Penguin
Recent Changes
Annotated edit history of perlfaq3(1) version 2 showing authors affecting page license. View with all changes included.
Rev Author # Line
1 perry 1 PERLFAQ3
2 !!!PERLFAQ3
3 NAME
4 DESCRIPTION
5 AUTHOR AND COPYRIGHT
6 ----
7 !!NAME
8
9
10 perlfaq3 - Programming Tools ($Revision: 1.38 $, $Date: 1999/05/23 16:08:30 $)
11 !!DESCRIPTION
12
13
14 This section of the FAQ answers questions
15 related to programmer tools and programming
16 support.
17
18
19 __How do I do (anything)?__
20
21
22 Have you looked at CPAN (see perlfaq2)? The
23 chances are that someone has already written a module that
24 can solve your problem. Have you read the appropriate man
25 pages? Here's a brief index:
26
27
28 Basics perldata, perlvar, perlsyn, perlop, perlsub
29 Execution perlrun, perldebug
30 Functions perlfunc
31 Objects perlref, perlmod, perlobj, perltie
32 Data Structures perlref, perllol, perldsc
33 Modules perlmod, perlmodlib, perlsub
34 Regexes perlre, perlfunc, perlop, perllocale
35 Moving to perl5 perltrap, perl
36 Linking w/C perlxstut, perlxs, perlcall, perlguts, perlembed
37 Various http://www.perl.com/CPAN/doc/FMTEYEWTK/index.html
38 (not a man-page but still useful)
39 A crude table of contents for the Perl man page set is found in perltoc.
40
41
42 __How can I use Perl interactively?__
43
44
45 The typical approach uses the Perl debugger, described in
46 the perldebug(1) man page, on an ``empty'' program,
47 like this:
48
49
50 perl -de 42
51 Now just type in any legal Perl code, and it will be immediately evaluated. You can also examine the symbol table, get stack backtraces, check variable values, set breakpoints, and other operations typically found in symbolic debuggers.
52
53
54 __Is there a Perl shell?__
55
56
57 In general, no. The Shell.pm module (distributed with Perl)
58 makes Perl try commands which aren't part of the Perl
59 language as shell commands. perlsh from the source
60 distribution is simplistic and uninteresting, but may still
61 be what you want.
62
63
64 __How do I debug my Perl programs?__
65
66
67 Have you tried use warnings or used -w?
68 They enable warnings to detect dubious
69 practices.
70
71
72 Have you tried use strict? It prevents you from
73 using symbolic references, makes you predeclare any
74 subroutines that you call as bare words, and (probably most
75 importantly) forces you to predeclare your variables with
76 my, our, or use
77 vars.
78
79
80 Did you check the return values of each and every system
81 call? The operating system (and thus Perl) tells you whether
82 they worked, and if not why.
83
84
85 open(FH,
86 Did you read perltrap? It's full of gotchas for old and new Perl programmers and even has sections for those of you who are upgrading from languages like ''awk'' and ''C''.
87
88
89 Have you tried the Perl debugger, described in perldebug?
90 You can step through your program and see what it's doing
91 and thus work out why what it's doing isn't what it should
92 be doing.
93
94
95 __How do I profile my Perl programs?__
96
97
98 You should get the Devel::DProf module from the standard
99 distribution (or separately on CPAN ) and
100 also use Benchmark.pm from the standard distribution. The
101 Benchmark module lets you time specific portions of your
102 code, while Devel::DProf gives detailed breakdowns of where
103 your code spends its time.
104
105
106 Here's a sample use of Benchmark:
107
108
109 use Benchmark;
110 @junk = `cat /etc/motd`;
111 $count = 10_000;
112 timethese($count, {
113 'map' =
114 This is what it prints (on one machine--your results will be dependent on your hardware, operating system, and the load on your machine):
115
116
117 Benchmark: timing 10000 iterations of for, map...
118 for: 4 secs ( 3.97 usr 0.01 sys = 3.98 cpu)
119 map: 6 secs ( 4.97 usr 0.00 sys = 4.97 cpu)
120 Be aware that a good benchmark is very hard to write. It only tests the data you give it and proves little about the differing complexities of contrasting algorithms.
