
93 |
|
94 |
Ok, PAR never has worked for me out of the box, and for some people, |
95 |
F<staticperl> does work out of the box, as they don't count "fiddling with |
96 |
module use lists" against it, but nevertheless, F<staticperl> is certainly |
97 |
a bit more difficult to use. |
98 |
|
99 |
=back |
100 |
|
101 |
=head1 HOW DOES IT WORK? |
102 |
|
103 |
Simple: F<staticperl> downloads, compile and installs a perl version of |
104 |
your choice in F<~/.staticperl>. You can add extra modules either by |
105 |
letting F<staticperl> install them for you automatically, or by using CPAN |
106 |
and doing it interactively. This usually takes 5-10 minutes, depending on |
107 |
the speed of your computer and your internet connection. |
108 |
|
109 |
It is possible to do program development at this stage, too. |
110 |
|
111 |
Afterwards, you create a list of files and modules you want to include, |
112 |
and then either build a new perl binary (that acts just like a normal perl |
113 |
except everything is compiled in), or you create bundle files (basically C |
114 |
sources you can use to embed all files into your project). |
115 |
|
116 |
This step is very fast (a few seconds if PPI is not used for stripping, or |
117 |
the stripped files are in the cache), and can be tweaked and repeated as |
118 |
often as necessary. |
119 |
|
120 |
=head1 THE F<STATICPERL> SCRIPT |
121 |
|
122 |
This module installs a script called F<staticperl> into your perl |
123 |
binary directory. The script is fully self-contained, and can be used |
124 |
without perl (for example, in an uClibc chroot environment). In fact, |
125 |
it can be extracted from the C<App::Staticperl> distribution tarball as |
126 |
F<bin/staticperl>, without any installation. |
127 |
|
128 |
F<staticperl> interprets the first argument as a command to execute, |
129 |
optionally followed by any parameters. |
130 |
|
131 |
There are two command categories: the "phase 1" commands which deal with |
132 |
installing perl and perl modules, and the "phase 2" commands, which deal |
133 |
with creating binaries and bundle files. |
134 |
|
135 |
=head2 PHASE 1 COMMANDS: INSTALLING PERL |
136 |
|
137 |
The most important command is F<install>, which does basically |
138 |
everything. The default is to download and install perl 5.12.2 and a few |
139 |
modules required by F<staticperl> itself, but all this can (and should) be |
140 |
changed - see L<CONFIGURATION>, below. |
141 |
|
142 |
The command |
143 |
|
144 |
staticperl install |
145 |
|
146 |
Is normally all you need: It installs the perl interpreter in |
147 |
F<~/.staticperl/perl>. It downloads, configures, builds and installs the |
148 |
perl interpreter if required. |
149 |
|
150 |
Most of the following commands simply run one or more steps of this |
151 |
sequence. |
152 |
|
153 |
To force recompilation or reinstallation, you need to run F<staticperl |
154 |
distclean> first. |
155 |
|
156 |
=over 4 |
157 |
|
158 |
=item F<staticperl fetch> |
159 |
|
160 |
Runs only the download and unpack phase, unless this has already happened. |
161 |
|
162 |
=item F<staticperl configure> |
163 |
|
164 |
Configures the unpacked perl sources, potentially after downloading them first. |
165 |
|
166 |
=item F<staticperl build> |
167 |
|
168 |
Builds the configured perl sources, potentially after automatically |
169 |
configuring them. |
170 |
|
171 |
=item F<staticperl install> |
172 |
|
173 |
Wipes the perl installation directory (usually F<~/.staticperl/perl>) and |
174 |
installs the perl distribution, potentially after building it first. |
175 |
|
176 |
=item F<staticperl cpan> [args...] |
177 |
|
178 |
Starts an interactive CPAN shell that you can use to install further |
179 |
modules. Installs the perl first if necessary, but apart from that, |
180 |
no magic is involved: you could just as well run it manually via |
181 |
F<~/.staticperl/perl/bin/cpan>. |
182 |
|
183 |
Any additional arguments are simply passed to the F<cpan> command. |
184 |
|
185 |
=item F<staticperl instcpan> module... |
186 |
|
187 |
Tries to install all the modules given and their dependencies, using CPAN. |
188 |
|
189 |
Example: |
190 |
|
191 |
staticperl instcpan EV AnyEvent::HTTPD Coro |
192 |
|
193 |
=item F<staticperl instsrc> directory... |
194 |
|
195 |
In the unlikely case that you have unpacked perl modules around and want |
196 |
to install from these instead of from CPAN, you can do this using this |
197 |
command by specifying all the directories with modules in them that you |
198 |
want to have built. |
199 |
|
200 |
=item F<staticperl clean> |
201 |
|
202 |
Deletes the perl source directory (and potentially cleans up other |
