This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
typos in perlboot.pod (from Randal L. Schwartz <merlyn@stonehenge.com>)
[perl5.git] / pod / perlcompile.pod
CommitLineData
54a137f5
GS
1=head1 NAME
2
3perlcompile - Introduction to the Perl Compiler-Translator
4
5=head1 DESCRIPTION
6
7Perl has always had a compiler: your source is compiled into an
8internal form (a parse tree) which is then optimized before being
9run. Since version 5.005, Perl has shipped with a module
10capable of inspecting the optimized parse tree (C<B>), and this has
11been used to write many useful utilities, including a module that lets
12you turn your Perl into C source code that can be compiled into an
13native executable.
14
15The C<B> module provides access to the parse tree, and other modules
16("back ends") do things with the tree. Some write it out as
17bytecode, C source code, or a semi-human-readable text. Another
18traverses the parse tree to build a cross-reference of which
19subroutines, formats, and variables are used where. Another checks
20your code for dubious constructs. Yet another back end dumps the
21parse tree back out as Perl source, acting as a source code beautifier
22or deobfuscator.
23
24Because its original purpose was to be a way to produce C code
25corresponding to a Perl program, and in turn a native executable, the
26C<B> module and its associated back ends are known as "the
27compiler", even though they don't really compile anything.
28Different parts of the compiler are more accurately a "translator",
29or an "inspector", but people want Perl to have a "compiler
30option" not an "inspector gadget". What can you do?
31
32This document covers the use of the Perl compiler: which modules
33it comprises, how to use the most important of the back end modules,
34what problems there are, and how to work around them.
35
36=head2 Layout
37
38The compiler back ends are in the C<B::> hierarchy, and the front-end
39(the module that you, the user of the compiler, will sometimes
40interact with) is the O module. Some back ends (e.g., C<B::C>) have
41programs (e.g., I<perlcc>) to hide the modules' complexity.
42
43Here are the important back ends to know about, with their status
44expressed as a number from 0 (outline for later implementation) to
4510 (if there's a bug in it, we're very surprised):
46
47=over 4
48
49=item B::Bytecode
50
51Stores the parse tree in a machine-independent format, suitable
52for later reloading through the ByteLoader module. Status: 5 (some
53things work, some things don't, some things are untested).
54
55=item B::C
56
57Creates a C source file containing code to rebuild the parse tree
58and resume the interpreter. Status: 6 (many things work adequately,
59including programs using Tk).
60
61=item B::CC
62
63Creates a C source file corresponding to the run time code path in
64the parse tree. This is the closest to a Perl-to-C translator there
65is, but the code it generates is almost incomprehensible because it
66translates the parse tree into a giant switch structure that
67manipulates Perl structures. Eventual goal is to reduce (given
68sufficient type information in the Perl program) some of the
69Perl data structure manipulations into manipulations of C-level
70ints, floats, etc. Status: 5 (some things work, including
71uncomplicated Tk examples).
72
73=item B::Lint
74
75Complains if it finds dubious constructs in your source code. Status:
766 (it works adequately, but only has a very limited number of areas
77that it checks).
78
79=item B::Deparse
80
81Recreates the Perl source, making an attempt to format it coherently.
82Status: 8 (it works nicely, but a few obscure things are missing).
83
84=item B::Xref
85
86Reports on the declaration and use of subroutines and variables.
87Status: 8 (it works nicely, but still has a few lingering bugs).
88
89=back
90
91=head1 Using The Back Ends
92
93The following sections describe how to use the various compiler back
94ends. They're presented roughly in order of maturity, so that the
95most stable and proven back ends are described first, and the most
96experimental and incomplete back ends are described last.
97
98The O module automatically enabled the B<-c> flag to Perl, which
99prevents Perl from executing your code once it has been compiled.
100This is why all the back ends print:
101
102 myperlprogram syntax OK
103
104before producing any other output.
105
106=head2 The Cross Referencing Back End (B::Xref)
107
108The cross referencing back end produces a report on your program,
109breaking down declarations and uses of subroutines and variables (and
110formats) by file and subroutine. For instance, here's part of the
111report from the I<pod2man> program that comes with Perl:
112
113 Subroutine clear_noremap
114 Package (lexical)
115 $ready_to_print i1069, 1079
116 Package main
117 $& 1086
118 $. 1086
119 $0 1086
120 $1 1087
121 $2 1085, 1085
122 $3 1085, 1085
123 $ARGV 1086
124 %HTML_Escapes 1085, 1085
125
126This shows the variables used in the subroutine C<clear_noremap>. The
127variable C<$ready_to_print> is a my() (lexical) variable,
128B<i>ntroduced (first declared with my()) on line 1069, and used on
129line 1079. The variable C<$&> from the main package is used on 1086,
130and so on.
131
132A line number may be prefixed by a single letter:
133
134=over 4
135
136=item i
137
138Lexical variable introduced (declared with my()) for the first time.
139
140=item &
141
142Subroutine or method call.