121
122
123 __How do I cross-reference my Perl
124 programs?__
125
126
127 The B::Xref module, shipped with the new, alpha-release Perl
128 compiler (not the general distribution prior to the 5.005
129 release), can be used to generate cross-reference reports
130 for Perl programs.
131
132
133 perl -MO=Xref[[,OPTIONS] scriptname.plx
134
135
136 __Is there a pretty-printer (formatter) for
137 Perl?__
138
139
140 There is no program that will reformat Perl as much as
141 indent(1) does for C. The complex feedback between
142 the scanner and the parser (this feedback is what confuses
143 the vgrind and emacs programs) makes it challenging at best
144 to write a stand-alone Perl parser.
145
146
147 Of course, if you simply follow the guidelines in perlstyle,
148 you shouldn't need to reformat. The habit of formatting your
149 code as you write it will help prevent bugs. Your editor can
150 and should help you with this. The perl-mode or newer
151 cperl-mode for emacs can provide remarkable amounts of help
152 with most (but not all) code, and even less programmable
153 editors can provide significant assistance. Tom swears by
154 the following settings in vi and its clones:
155
156
157 set ai sw=4
158 map! ^O {^M}^[[O^T
159 Now put that in your ''.exrc'' file (replacing the caret characters with control characters) and away you go. In insert mode, ^T is for indenting, ^D is for undenting, and ^O is for blockdenting-- as it were. If you haven't used the last one, you're missing a lot. A more complete example, with comments, can be found at http://www.perl.com/CPAN-local/authors/id/TOMC/scripts/toms.exrc.gz
160
161
162 If you are used to using the ''vgrind'' program for
163 printing out nice code to a laser printer, you can take a
164 stab at this using
165 http://www.perl.com/CPAN/doc/misc/tips/working.vgrind.entry,
166 but the results are not particularly satisfying for
167 sophisticated code.
168
169
170 The a2ps at http://www.infres.enst.fr/%7Edemaille/a2ps/ does
171 lots of things related to generating nicely printed output
172 of documents.
173
174
175 __Is there a ctags for Perl?__
176
177
178 There's a simple one at
179 http://www.perl.com/CPAN/authors/id/TOMC/scripts/ptags.gz
180 which may do the trick. And if not, it's easy to hack into
181 what you want.
182
183
184 __Is there an IDE or Windows Perl
185 Editor?__
186
187
188 Perl programs are just plain text, so any editor will
189 do.
190
191
192 If you're on Unix, you already have an IDE--Unix itself. The
193 UNIX philosophy is the philosophy of several
194 small tools that each do one thing and do it well. It's like
195 a carpenter's toolbox.
196
197
198 If you want a Windows IDE , check the
199 following:
200
201
202 CodeMagicCD
203
204
205 http://www.codemagiccd.com/
206
207
208 Komodo
209
210
211 !ActiveState's cross-platform, multi-language
212 IDE has Perl support, including a regular
213 expression debugger and remote debugging
214 (http://www.!ActiveState.com/Products/Komodo/index.html).
215 (Visual Perl, a Visual Studio.NET plug-in is currently
216 (early 2001) in beta
217 (http://www.!ActiveState.com/Products/!VisualPerl/index.html)).
218
219
220 The Object System
221
222
223 (http://www.castlelink.co.uk/object_system/) is a Perl web
224 applications development IDE .
225
226
227 !PerlBuilder
228
229
230 (http://www.solutionsoft.com/perl.htm) is an integrated
231 development environment for Windows that supports Perl
232 development.
233
234
235 Perl code magic
236
237
238 (http://www.petes-place.com/codemagic.html).
239
240
241 visiPerl+
242
243
244 http://helpconsulting.net/visiperl/, from Help
245 Consulting.
246
247
248 For editors: if you're on Unix you probably have vi or a vi
249 clone already, and possibly an emacs too, so you may not
250 need to download anything. In any emacs the cperl-mode (M-x
251 cperl-mode) gives you perhaps the best available Perl
252 editing mode in any editor.