203 |
intermediate files). This can be used to clean up files only needed for |
204 |
building perl, without removing the installed perl interpreter, or to |
205 |
force a re-build from scratch. |
206 |
|
207 |
At the moment, it doesn't delete downloaded tarballs. |
208 |
|
209 |
=item F<staticperl distclean> |
210 |
|
211 |
This wipes your complete F<~/.staticperl> directory. Be careful with this, |
212 |
it nukes your perl download, perl sources, perl distribution and any |
213 |
installed modules. It is useful if you wish to start over "from scratch" |
214 |
or when you want to uninstall F<staticperl>. |
215 |
|
216 |
=back |
217 |
|
218 |
=head2 PHASE 2 COMMANDS: BUILDING PERL BUNDLES |
219 |
|
220 |
Building (linking) a new F<perl> binary is handled by a separate |
221 |
script. To make it easy to use F<staticperl> from a F<chroot>, the script |
222 |
is embedded into F<staticperl>, which will write it out and call for you |
223 |
with any arguments you pass: |
224 |
|
225 |
staticperl mkbundle mkbundle-args... |
226 |
|
227 |
In the oh so unlikely case of something not working here, you |
228 |
can run the script manually as well (by default it is written to |
229 |
F<~/.staticperl/mkbundle>). |
230 |
|
231 |
F<mkbundle> is a more conventional command and expect the argument |
232 |
syntax commonly used on UNIX clones. For example, this command builds |
233 |
a new F<perl> binary and includes F<Config.pm> (for F<perl -V>), |
234 |
F<AnyEvent::HTTPD>, F<URI> and a custom F<httpd> script (from F<eg/httpd> |
235 |
in this distribution): |
236 |
|
237 |
# first make sure we have perl and the required modules |
238 |
staticperl instcpan AnyEvent::HTTPD |
239 |
|
240 |
# now build the perl |
241 |
staticperl mkperl -M'"Config_heavy.pl"' -MAnyEvent::Impl::Perl \ |
242 |
-MAnyEvent::HTTPD -MURI::http \ |
243 |
--add 'eg/httpd httpd.pm' |
244 |
|
245 |
# finally, invoke it |
246 |
./perl -Mhttpd |
247 |
|
248 |
As you can see, things are not quite as trivial: the L<Config> module has |
249 |
a hidden dependency which is not even a perl module (F<Config_heavy.pl>), |
250 |
L<AnyEvent> needs at least one event loop backend that we have to |
251 |
specify manually (here L<AnyEvent::Impl::Perl>), and the F<URI> module |
252 |
(required by L<AnyEvent::HTTPD>) implements various URI schemes as extra |
253 |
modules - since L<AnyEvent::HTTPD> only needs C<http> URIs, we only need |
254 |
to include that module. I found out about these dependencies by carefully |
255 |
watching any error messages about missing modules... |
256 |
|
257 |
Instead of building a new perl binary, you can also build a standalone |
258 |
application: |
259 |
|
260 |
# build the app |
261 |
staticperl mkapp app --boot eg/httpd \ |
262 |
-MAnyEvent::Impl::Perl -MAnyEvent::HTTPD -MURI::http |
263 |
|
264 |
# run it |
265 |
./app |
266 |
|
267 |
=head3 OPTION PROCESSING |
268 |
|
269 |
All options can be given as arguments on the command line (typically |
270 |
using long (e.g. C<--verbose>) or short option (e.g. C<-v>) style). Since |
271 |
specifying a lot of modules can make the command line very cumbersome, |
272 |
you can put all long options into a "bundle specification file" (with or |
273 |
without C<--> prefix) and specify this bundle file instead. |
274 |
|
275 |
For example, the command given earlier could also look like this: |
276 |
|
277 |
staticperl mkperl httpd.bundle |
278 |
|
279 |
And all options could be in F<httpd.bundle>: |
280 |
|
281 |
use "Config_heavy.pl" |
282 |
use AnyEvent::Impl::Perl |
283 |
use AnyEvent::HTTPD |
284 |
use URI::http |
285 |
add eg/httpd httpd.pm |
286 |
|
287 |
All options that specify modules or files to be added are processed in the |
288 |
order given on the command line (that affects the C<--use> and C<--eval> |
289 |
options at the moment). |
290 |
|
291 |
=head3 MKBUNDLE OPTIONS |
292 |
|
293 |
=over 4 |
294 |
|
295 |
=item --verbose | -v |
296 |
|
297 |
Increases the verbosity level by one (the default is C<1>). |
298 |
|
299 |
=item --quiet | -q |
300 |
|
301 |
Decreases the verbosity level by one. |
302 |
|
303 |
=item --strip none|pod|ppi |
304 |
|
305 |
Specify the stripping method applied to reduce the file of the perl |
306 |
sources included. |
307 |
|
308 |
The default is C<pod>, which uses the L<Pod::Strip> module to remove all |
309 |
pod documentation, which is very fast and reduces file size a lot. |
310 |
|
311 |
The C<ppi> method uses L<PPI> to parse and condense the perl sources. This |
312 |
saves a lot more than just L<Pod::Strip>, and is generally safer, |
313 |
but is also a lot slower (some files take almost a minute to strip - |