143
144=item s
145
146Subroutine defined.
147
148=item r
149
150Format defined.
151
152=back
153
154The most useful option the cross referencer has is to save the report
155to a separate file. For instance, to save the report on
156I<myperlprogram> to the file I<report>:
157
158 $ perl -MO=Xref,-oreport myperlprogram
159
160=head2 The Decompiling Back End
161
162The Deparse back end turns your Perl source back into Perl source. It
163can reformat along the way, making it useful as a de-obfuscator. The
164most basic way to use it is:
165
166 $ perl -MO=Deparse myperlprogram
167
168You'll notice immediately that Perl has no idea of how to paragraph
169your code. You'll have to separate chunks of code from each other
170with newlines by hand. However, watch what it will do with
171one-liners:
172
173 $ perl -MO=Deparse -e '$op=shift||die "usage: $0
174 code [...]";chomp(@ARGV=<>)unless@ARGV; for(@ARGV){$was=$_;eval$op;
175 die$@ if$@; rename$was,$_ unless$was eq $_}'
176 -e syntax OK
177 $op = shift @ARGV || die("usage: $0 code [...]");
178 chomp(@ARGV = <ARGV>) unless @ARGV;
179 foreach $_ (@ARGV) {
180 $was = $_;
181 eval $op;
182 die $@ if $@;
183 rename $was, $_ unless $was eq $_;
184 }
185
186(this is the I<rename> program that comes in the I<eg/> directory
187of the Perl source distribution).
188
189The decompiler has several options for the code it generates. For
190instance, you can set the size of each indent from 4 (as above) to
1912 with:
192
193 $ perl -MO=Deparse,-si2 myperlprogram
194
195The B<-p> option adds parentheses where normally they are omitted:
196
197 $ perl -MO=Deparse -e 'print "Hello, world\n"'
198 -e syntax OK
199 print "Hello, world\n";
200 $ perl -MO=Deparse,-p -e 'print "Hello, world\n"'
201 -e syntax OK
202 print("Hello, world\n");
203
204See L<B::Deparse> for more information on the formatting options.
205
206=head2 The Lint Back End (B::Lint)
207
208The lint back end inspects programs for poor style. One programmer's
209bad style is another programmer's useful tool, so options let you
210select what is complained about.
211
212To run the style checker across your source code:
213
214 $ perl -MO=Lint myperlprogram
215
216To disable context checks and undefined subroutines:
217
218 $ perl -MO=Lint,-context,-undefined-subs myperlprogram
219
220See L<B::Lint> for information on the options.
221
222=head2 The Simple C Back End
223
224This module saves the internal compiled state of your Perl program
225to a C source file, which can be turned into a native executable
226for that particular platform using a C compiler. The resulting
227program links against the Perl interpreter library, so it
228will not save you disk space (unless you build Perl with a shared
229library) or program size. It may, however, save you startup time.
230
231The C<perlcc> tool generates such executables by default.
232
233 perlcc myperlprogram.pl
234
235=head2 The Bytecode Back End
236
237This back end is only useful if you also have a way to load and
238execute the bytecode that it produces. The ByteLoader module provides
239this functionality.
240
241To turn a Perl program into executable byte code, you can use C<perlcc>
242with the C<-b> switch:
243
244 perlcc -b myperlprogram.pl
245
246The byte code is machine independent, so once you have a compiled
247module or program, it is as portable as Perl source (assuming that
248the user of the module or program has a modern-enough Perl interpreter
249to decode the byte code).
250
251See B<B::Bytecode> for information on options to control the
252optimization and nature of the code generated by the Bytecode module.
253
254=head2 The Optimized C Back End
255
256The optimized C back end will turn your Perl program's run time
257code-path into an equivalent (but optimized) C program that manipulates
258the Perl data structures directly. The program will still link against
259the Perl interpreter library, to allow for eval(), C<s///e>,
260C<require>, etc.
261
262The C<perlcc> tool generates such executables when using the -opt
263switch. To compile a Perl program (ending in C<.pl>
264or C<.p>):
265
266 perlcc -opt myperlprogram.pl
267
268To produce a shared library from a Perl module (ending in C<.pm>):
269
270 perlcc -opt Myperlmodule.pm
271
272For more information, see L<perlcc> and L<B::CC>.
273
274=over 4
275
276=item B
277
278This module is the introspective ("reflective" in Java terms)
279module, which allows a Perl program to inspect its innards. The
280back end modules all use this module to gain access to the compiled
281parse tree. You, the user of a back end module, will not need to
282interact with B.
283
284=item O
285
286This module is the front-end to the compiler's back ends. Normally
287called something like this:
288
289 $ perl -MO=Deparse myperlprogram
290
291This is like saying C<use O 'Deparse'> in your Perl program.
292
293=item B::Asmdata
294
295This module is used by the B::Assembler module, which is in turn used
296by the B::Bytecode module, which stores a parse-tree as
297bytecode for later loading. It's not a back end itself, but rather a
298component of a back end.