253
254
255 For Windows editors: you can download an Emacs
256
257
258 GNU Emacs
259
260
261 http://www.gnu.org/software/emacs/windows/ntemacs.html
262
263
264 MicroEMACS
265
266
267 http://members.nbci.com/uemacs/
268
269
270 XEmacs
271
272
273 http://www.xemacs.org/Download/index.html
274
275
276 or a vi clone such as
277
278
279 Elvis
280
281
282 ftp://ftp.cs.pdx.edu/pub/elvis/
283 http://www.fh-wedel.de/elvis/
284
285
286 Vile
287
288
289 http://vile.cx/
290
291
292 Vim
293
294
295 http://www.vim.org/
296
297
298 win32: http://www.cs.vu.nl/%7Etmgil/vi.html
299
300
301 For vi lovers in general, Windows or elsewhere:
302 http://www.thomer.com/thomer/vi/vi.html.
303
304
305 nvi (http://www.bostic.com/vi/, available from
306 CPAN in src/misc/) is yet another vi clone,
307 unfortunately not available for Windows, but in
308 UNIX platforms you might be interested in
309 trying it out, firstly because strictly speaking it is not a
310 vi clone, it is the real vi, or the new incarnation of it,
311 and secondly because you can embed Perl inside it to use
312 Perl as the scripting language. nvi is not alone in this,
313 though: at least also vim and vile offer an embedded
314 Perl.
315
316
317 The following are Win32 multilanguage editor/IDESs that
318 support Perl:
319
320
321 Codewright
322
323
324 http://www.starbase.com/
325
326
327 !MultiEdit
328
329
330 http://www.!MultiEdit.com/
331
332
333 !SlickEdit
334
335
336 http://www.slickedit.com/
337
338
339 There is also a toyedit Text widget based editor written in
340 Perl that is distributed with the Tk module on
341 CPAN . The ptkdb
342 (http://world.std.com/~aep/ptkdb/) is a Perl/tk based
343 debugger that acts as a development environment of sorts.
344 Perl Composer
345 (http://perlcomposer.sourceforge.net/vperl.html) is an
346 IDE for Perl/Tk GUI
347 creation.
348
349
350 In addition to an editor/IDE you might be interested in a
351 more powerful shell environment for Win32. Your options
352 include
353
354
355 Bash
356
357
358 from the Cygwin package
359 (http://sources.redhat.com/cygwin/)
360
361
362 Ksh
363
364
365 from the MKS Toolkit (http://www.mks.com/),
366 or the Bourne shell of the U/WIN environment
367 (http://www.research.att.com/sw/tools/uwin/)
368
369
370 Tcsh
371
372
373 ftp://ftp.astron.com/pub/tcsh/, see also
374 http://www.primate.wisc.edu/software/csh-tcsh-book/
375
376
377 Zsh
378
379
380 ftp://ftp.blarg.net/users/amol/zsh/, see also
381 http://www.zsh.org/
382
383
384 MKS and U/WIN are commercial (U/WIN is free
385 for educational and research purposes), Cygwin is covered by
386 the GNU Public License (but that shouldn't
387 matter for Perl use). The Cygwin, MKS , and
388 U/WIN all contain (in addition to the shells) a
389 comprehensive set of standard UNIX toolkit
390 utilities.
391
392
393 If you're transferring text files between Unix and Windows
394 using FTP be sure to transfer them in
395 ASCII mode so the ends of lines are
396 appropriately converted.
397
398
399 On Mac OS the !MacPerl Application comes with
400 a simple 32k text editor that behaves like a rudimentary
401 IDE . In contrast to the !MacPerl Application
402 the MPW Perl tool can make use of the
403 MPW Shell itself as an editor (with no 32k
404 limit).
405
406
407 BBEdit and BBEdit Lite
408
409
410 are text editors for Mac OS that have a Perl
411 sensitivity mode (http://web.barebones.com/).
412
413
414 Alpha
415
416
417 is an editor, written and extensible in Tcl, that
418 nonetheless has built in support for several popular markup
419 and programming languages including Perl and
420 HTML (http://alpha.olm.net/).
421
422
423 Pepper and Pe are programming language sensitive text
424 editors for Mac OS X and BeOS respectively
425 (http://www.hekkelman.com/).
426
427
428 __Where can I get Perl macros for vi?__
429
430
431 For a complete version of Tom Christiansen's vi
432 configuration file, see
433 http://www.perl.com/CPAN/authors/Tom_Christiansen/scripts/toms.exrc.gz
434 , the standard benchmark file for vi emulators. The file
435 runs best with nvi, the current version of vi out of
436 Berkeley, which incidentally can be built with an embedded
437 Perl interpreter--see
438 http://www.perl.com/CPAN/src/misc.