314 |
F<staticperl> maintains a cache of stripped files to speed up subsequent |
315 |
runs for this reason). Note that this method doesn't optimise for raw file |
316 |
size, but for best compression (that means that the uncompressed file size |
317 |
is a bit larger, but the files compress better, e.g. with F<upx>). |
318 |
|
319 |
Last not least, if you need accurate line numbers in error messages, |
320 |
or in the unlikely case where C<pod> is too slow, or some module gets |
321 |
mistreated, you can specify C<none> to not mangle included perl sources in |
322 |
any way. |
323 |
|
324 |
=item --perl |
325 |
|
326 |
After writing out the bundle files, try to link a new perl interpreter. It |
327 |
will be called F<perl> and will be left in the current working |
328 |
directory. The bundle files will be removed. |
329 |
|
330 |
This switch is automatically used when F<staticperl> is invoked with the |
331 |
C<mkperl> command (instead of C<mkbundle>): |
332 |
|
333 |
# build a new ./perl with only common::sense in it - very small :) |
334 |
staticperl mkperl -Mcommon::sense |
335 |
|
336 |
=item --app name |
337 |
|
338 |
After writing out the bundle files, try to link a new standalone |
339 |
program. It will be called C<name>, and the bundle files get removed after |
340 |
linking it. |
341 |
|
342 |
The difference to the (mutually exclusive) C<--perl> option is that the |
343 |
binary created by this option will not try to act as a perl interpreter - |
344 |
instead it will simply initialise the perl interpreter, clean it up and |
345 |
exit. |
346 |
|
347 |
This switch is automatically used when F<staticperl> is invoked with the |
348 |
C<mkapp> command (instead of C<mkbundle>): |
349 |
|
350 |
To let it do something useful you I<must> add some boot code, e.g. with |
351 |
the C<--boot> option. |
352 |
|
353 |
Example: create a standalone perl binary that will execute F<appfile> when |
354 |
it is started. |
355 |
|
356 |
staticperl mkbundle --app myexe --boot appfile |
357 |
|
358 |
=item --use module | -Mmodule |
359 |
|
360 |
Include the named module and all direct dependencies. This is done by |
361 |
C<require>'ing the module in a subprocess and tracing which other modules |
362 |
and files it actually loads. If the module uses L<AutoLoader>, then all |
363 |
splitfiles will be included as well. |
364 |
|
365 |
Example: include AnyEvent and AnyEvent::Impl::Perl. |
366 |
|
367 |
staticperl mkbundle --use AnyEvent --use AnyEvent::Impl::Perl |
368 |
|
369 |
Sometimes you want to load old-style "perl libraries" (F<.pl> files), or |
370 |
maybe other weirdly named files. To do that, you need to quote the name in |
371 |
single or double quotes. When given on the command line, you probably need |
372 |
to quote once more to avoid your shell interpreting it. Common cases that |
373 |
need this are F<Config_heavy.pl> and F<utf8_heavy.pl>. |
374 |
|
375 |
Example: include the required files for F<perl -V> to work in all its |
376 |
glory (F<Config.pm> is included automatically by this). |
377 |
|
378 |
# bourne shell |
379 |
staticperl mkbundle --use '"Config_heavy.pl"' |
380 |
|
381 |
# bundle specification file |
382 |
use "Config_heavy.pl" |
383 |
|
384 |
The C<-Mmodule> syntax is included as an alias that might be easier to |
385 |
remember than C<use>. Or maybe it confuses people. Time will tell. Or |
386 |
maybe not. Argh. |
387 |
|
388 |
=item --eval "perl code" | -e "perl code" |
389 |
|
390 |
Sometimes it is easier (or necessary) to specify dependencies using perl |
391 |
code, or maybe one of the modules you use need a special use statement. In |
392 |
that case, you can use C<eval> to execute some perl snippet or set some |
393 |
variables or whatever you need. All files C<require>'d or C<use>'d in the |
394 |
script are included in the final bundle. |
395 |
|
396 |
Keep in mind that F<mkbundle> will only C<require> the modules named |
397 |
by the C<--use> option, so do not expect the symbols from modules you |
398 |
C<--use>'d earlier on the command line to be available. |
399 |
|
400 |
Example: force L<AnyEvent> to detect a backend and therefore include it |
401 |
in the final bundle. |
402 |
|
403 |
staticperl mkbundle --eval 'use AnyEvent; AnyEvent::detect' |
404 |
|
405 |
# or like this |
406 |
staticperl mkbundle -MAnyEvent --eval 'use AnyEvent; AnyEvent::detect' |
407 |
|
408 |
Example: use a separate "bootstrap" script that C<use>'s lots of modules |
409 |
and include this in the final bundle, to be executed automatically. |
410 |
|
411 |
staticperl mkbundle --eval 'do "bootstrap"' --boot bootstrap |
412 |
|
413 |