299
300=item B::Assembler
301
302This module turns a parse-tree into data suitable for storing
303and later decoding back into a parse-tree. It's not a back end
304itself, but rather a component of a back end. It's used by the
305I<assemble> program that produces bytecode.
306
307=item B::Bblock
308
309This module is used by the B::CC back end. It walks "basic blocks",
310whatever they may be.
311
312=item B::Bytecode
313
314This module is a back end that generates bytecode from a
315program's parse tree. This bytecode is written to a file, from where
316it can later be reconstructed back into a parse tree. The goal is to
317do the expensive program compilation once, save the interpreter's
318state into a file, and then restore the state from the file when the
319program is to be executed. See L</"The Bytecode Back End">
320for details about usage.
321
322=item B::C
323
324This module writes out C code corresponding to the parse tree and
325other interpreter internal structures. You compile the corresponding
326C file, and get an executable file that will restore the internal
327structures and the Perl interpreter will begin running the
328program. See L</"The Simple C Back End"> for details about usage.
329
330=item B::CC
331
332This module writes out C code corresponding to your program's
333operations. Unlike the B::C module, which merely stores the
334interpreter and its state in a C program, the B::CC module makes a
335C program that does not involve the interpreter. As a consequence,
336programs translated into C by B::CC can execute faster than normal
337interpreted programs. See L</"The Optimized C Back End"> for
338details about usage.
339
340=item B::Debug
341
342This module dumps the Perl parse tree in verbose detail to STDOUT.
343It's useful for people who are writing their own back end, or who
344are learning about the Perl internals. It's not useful to the
345average programmer.
346
347=item B::Deparse
348
349This module produces Perl source code from the compiled parse tree.
350It is useful in debugging and deconstructing other people's code,
351also as a pretty-printer for your own source. See
352L</"The Decompiling Back End"> for details about usage.
353
354=item B::Disassembler
355
356This module turns bytecode back into a parse tree. It's not a back
357end itself, but rather a component of a back end. It's used by the
358I<disassemble> program that comes with the bytecode.
359
360=item B::Lint
361
362This module inspects the compiled form of your source code for things
363which, while some people frown on them, aren't necessarily bad enough
364to justify a warning. For instance, use of an array in scalar context
365without explicitly saying C<scalar(@array)> is something that Lint
366can identify. See L</"The Lint Back End"> for details about usage.
367
368=item B::Showlex
369
370This module prints out the my() variables used in a function or a
371file. To gt a list of the my() variables used in the subroutine
372mysub() defined in the file myperlprogram:
373
374 $ perl -MO=Showlex,mysub myperlprogram
375
376To gt a list of the my() variables used in the file myperlprogram:
377
378 $ perl -MO=Showlex myperlprogram
379
380[BROKEN]
381
382=item B::Stackobj
383
384This module is used by the B::CC module. It's not a back end itself,
385but rather a component of a back end.
386
387=item B::Stash
388
389This module is used by the L<perlcc> program, which compiles a module
390into an executable. B::Stash prints the symbol tables in use by a
391program, and is used to prevent B::CC from producing C code for the
392B::* and O modules. It's not a back end itself, but rather a
393component of a back end.
394
395=item B::Terse
396
397This module prints the contents of the parse tree, but without as much
398information as B::Debug. For comparison, C<print "Hello, world.">
399produced 96 lines of output from B::Debug, but only 6 from B::Terse.
400
401This module is useful for people who are writing their own back end,
402or who are learning about the Perl internals. It's not useful to the
403average programmer.
404
405=item B::Xref
406
407This module prints a report on where the variables, subroutines, and
408formats are defined and used within a program and the modules it
409loads. See L</"The Cross Referencing Back End"> for details about
410usage.
411
a45bd81d 412=back
54a137f5
GS
413
414=head1 KNOWN PROBLEMS
415
416The simple C backend currently only saves typeglobs with alphanumeric
417names.
418
419The optimized C backend outputs code for more modules than it should
420(e.g., DirHandle). It also has little hope of properly handling
421C<goto LABEL> outside the running subroutine (C<goto &sub> is ok).
422C<goto LABEL> currently does not work at all in this backend.
423It also creates a huge initialization function that gives
424C compilers headaches. Splitting the initialization function gives
425better results. Other problems include: unsigned math does not
426work correctly; some opcodes are handled incorrectly by default
427opcode handling mechanism.
428
429BEGIN{} blocks are executed while compiling your code. Any external
430state that is initialized in BEGIN{}, such as opening files, initiating
431database connections etc., do not behave properly. To work around
432this, Perl has an INIT{} block that corresponds to code being executed
433before your program begins running but after your program has finished
434being compiled. Execution order: BEGIN{}, (possible save of state
435through compiler back-end), INIT{}, program runs, END{}.
436
437=head1 AUTHOR
438
439This document was originally written by Nathan Torkington, and is now
440maintained by the perl5-porters mailing list
441I<perl5-porters@perl.org>.
442
443=cut