439
440
441 __Where can I get perl-mode for emacs?__
442
443
444 Since Emacs version 19 patchlevel 22 or so, there have been
445 both a perl-mode.el and support for the Perl debugger built
446 in. These should come with the standard Emacs 19
447 distribution.
448
449
450 In the Perl source directory, you'll find a directory called
451 ``emacs'', which contains a cperl-mode that color-codes
452 keywords, provides context-sensitive help, and other nifty
453 things.
454
455
456 Note that the perl-mode of emacs will have fits with
457 (single quote), and mess up
458 the indentation and highlighting. You are probably using
459 in new Perl code anyway, so
460 this shouldn't be an issue.
461
462
463 __How can I use curses with Perl?__
464
465
466 The Curses module from CPAN provides a
467 dynamically loadable object module interface to a curses
468 library. A small demo can be found at the directory
469 http://www.perl.com/CPAN/authors/Tom_Christiansen/scripts/rep;
470 this program repeats a command and updates the screen as
471 needed, rendering __rep ps axu__ similar to
472 __top__.
473
474
475 __How can I use X or Tk with Perl?__
476
477
478 Tk is a completely Perl-based, object-oriented interface to
479 the Tk toolkit that doesn't force you to use Tcl just to get
480 at Tk. Sx is an interface to the Athena Widget set. Both are
481 available from CPAN . See the directory
482 http://www.perl.com/CPAN/modules/by-category/08_User_Interfaces/
483
484
485 Invaluable for Perl/Tk programming are the Perl/Tk
486 FAQ at
487 http://w4.lns.cornell.edu/%7Epvhp/ptk/ptkTOC.html , the
488 Perl/Tk Reference Guide available at
489 http://www.perl.com/CPAN-local/authors/Stephen_O_Lidie/ ,
490 and the online manpages at
491 http://www-users.cs.umn.edu/%7Eamundson/perl/perltk/toc.html
492 .
493
494
495 __How can I generate simple menus without using
496 CGI or Tk?__
497
498
499 The
500 http://www.perl.com/CPAN/authors/id/SKUNZ/perlmenu.v4.0.tar.gz
501 module, which is curses-based, can help with
502 this.
503
504
505 __What is undump?__
506
507
508 See the next question on ``How can I make my Perl program
509 run faster?''
510
511
512 __How can I make my Perl program run
513 faster?__
514
515
516 The best way to do this is to come up with a better
517 algorithm. This can often make a dramatic difference. Jon
518 Bentley's book ``Programming Pearls'' (that's not a
519 misspelling!) has some good tips on optimization, too.
520 Advice on benchmarking boils down to: benchmark and profile
521 to make sure you're optimizing the right part, look for
522 better algorithms instead of microtuning your code, and when
523 all else fails consider just buying faster
524 hardware.
525
526
527 A different approach is to autoload seldom-used Perl code.
528 See the !AutoSplit and !AutoLoader modules in the standard
529 distribution for that. Or you could locate the bottleneck
530 and think about writing just that part in C, the way we used
531 to take bottlenecks in C code and write them in assembler.
532 Similar to rewriting in C, modules that have critical
533 sections can be written in C (for instance, the
534 PDL module from CPAN
535 ).
536
537
538 In some cases, it may be worth it to use the backend
539 compiler to produce byte code (saving compilation time) or
540 compile into C, which will certainly save compilation time
541 and sometimes a small amount (but not much) execution time.
542 See the question about compiling your Perl programs for more
543 on the compiler--the wins aren't as obvious as you'd
544 hope.
545
546
547 If you're currently linking your perl executable to a shared
548 ''libc.so'', you can often gain a 10-25% performance
549 benefit by rebuilding it to link with a static libc.a
550 instead. This will make a bigger perl executable, but your
551 Perl programs (and programmers) may thank you for it. See
552 the ''INSTALL'' file in the source
553 distribution for more information.
554
555
556 Unsubstantiated reports allege that Perl interpreters that
557 use sfio outperform those that don't (for I/O intensive
558 applications). To try this, see the
559 ''INSTALL'' file in the source
560 distribution, especially the ``Selecting File I/O
561 mechanisms'' section.