=item --boot filename |
414 |
|
415 |
Include the given file in the bundle and arrange for it to be executed |
416 |
(using a C<require>) before anything else when the new perl is |
417 |
initialised. This can be used to modify C<@INC> or anything else before |
418 |
the perl interpreter executes scripts given on the command line (or via |
419 |
C<-e>). This works even in an embedded interpreter. |
420 |
|
421 |
=item --incglob pattern |
422 |
|
423 |
This goes through all library directories and tries to match any F<.pm> |
424 |
and F<.pl> files against the extended glob pattern (see below). If a file |
425 |
matches, it is added. This switch will automatically detect L<AutoLoader> |
426 |
files and the required link libraries for XS modules, but it will I<not> |
427 |
scan the file for dependencies (at the moment). |
428 |
|
429 |
This is mainly useful to include "everything": |
430 |
|
431 |
--incglob '*' |
432 |
|
433 |
Or to include perl libraries, or trees of those, such as the unicode |
434 |
database files needed by many other modules: |
435 |
|
436 |
--incglob '/unicore/**.pl' |
437 |
|
438 |
=item --add file | --add "file alias" |
439 |
|
440 |
Adds the given (perl) file into the bundle (and optionally call it |
441 |
"alias"). This is useful to include any custom files into the bundle. |
442 |
|
443 |
Example: embed the file F<httpd> as F<httpd.pm> when creating the bundle. |
444 |
|
445 |
staticperl mkperl --add "httpd httpd.pm" |
446 |
|
447 |
It is also a great way to add any custom modules: |
448 |
|
449 |
# specification file |
450 |
add file1 myfiles/file1 |
451 |
add file2 myfiles/file2 |
452 |
add file3 myfiles/file3 |
453 |
|
454 |
=item --binadd file | --add "file alias" |
455 |
|
456 |
Just like C<--add>, except that it treats the file as binary and adds it |
457 |
without any processing. |
458 |
|
459 |
You should probably add a C</> prefix to avoid clashing with embedded |
460 |
perl files (whose paths do not start with C</>), and/or use a special |
461 |
directory, such as C</res/name>. |
462 |
|
463 |
You can later get a copy of these files by calling C<staticperl::find |
464 |
"alias">. |
465 |
|
466 |
=item --include pattern | -i pattern | --exclude pattern | -x pattern |
467 |
|
468 |
These two options define an include/exclude filter that is used after all |
469 |
files selected by the other options have been found. Each include/exclude |
470 |
is applied to all files found so far - an include makes sure that the |
471 |
given files will be part of the resulting file set, an exclude will |
472 |
exclude files. The patterns are "extended glob patterns" (see below). |
473 |
|
474 |
For example, to include everything, except C<Devel> modules, but still |
475 |
include F<Devel::PPPort>, you could use this: |
476 |
|
477 |
--incglob '*' -i '/Devel/PPPort.pm' -x '/Devel/**' |
478 |
|
479 |
=item --static |
480 |
|
481 |
When C<--perl> is also given, link statically instead of dynamically. The |
482 |
default is to link the new perl interpreter fully dynamic (that means all |
483 |
perl modules are linked statically, but all external libraries are still |
484 |
referenced dynamically). |
485 |
|
486 |
Keep in mind that Solaris doesn't support static linking at all, and |
487 |
systems based on GNU libc don't really support it in a usable fashion |
488 |
either. Try uClibc if you want to create fully statically linked |
489 |
executables, or try the C<--staticlibs> option to link only some libraries |
490 |
statically. |
491 |
|
492 |
=item --staticlib libname |
493 |
|
494 |
When not linking fully statically, this option allows you to link specific |
495 |
libraries statically. What it does is simply replace all occurances of |
496 |
C<-llibname> with the GCC-specific C<-Wl,-Bstatic -llibname -Wl,-Bdynamic> |
497 |
option. |
498 |
|
499 |
This will have no effect unless the library is actually linked against, |
500 |
specifically, C<--staticlib> will not link against the named library |
501 |
unless it would be linked against anyway. |
502 |
|
503 |
Example: link libcrypt statically into the binary. |
504 |
|
505 |
staticperl mkperl -MIO::AIO --staticlib crypt |
506 |
|
507 |
# ldopts might nwo contain: |
508 |
# -lm -Wl,-Bstatic -lcrypt -Wl,-Bdynamic -lpthread |
509 |
|
510 |
=item any other argument |
511 |
|
512 |
Any other argument is interpreted as a bundle specification file, which |
513 |
supports most long options (without extra quoting), one option per line. |
514 |
|
515 |
=back |
516 |
|
517 |
=head3 EXTENDED GLOB PATTERNS |
518 |
|
519 |
Some options of F<staticperl mkbundle> expect an I<extended glob |
520 |