562
563
564 The undump program was an old attempt to speed up your Perl
565 program by storing the already-compiled form to disk. This
566 is no longer a viable option, as it only worked on a few
567 architectures, and wasn't a good solution
568 anyway.
569
570
571 __How can I make my Perl program take less
572 memory?__
573
574
575 When it comes to time-space tradeoffs, Perl nearly always
576 prefers to throw memory at a problem. Scalars in Perl use
577 more memory than strings in C, arrays take more than that,
578 and hashes use even more. While there's still a lot to be
579 done, recent releases have been addressing these issues. For
580 example, as of 5.004, duplicate hash keys are shared amongst
581 all hashes using them, so require no
582 reallocation.
583
584
585 In some cases, using ''substr()'' or ''vec()'' to
586 simulate arrays can be highly beneficial. For example, an
587 array of a thousand booleans will take at least 20,000 bytes
588 of space, but it can be turned into one 125-byte bit
589 vector--a considerable memory savings. The standard
590 Tie::!SubstrHash module can also help for certain types of
591 data structure. If you're working with specialist data
592 structures (matrices, for instance) modules that implement
593 these in C may use less memory than equivalent Perl
594 modules.
595
596
597 Another thing to try is learning whether your Perl was
598 compiled with the system malloc or with Perl's builtin
599 malloc. Whichever one it is, try using the other one and see
600 whether this makes a difference. Information about malloc is
601 in the ''INSTALL'' file in the source
602 distribution. You can find out whether you are using perl's
603 malloc by typing perl -V:usemymalloc.
604
605
606 __Is it unsafe to return a pointer to local
607 data?__
608
609
610 No, Perl's garbage collection system takes care of
611 this.
612
613
614 sub makeone {
615 my @a = ( 1 .. 10 );
616 return @a;
617 }
618 for $i ( 1 .. 10 ) {
619 push @many, makeone();
620 }
621 print $many[[4][[5],
622 print
623
624
625 __How can I free an array or hash so my program
626 shrinks?__
627
628
629 You can't. On most operating systems, memory allocated to a
630 program can never be returned to the system. That's why
631 long-running programs sometimes re-exec themselves. Some
632 operating systems (notably, FreeBSD and Linux) allegedly
633 reclaim large chunks of memory that is no longer used, but
634 it doesn't appear to happen with Perl (yet). The Mac appears
635 to be the only platform that will reliably (albeit, slowly)
636 return memory to the OS .
637
638
639 We've had reports that on Linux (Redhat 5.1) on Intel,
640 undef $scalar will return memory to the system,
641 while on Solaris 2.6 it won't. In general, try it yourself
642 and see.
643
644
645 However, judicious use of ''my()'' on your variables will
646 help make sure that they go out of scope so that Perl can
647 free up that space for use in other parts of your program. A
648 global variable, of course, never goes out of scope, so you
649 can't get its space automatically reclaimed, although
650 ''undef()''ing and/or ''delete()''ing it will achieve
651 the same effect. In general, memory allocation and
652 de-allocation isn't something you can or should be worrying
653 about much in Perl, but even this capability (preallocation
654 of data types) is in the works.
655
656
657 __How can I make my CGI script more
658 efficient?__
659
660
661 Beyond the normal measures described to make general Perl
662 programs faster or smaller, a CGI program has
663 additional issues. It may be run several times per second.
664 Given that each time it runs it will need to be re-compiled
665 and will often allocate a megabyte or more of system memory,
666 this can be a killer. Compiling into C __isn't going to
667 help you__ because the process start-up overhead is where
668 the bottleneck is.
669
670
671 There are two popular ways to avoid this overhead. One
672 solution involves running the Apache HTTP
673 server (available from http://www.apache.org/) with either
674 of the mod_perl or mod_fastcgi plugin modules.
675
676
677 With mod_perl and the Apache::Registry module (distributed
678 with mod_perl), httpd will run with an embedded Perl
679 interpreter which pre-compiles your script and then executes
680 it within the same address space without forking. The Apache
681 extension also gives Perl access to the internal server
682 API , so modules written in Perl can do just
683 about anything a module written in C can. For more on
684 mod_perl, see http://perl.apache.org/
685
686
687 With the FCGI module (from
688 CPAN ) and the mod_fastcgi module (available
689 from http://www.fastcgi.com/) each of your Perl programs
690 becomes a permanent CGI daemon
691 process.
692
693
694 Both of these solutions can have far-reaching effects on
695 your system and on the way you write your CGI
696 programs, so investigate them with care.
697
698
699 See
700 http://www.perl.com/CPAN/modules/by-category/15_World_Wide_Web_HTML_HTTP_CGI/
701 .
702
703
704 A non-free, commercial product, ``The Velocity Engine for
705 Perl'', (http://www.binevolve.com/ or
706 http://www.binevolve.com/velocigen/ ) might also be worth
707 looking at. It will allow you to increase the performance of
708 your Perl programs, running programs up to 25 times faster
709 than normal CGI Perl when running in
710 persistent Perl mode or 4 to 5 times faster without any
711 modification to your existing CGI programs.
712 Fully functional evaluation copies are available from the
713 web site.
714
715
716 __How can I hide the source for my Perl
717 program?__
718
719
720 Delete it. :-) Seriously, there are a number of (mostly
721 unsatisfactory) solutions with varying levels of
722 ``security''.
723
724
725 First of all, however, you ''can't'' take away read
726 permission, because the source code has to be readable in
727 order to be compiled and interpreted. (That doesn't mean
728 that a CGI script's source is readable by
729 people on the web, though--only by people with access to the
730 filesystem.) So you have to leave the permissions at the
731 socially friendly 0755 level.
732
733
734 Some people regard this as a security problem. If your
735 program does insecure things and relies on people not
736 knowing how to exploit those insecurities, it is not secure.
737 It is often possible for someone to determine the insecure
738 things and exploit them without viewing the source. Security
739 through obscurity, the name for hiding your bugs instead of
740 fixing them, is little security indeed.
741
742
743 You can try using encryption via source filters (Filter::*
744 from CPAN ), but any decent programmer will
745 be able to decrypt it. You can try using the byte code
746 compiler and interpreter described below, but the curious
747 might still be able to de-compile it. You can try using the
748 native-code compiler described below, but crackers might be
749 able to disassemble it. These pose varying degrees of
750 difficulty to people wanting to get at your code, but none
751 can definitively conceal it (true of every language, not
752 just Perl).
753
754
755 If you're concerned about people profiting from your code,
756 then the bottom line is that nothing but a restrictive
757 license will give you legal security. License your software
758 and pepper it with threatening statements like ``This is
759 unpublished proprietary software of XYZ Corp.
760 Your access to it does not give you permission to use it
761 blah blah blah.'' We are not lawyers, of course, so you
762 should see a lawyer if you want to be sure your license's
763 wording will stand up in court.
764
765
766 __How can I compile my Perl program into byte code or
767 C?__
768
769
770 Malcolm Beattie has written a multifunction backend
771 compiler, available from CPAN , that can do
772 both these things. It is included in the perl5.005 release,
773 but is still considered experimental. This means it's fun to
774 play with if you're a programmer but not really for people
775 looking for turn-key solutions.
776
777
778 Merely compiling into C does not in and of itself guarantee
779 that your code will run very much faster. That's because
780 except for lucky cases where a lot of native type
781 inferencing is possible, the normal Perl run-time system is
782 still present and so your program will take just as long to
783 run and be just as big. Most programs save little more than
784 compilation time, leaving execution no more than 10-30%
785 faster. A few rare programs actually benefit significantly
786 (even running several times faster), but this takes some
787 tweaking of your code.
788
789
790 You'll probably be astonished to learn that the current
791 version of the compiler generates a compiled form of your
792 script whose executable is just as big as the original perl
793 executable, and then some. That's because as currently
794 written, all programs are prepared for a full ''eval()''
795 statement. You can tremendously reduce this cost by building
796 a shared ''libperl.so'' library and linking against that.
797 See the ''INSTALL'' podfile in the Perl
798 source distribution for details. If you link your main perl
799 binary with this, it will make it minuscule. For example, on
800 one author's system, ''/usr/bin/perl'' is only 11k in
801 size!