pattern>. This is neither a normal shell glob nor a regex, but something |
521 |
in between. The idea has been copied from rsync, and there are the current |
522 |
matching rules: |
523 |
|
524 |
=over 4 |
525 |
|
526 |
=item Patterns starting with F</> will be a anchored at the root of the library tree. |
527 |
|
528 |
That is, F</unicore> will match the F<unicore> directory in C<@INC>, but |
529 |
nothing inside, and neither any other file or directory called F<unicore> |
530 |
anywhere else in the hierarchy. |
531 |
|
532 |
=item Patterns not starting with F</> will be anchored at the end of the path. |
533 |
|
534 |
That is, F<idna.pl> will match any file called F<idna.pl> anywhere in the |
535 |
hierarchy, but not any directories of the same name. |
536 |
|
537 |
=item A F<*> matches any single component. |
538 |
|
539 |
That is, F</unicore/*.pl> would match all F<.pl> files directly inside |
540 |
C</unicore>, not any deeper level F<.pl> files. Or in other words, F<*> |
541 |
will not match slashes. |
542 |
|
543 |
=item A F<**> matches anything. |
544 |
|
545 |
That is, F</unicore/**.pl> would match all F<.pl> files under F</unicore>, |
546 |
no matter how deeply nested they are inside subdirectories. |
547 |
|
548 |
=item A F<?> matches a single character within a component. |
549 |
|
550 |
That is, F</Encode/??.pm> matches F</Encode/JP.pm>, but not the |
551 |
hypothetical F</Encode/J/.pm>, as F<?> does not match F</>. |
552 |
|
553 |
=back |
554 |
|
555 |
=head2 F<STATICPERL> CONFIGURATION AND HOOKS |
556 |
|
557 |
During (each) startup, F<staticperl> tries to source the following shell |
558 |
files in order: |
559 |
|
560 |
/etc/staticperlrc |
561 |
~/.staticperlrc |
562 |
$STATICPERL/rc |
563 |
|
564 |
They can be used to override shell variables, or define functions to be |
565 |
called at specific phases. |
566 |
|
567 |
Note that the last file is erased during F<staticperl distclean>, so |
568 |
generally should not be used. |
569 |
|
570 |
=head3 CONFIGURATION VARIABLES |
571 |
|
572 |
=head4 Variables you I<should> override |
573 |
|
574 |
=over 4 |
575 |
|
576 |
=item C<EMAIL> |
577 |
|
578 |
The e-mail address of the person who built this binary. Has no good |
579 |
default, so should be specified by you. |
580 |
|
581 |
=item C<CPAN> |
582 |
|
583 |
The URL of the CPAN mirror to use (e.g. L<http://mirror.netcologne.de/cpan/>). |
584 |
|
585 |
=item C<EXTRA_MODULES> |
586 |
|
587 |
Additional modules installed during F<staticperl install>. Here you can |
588 |
set which modules you want have to installed from CPAN. |
589 |
|
590 |
Example: I really really need EV, AnyEvent, Coro and AnyEvent::AIO. |
591 |
|
592 |
EXTRA_MODULES="EV AnyEvent Coro AnyEvent::AIO" |
593 |
|
594 |
Note that you can also use a C<postinstall> hook to achieve this, and |
595 |
more. |
596 |
|
597 |
=back |
598 |
|
599 |
=head4 Variables you might I<want> to override |
600 |
|
601 |
=over 4 |
602 |
|
603 |
=item C<STATICPERL> |
604 |
|
605 |
The directory where staticperl stores all its files |
606 |
(default: F<~/.staticperl>). |
607 |
|
608 |
=item C<PERL_MM_USE_DEFAULT>, C<EV_EXTRA_DEFS>, ... |
609 |
|
610 |
Usually set to C<1> to make modules "less inquisitive" during their |
611 |
installation, you can set any environment variable you want - some modules |
612 |
(such as L<Coro> or L<EV>) use environment variables for further tweaking. |
613 |
|
614 |
=item C<PERL_VERSION> |
615 |
|
616 |
The perl version to install - default is currently C<5.12.2>, but C<5.8.9> |
617 |
is also a good choice (5.8.9 is much smaller than 5.12.2, while 5.10.1 is |
618 |
about as big as 5.12.2). |
619 |
|
620 |
=item C<PERL_PREFIX> |
621 |
|
622 |
The prefix where perl gets installed (default: F<$STATICPERL/perl>), |
623 |
i.e. where the F<bin> and F<lib> subdirectories will end up. |
624 |
|
625 |
=item C<PERL_CONFIGURE> |
626 |
|
627 |
Additional Configure options - these are simply passed to the perl |
628 |
Configure script. For example, if you wanted to enable dynamic loading, |
629 |
you could pass C<-Dusedl>. To enable ithreads (Why would you want that |
630 |
insanity? Don't! Use L<forks> instead!) you would pass C<-Duseithreads> |
631 |
and so on. |
632 |
|
633 |
More commonly, you would either activate 64 bit integer support |
634 |
(C<-Duse64bitint>), or disable large files support (-Uuselargefiles), to |
635 |
reduce filesize further. |
636 |
|
637 |
=item C<PERL_CPPFLAGS>, C<PERL_OPTIMIZE>, C<PERL_LDFLAGS>, C<PERL_LIBS> |
638 |
|
639 |
These flags are passed to perl's F<Configure> script, and are generally |
640 |
optimised for small size (at the cost of performance). Since they also |
641 |
contain subtle workarounds around various build issues, changing these |
642 |
usually requires understanding their default values - best look at the top |
643 |
of the F<staticperl> script for more info on these. |
644 |
|
645 |
=back |
646 |
|
647 |
=head4 Variables you probably I<do not want> to override |
648 |
|
649 |
=over 4 |
650 |
|
651 |
=item C<MKBUNDLE> |
652 |
|
653 |
Where F<staticperl> writes the C<mkbundle> command to |
654 |
(default: F<$STATICPERL/mkbundle>). |
655 |
|
656 |
=item C<STATICPERL_MODULES> |
657 |
|
658 |
Additional modules needed by C<mkbundle> - should therefore not be changed |
659 |
unless you know what you are doing. |
660 |
|
661 |
=back |
662 |
|
663 |
=head3 OVERRIDABLE HOOKS |
664 |
|
665 |
In addition to environment variables, it is possible to provide some |
666 |
shell functions that are called at specific times. To provide your own |
667 |
commands, just define the corresponding function. |
668 |
|
669 |
Example: install extra modules from CPAN and from some directories |
670 |
at F<staticperl install> time. |
671 |
|
672 |
postinstall() { |
673 |
rm -rf lib/threads* # weg mit Schaden |
674 |
instcpan IO::AIO EV |
675 |
instsrc ~/src/AnyEvent |
676 |
instsrc ~/src/XML-Sablotron-1.0100001 |
677 |
instcpan Anyevent::AIO AnyEvent::HTTPD |
678 |
} |
679 |
|
680 |
=over 4 |
681 |
|
682 |
=item preconfigure |
683 |
|
684 |
Called just before running F<./Configur> in the perl source |
685 |
directory. Current working directory is the perl source directory. |
686 |
|
687 |
This can be used to set any C<PERL_xxx> variables, which might be costly |
688 |
to compute. |
689 |
|
690 |
=item postconfigure |
691 |
|
692 |
Called after configuring, but before building perl. Current working |
693 |
directory is the perl source directory. |
694 |
|
695 |
Could be used to tailor/patch config.sh (followed by F<sh Configure -S>) |
696 |
or do any other modifications. |
697 |
|
698 |
=item postbuild |
699 |
|
700 |
Called after building, but before installing perl. Current working |
701 |
directory is the perl source directory. |
702 |
|
703 |
I have no clue what this could be used for - tell me. |
704 |
|
705 |
=item postinstall |
706 |
|
707 |
Called after perl and any extra modules have been installed in C<$PREFIX>, |
708 |
but before setting the "installation O.K." flag. |
709 |
|
710 |
The current working directory is C<$PREFIX>, but maybe you should not rely |
711 |
on that. |
712 |
|
713 |
This hook is most useful to customise the installation, by deleting files, |
714 |
or installing extra modules using the C<instcpan> or C<instsrc> functions. |
715 |
|
716 |
The script must return with a zero exit status, or the installation will |
717 |
fail. |
718 |
|
719 |
=back |
720 |
|
721 |
=head1 ANATOMY OF A BUNDLE |
722 |
|
723 |
When not building a new perl binary, C<mkbundle> will leave a number of |
724 |
files in the current working directory, which can be used to embed a perl |
725 |
interpreter in your program. |
726 |
|
727 |
Intimate knowledge of L<perlembed> and preferably some experience with |
728 |
embedding perl is highly recommended. |
729 |
|
730 |
C<mkperl> (or the C<--perl> option) basically does this to link the new |
731 |
interpreter (it also adds a main program to F<bundle.>): |
732 |
|
733 |
$Config{cc} $(cat bundle.ccopts) -o perl bundle.c $(cat bundle.ldopts) |
734 |
|
735 |
=over 4 |
736 |
|
737 |
=item bundle.h |
738 |
|
739 |
A header file that contains the prototypes of the few symbols "exported" |
740 |
by bundle.c, and also exposes the perl headers to the application. |
741 |
|
742 |
=over 4 |
743 |
|
744 |
=item staticperl_init () |
745 |
|
746 |
Initialises the perl interpreter. You can use the normal perl functions |
747 |
after calling this function, for example, to define extra functions or |
748 |
to load a .pm file that contains some initialisation code, or the main |
749 |
program function: |
750 |
|
751 |
XS (xsfunction) |
752 |
{ |
753 |
dXSARGS; |
754 |
|
755 |
// now we have items, ST(i) etc. |
756 |
} |
757 |
|
758 |
static void |
759 |
run_myapp(void) |
760 |
{ |
761 |
staticperl_init (); |
762 |
newXSproto ("myapp::xsfunction", xsfunction, __FILE__, "$$;$"); |
763 |
eval_pv ("require myapp::main", 1); // executes "myapp/main.pm" |
764 |
} |
765 |
|
766 |
=item staticperl_xs_init (pTHX) |
767 |
|
768 |
Sometimes you need direct control over C<perl_parse> and C<perl_run>, in |
769 |
which case you do not want to use C<staticperl_init> but call them on your |
770 |
own. |
771 |
|
772 |
Then you need this function - either pass it directly as the C<xs_init> |
773 |
function to C<perl_parse>, or call it from your own C<xs_init> function. |