802
803
804 In general, the compiler will do nothing to make a Perl
805 program smaller, faster, more portable, or more secure. In
806 fact, it can make your situation worse. The executable will
807 be bigger, your VM system may take longer to
808 load the whole thing, the binary is fragile and hard to fix,
809 and compilation never stopped software piracy in the form of
810 crackers, viruses, or bootleggers. The real advantage of the
811 compiler is merely packaging, and once you see the size of
812 what it makes (well, unless you use a shared
813 ''libperl.so''), you'll probably want a complete Perl
814 install anyway.
815
816
817 __How can I compile Perl into Java?__
818
819
820 You can also integrate Java and Perl with the Perl Resource
821 Kit from O'Reilly and Associates. See
822 http://www.oreilly.com/catalog/prkunix/ .
823
824
825 Perl 5.6 comes with Java Perl Lingo, or JPL .
826 JPL , still in development, allows Perl code
827 to be called from Java. See jpl/README in the Perl source
828 tree.
829
830
831 __How can I get__ #!perl __to work on [[
832 MS-DOS ,NT,...]?__
833
834
835 For OS/2 just use
836
837
838 extproc perl -S -your_switches
839 as the first line in *.cmd file (-S due to a bug in cmd.exe's `extproc' handling). For DOS one should first invent a corresponding batch file and codify it in ALTERNATIVE_SHEBANG (see the ''INSTALL'' file in the source distribution for more information).
840
841
842 The Win95/NT installation, when using the !ActiveState port
843 of Perl, will modify the Registry to associate the
844 .pl extension with the perl interpreter. If you
845 install another port, perhaps even building your own
846 Win95/NT Perl from the standard sources by using a Windows
847 port of gcc (e.g., with cygwin or mingw32), then you'll have
848 to modify the Registry yourself. In addition to associating
849 .pl with the interpreter, NT people
850 can use: SET PATHEXT=%PATHEXT%;.PL to let them run
851 the program install-linux.pl merely by typing
852 install-linux.
853
854
855 Macintosh Perl programs will have the appropriate Creator
856 and Type, so that double-clicking them will invoke the Perl
857 application.
858
859
860 ''IMPORTANT !'': Whatever you do,
861 PLEASE don't get frustrated, and just throw
862 the perl interpreter into your cgi-bin directory, in order
863 to get your programs working for a web server. This is an
864 EXTREMELY big security risk. Take the time to
865 figure out how to do it correctly.
866
867
868 __Can I write useful Perl programs on the command
869 line?__
870
871
872 Yes. Read perlrun for more information. Some examples
873 follow. (These assume standard Unix shell quoting
874 rules.)
875
876
877 # sum first and last fields
878 perl -lane 'print $F[[0] + $F[[-1]' *
879 # identify text files
880 perl -le 'for(@ARGV) {print if -f
881 # remove (most) comments from C program
882 perl -0777 -pe 's{/*.*?*/}{}gs' foo.c
883 # make file a month younger than today, defeating reaper daemons
884 perl -e '$X=24*60*60; utime(time(),time() + 30 * $X,@ARGV)' *
885 # find first unused uid
886 perl -le '$i++ while getpwuid($i); print $i'
887 # display reasonable manpath
888 echo $PATH perl -nl -072 -e '
889 s![[^/+]*$!man!
890 OK , the last one was actually an Obfuscated Perl Contest entry. :-)
891
892
893 __Why don't Perl one-liners work on my DOS/Mac/VMS
894 system?__
895
896
897 The problem is usually that the command interpreters on
898 those systems have rather different ideas about quoting than
899 the Unix shells under which the one-liners were created. On
900 some systems, you may have to change single-quotes to double
901 ones, which you must ''NOT'' do on Unix or
902 Plan9 systems. You might also have to change a single % to a
903 %%.
904
905
906 For example:
907
908
909 # Unix
910 perl -e 'print
911 # DOS, etc.
912 perl -e
913 # Mac
914 print
915 # VMS
916 perl -e
917 The problem is that none of these examples are reliable: they depend on the command interpreter. Under Unix, the first two often work. Under DOS , it's entirely possible that neither works. If 4DOS was the command shell, you'd probably have better luck like this:
918
919
920 perl -e
921 Under the Mac, it depends which environment you are using. The !MacPerl shell, or MPW , is much like Unix shells in its support for several quoting variants, except that it makes free use of the Mac's non-ASCII characters as control characters.