774 |
|
775 |
=item staticperl_cleanup () |
776 |
|
777 |
In the unlikely case that you want to destroy the perl interpreter, here |
778 |
is the corresponding function. |
779 |
|
780 |
=item PerlInterpreter *staticperl |
781 |
|
782 |
The perl interpreter pointer used by staticperl. Not normally so useful, |
783 |
but there it is. |
784 |
|
785 |
=back |
786 |
|
787 |
=item bundle.ccopts |
788 |
|
789 |
Contains the compiler options required to compile at least F<bundle.c> and |
790 |
any file that includes F<bundle.h> - you should probably use it in your |
791 |
C<CFLAGS>. |
792 |
|
793 |
=item bundle.ldopts |
794 |
|
795 |
The linker options needed to link the final program. |
796 |
|
797 |
=back |
798 |
|
799 |
=head1 RUNTIME FUNCTIONALITY |
800 |
|
801 |
Binaries created with C<mkbundle>/C<mkperl> contain extra functions, which |
802 |
are required to access the bundled perl sources, but might be useful for |
803 |
other purposes. |
804 |
|
805 |
In addition, for the embedded loading of perl files to work, F<staticperl> |
806 |
overrides the C<@INC> array. |
807 |
|
808 |
=over 4 |
809 |
|
810 |
=item $file = staticperl::find $path |
811 |
|
812 |
Returns the data associated with the given C<$path> |
813 |
(e.g. C<Digest/MD5.pm>, C<auto/POSIX/autosplit.ix>), which is basically |
814 |
the UNIX path relative to the perl library directory. |
815 |
|
816 |
Returns C<undef> if the file isn't embedded. |
817 |
|
818 |
=item @paths = staticperl::list |
819 |
|
820 |
Returns the list of all paths embedded in this binary. |
821 |
|
822 |
=back |
823 |
|
824 |
=head1 FULLY STATIC BINARIES - BUILDROOT |
825 |
|
826 |
To make truly static (Linux-) libraries, you might want to have a look at |
827 |
buildroot (L<http://buildroot.uclibc.org/>). |
828 |
|
829 |
Buildroot is primarily meant to set up a cross-compile environment (which |
830 |
is not so useful as perl doesn't quite like cross compiles), but it can also compile |
831 |
a chroot environment where you can use F<staticperl>. |
832 |
|
833 |
To do so, download buildroot, and enable "Build options => development |
834 |
files in target filesystem" and optionally "Build options => gcc |
835 |
optimization level (optimize for size)". At the time of writing, I had |
836 |
good experiences with GCC 4.4.x but not GCC 4.5. |
837 |
|
838 |
To minimise code size, I used C<-pipe -ffunction-sections -fdata-sections |
839 |
-finline-limit=8 -fno-builtin-strlen -mtune=i386>. The C<-mtune=i386> |
840 |
doesn't decrease codesize much, but it makes the file much more |
841 |
compressible. |
842 |
|
843 |
If you don't need Coro or threads, you can go with "linuxthreads.old" (or |
844 |
no thread support). For Coro, it is highly recommended to switch to a |
845 |
uClibc newer than 0.9.31 (at the time of this writing, I used the 20101201 |
846 |
snapshot) and enable NPTL, otherwise Coro needs to be configured with the |
847 |
ultra-slow pthreads backend to work around linuxthreads bugs (it also uses |
848 |
twice the address space needed for stacks). |
849 |
|
850 |
If you use C<linuxthreads.old>, then you should also be aware that |
851 |
uClibc shares C<errno> between all threads when statically linking. See |
852 |
L<http://lists.uclibc.org/pipermail/uclibc/2010-June/044157.html> for a |
853 |
workaround (And L<https://bugs.uclibc.org/2089> for discussion). |
854 |
|
855 |
C<ccache> support is also recommended, especially if you want |
856 |
to play around with buildroot options. Enabling the C<miniperl> |
857 |
package will probably enable all options required for a successful |
858 |
perl build. F<staticperl> itself additionally needs either C<wget> |
859 |
(recommended, for CPAN) or C<curl>. |
860 |
|
861 |
As for shells, busybox should provide all that is needed, but the default |
862 |
busybox configuration doesn't include F<comm> which is needed by perl - |
863 |
either make a custom busybox config, or compile coreutils. |
864 |
|
865 |
For the latter route, you might find that bash has some bugs that keep |
866 |
it from working properly in a chroot - either use dash (and link it to |
867 |
F</bin/sh> inside the chroot) or link busybox to F</bin/sh>, using it's |
868 |
built-in ash shell. |
869 |
|
870 |
Finally, you need F</dev/null> inside the chroot for many scripts to work |
871 |
- F<cp /dev/null output/target/dev> or bind-mounting your F</dev> will |
872 |
both provide this. |
873 |
|
874 |
After you have compiled and set up your buildroot target, you can copy |
875 |
F<staticperl> from the C<App::Staticperl> distribution or from your |
876 |
perl f<bin> directory (if you installed it) into the F<output/target> |
877 |
filesystem, chroot inside and run it. |
878 |
|
879 |
=head1 RECIPES / SPECIFIC MODULES |
880 |
|
881 |
This section contains some common(?) recipes and information about |
882 |
problems with some common modules or perl constructs that require extra |
883 |
files to be included. |
884 |
|
885 |
=head2 MODULES |
886 |
|
887 |
=over 4 |
888 |
|
889 |
=item utf8 |
890 |
|
891 |
Some functionality in the utf8 module, such as swash handling (used |
892 |
for unicode character ranges in regexes) is implemented in the |
893 |
C<"utf8_heavy.pl"> library: |
894 |
|
895 |
-M'"utf8_heavy.pl"' |
896 |
|
897 |
Many Unicode properties in turn are defined in separate modules, |
898 |
such as C<"unicore/Heavy.pl"> and more specific data tables such as |
899 |
C<"unicore/To/Digit.pl"> or C<"unicore/lib/Perl/Word.pl">. These tables |
900 |
are big (7MB uncompressed, although F<staticperl> contains special |
901 |
handling for those files), so including them on demand by your application |
902 |
only might pay off. |
903 |
|
904 |
To simply include the whole unicode database, use: |
905 |
|
906 |
--incglob '/unicore/*.pl' |
907 |
|
908 |
=item AnyEvent |
909 |
|
910 |
AnyEvent needs a backend implementation that it will load in a delayed |
911 |
fashion. The L<AnyEvent::Impl::Perl> backend is the default choice |
912 |
for AnyEvent if it can't find anything else, and is usually a safe |
913 |
fallback. If you plan to use e.g. L<EV> (L<POE>...), then you need to |
914 |
include the L<AnyEvent::Impl::EV> (L<AnyEvent::Impl::POE>...) backend as |
915 |
well. |
916 |
|
917 |
If you want to handle IRIs or IDNs (L<AnyEvent::Util> punycode and idn |
918 |
functions), you also need to include C<"AnyEvent/Util/idna.pl"> and |
919 |
C<"AnyEvent/Util/uts46data.pl">. |
920 |
|
921 |
=item Carp |
922 |
|
923 |
Carp had (in older versions of perl) a dependency on L<Carp::Heavy>. As of |
924 |
perl 5.12.2 (maybe earlier), this dependency no longer exists. |
925 |
|
926 |
=item Config |
927 |
|
928 |
The F<perl -V> switch (as well as many modules) needs L<Config>, which in |
929 |
turn might need L<"Config_heavy.pl">. Including the latter gives you |
930 |
both. |
931 |
|
932 |
=item Term::ReadLine::Perl |
933 |
|
934 |
Also needs L<Term::ReadLine::readline>. |
935 |
|
936 |
=item URI |
937 |
|
938 |
URI implements schemes as separate modules - the generic URL scheme is |
939 |
implemented in L<URI::_generic>, HTTP is implemented in L<URI::http>. If |
940 |
you need to use any of these schemes, you should include these manually. |
941 |
|
942 |
=back |
943 |
|
944 |
=head2 RECIPES |
945 |
|
946 |
=over 4 |
947 |
|
948 |
=item Linking everything in |
949 |
|
950 |
To link just about everything installed in the perl library into a new |
951 |
perl, try this: |
952 |
|
953 |
staticperl mkperl --strip ppi --incglob '*' |
954 |
|
955 |
=item Getting rid of netdb function |
956 |
|
957 |
The perl core has lots of netdb functions (C<getnetbyname>, C<getgrent> |
958 |
and so on) that few applications use. You can avoid compiling them in by |
959 |
putting the following fragment into a C<preconfigure> hook: |
960 |
|
961 |
preconfigure() { |
962 |
for sym in \ |
963 |
d_getgrnam_r d_endgrent d_endgrent_r d_endhent \ |
964 |
d_endhostent_r d_endnent d_endnetent_r d_endpent \ |
965 |
d_endprotoent_r d_endpwent d_endpwent_r d_endsent \ |
966 |
d_endservent_r d_getgrent d_getgrent_r d_getgrgid_r \ |
967 |
d_getgrnam_r d_gethbyaddr d_gethent d_getsbyport \ |
968 |
d_gethostbyaddr_r d_gethostbyname_r d_gethostent_r \ |
969 |
d_getlogin_r d_getnbyaddr d_getnbyname d_getnent \ |
970 |
d_getnetbyaddr_r d_getnetbyname_r d_getnetent_r \ |
971 |
d_getpent d_getpbyname d_getpbynumber d_getprotobyname_r \ |
972 |
d_getprotobynumber_r d_getprotoent_r d_getpwent \ |
973 |
d_getpwent_r d_getpwnam_r d_getpwuid_r d_getsent \ |
974 |
d_getservbyname_r d_getservbyport_r d_getservent_r \ |
975 |
d_getspnam_r d_getsbyname |
976 |
# d_gethbyname |
977 |
do |
978 |
PERL_CONFIGURE="$PERL_CONFIGURE -U$sym" |
979 |
done |
980 |
} |
981 |
|
982 |
This mostly gains space when linking staticaly, as the functions will |
983 |
liekly not be linked in. The gain for dynamically-linked binaries is |
984 |
smaller. |
985 |
|
986 |
Also, this leaves C<gethostbyname> in - not only is it actually used |
987 |
often, the L<Socket> module also exposes it, so leaving it out usually |
988 |
gains little. Why Socket exposes a C function that is in the core already |
989 |
is anybody's guess. |
990 |
|
991 |
=back |
992 |
|
993 |
=head1 AUTHOR |
994 |
|
995 |
Marc Lehmann <schmorp@schmorp.de> |
996 |
http://software.schmorp.de/pkg/staticperl.html |