922
923
924 Using ''qq()'', q(), and ''qx()'', instead of ``double
925 quotes'', 'single quotes', and `backticks`, may make
926 one-liners easier to write.
927
928
929 There is no general solution to all of this. It is a mess,
930 pure and simple. Sucks to be away from Unix, huh?
931 :-)
932
933
934 [[Some of this answer was contributed by Kenneth
935 Albanowski.]
936
937
938 __Where can I learn about CGI or Web
939 programming in Perl?__
940
941
942 For modules, get the CGI or
943 LWP modules from CPAN . For
944 textbooks, see the two especially dedicated to web stuff in
945 the question on books. For problems and questions related to
946 the web, like ``Why do I get 500 Errors'' or ``Why doesn't
947 it run from the browser right when it runs fine on the
948 command line'', see these sources:
949
950
951 WWW Security FAQ
952 http://www.w3.org/Security/Faq/
953 Web FAQ
954 http://www.boutell.com/faq/
955 CGI FAQ
956 http://www.webthing.com/tutorials/cgifaq.html
957 HTTP Spec
958 http://www.w3.org/pub/WWW/Protocols/HTTP/
959 HTML Spec
960 http://www.w3.org/TR/REC-html40/
961 http://www.w3.org/pub/WWW/!MarkUp/
962 CGI Spec
963 http://www.w3.org/CGI/
964 CGI Security FAQ
965 http://www.go2net.com/people/paulp/cgi-security/safe-cgi.txt
966
967
968 __Where can I learn about object-oriented Perl
969 programming?__
970
971
972 A good place to start is perltoot, and you can use perlobj,
973 perlboot, and perlbot for reference. Perltoot didn't come
974 out until the 5.004 release; you can get a copy (in pod,
975 html, or postscript) from
976 http://www.perl.com/CPAN/doc/FMTEYEWTK/ .
977
978
979 __Where can I learn about linking C with Perl? [[h2xs,
980 xsubpp]__
981
982
983 If you want to call C from Perl, start with perlxstut,
984 moving on to perlxs, xsubpp, and perlguts. If you want to
985 call Perl from C, then read perlembed, perlcall, and
986 perlguts. Don't forget that you can learn a lot from looking
987 at how the authors of existing extension modules wrote their
988 code and solved their problems.
989
990
991 __I've read perlembed, perlguts, etc., but I can't embed
992 perl in my C program; what am I doing
993 wrong?__
994
995
996 Download the !ExtUtils::Embed kit from CPAN
997 and run `make test'. If the tests pass, read the pods again
998 and again and again. If they fail, see perlbug and send a
999 bug report with the output of make test
1000 TEST_VERBOSE=1 along with perl
1001 -V.
1002
1003
1004 __When I tried to run my script, I got this message. What
1005 does it mean?__
1006
1007
1008 A complete list of Perl's error messages and warnings with
1009 explanatory text can be found in perldiag. You can also use
1010 the splain program (distributed with Perl) to explain the
1011 error messages:
1012
1013
1014 perl program 2
1015 or change your program to explain the messages for you:
1016
1017
1018 use diagnostics;
1019 or
1020
1021
1022 use diagnostics -verbose;
1023
1024
1025 __What's !MakeMaker?__
1026
1027
1028 This module (part of the standard Perl distribution) is
1029 designed to write a Makefile for an extension module from a
1030 Makefile.PL. For more information, see
1031 !ExtUtils::!MakeMaker.
1032 !!AUTHOR AND COPYRIGHT
1033
1034
1035 Copyright (c) 1997-1999 Tom Christiansen and Nathan
1036 Torkington. All rights reserved.
1037
1038
1039 When included as an integrated part of the Standard
1040 Distribution of Perl or of its documentation (printed or
1041 otherwise), this works is covered under Perl's Artistic
1042 License. For separate distributions of all or part of this
1043 FAQ outside of that, see
1044 perlfaq.
1045
1046
1047 Irrespective of its distribution, all code examples here are
1048 in the public domain. You are permitted and encouraged to
1049 use this code and any derivatives thereof in your own
1050 programs for fun or for profit as you see fit. A simple
1051 comment in the code giving credit to the FAQ
1052 would be courteous but is not required.
1053 ----
This page is a man page (or other imported legacy content). We are unable to automatically determine the license status of this page.