3 perlmod - Perl modules (packages)
9 Perl provides a mechanism for alternative namespaces to protect packages
10 from stomping on each other's variables. In fact, apart from certain
11 magical variables, there's really no such thing as a global variable in
12 Perl. The package statement declares the compilation unit as being in the
13 given namespace. The scope of the package declaration is from the
14 declaration itself through the end of the enclosing block (the same scope
15 as the local() operator). All further unqualified dynamic identifiers
16 will be in this namespace. A package statement only affects dynamic
17 variables--including those you've used local() on--but I<not> lexical
18 variables created with my(). Typically it would be the first declaration
19 in a file to be included by the C<require> or C<use> operator. You can
20 switch into a package in more than one place; it merely influences which
21 symbol table is used by the compiler for the rest of that block. You can
22 refer to variables and filehandles in other packages by prefixing the
23 identifier with the package name and a double colon:
24 C<$Package::Variable>. If the package name is null, the C<main> package
25 is assumed. That is, C<$::sail> is equivalent to C<$main::sail>.
27 (The old package delimiter was a single quote, but double colon
28 is now the preferred delimiter, in part because it's more readable
29 to humans, and in part because it's more readable to B<emacs> macros.
30 It also makes C++ programmers feel like they know what's going on.)
32 Packages may be nested inside other packages: C<$OUTER::INNER::var>. This
33 implies nothing about the order of name lookups, however. All symbols
34 are either local to the current package, or must be fully qualified
35 from the outer package name down. For instance, there is nowhere
36 within package C<OUTER> that C<$INNER::var> refers to C<$OUTER::INNER::var>.
37 It would treat package C<INNER> as a totally separate global package.
39 Only identifiers starting with letters (or underscore) are stored in a
40 package's symbol table. All other symbols are kept in package C<main>,
41 including all of the punctuation variables like $_. In addition, the
42 identifiers STDIN, STDOUT, STDERR, ARGV, ARGVOUT, ENV, INC and SIG are
43 forced to be in package C<main>, even when used for other purposes than
44 their built-in one. Note also that, if you have a package called C<m>,
45 C<s> or C<y>, then you can't use the qualified form of an identifier
46 because it will be interpreted instead as a pattern match, a substitution,
49 (Variables beginning with underscore used to be forced into package
50 main, but we decided it was more useful for package writers to be able
51 to use leading underscore to indicate private variables and method names.
52 $_ is still global though.)
54 Eval()ed strings are compiled in the package in which the eval() was
55 compiled. (Assignments to C<$SIG{}>, however, assume the signal
56 handler specified is in the C<main> package. Qualify the signal handler
57 name if you wish to have a signal handler in a package.) For an
58 example, examine F<perldb.pl> in the Perl library. It initially switches
59 to the C<DB> package so that the debugger doesn't interfere with variables
60 in the script you are trying to debug. At various points, however, it
61 temporarily switches back to the C<main> package to evaluate various
62 expressions in the context of the C<main> package (or wherever you came
63 from). See L<perldebug>.
65 See L<perlsub> for other scoping issues related to my() and local(),
66 or L<perlref> regarding closures.
70 The symbol table for a package happens to be stored in the associative
71 array of that name appended with two colons. The main symbol table's
72 name is thus C<%main::>, or C<%::> for short. Likewise symbol table for
73 the nested package mentioned earlier is named C<%OUTER::INNER::>.
75 The value in each entry of the associative array is what you are referring
76 to when you use the C<*name> typeglob notation. In fact, the following
77 have the same effect, though the first is more efficient because it does
78 the symbol table lookups at compile time:
80 local(*main::foo) = *main::bar; local($main::{'foo'}) =
83 You can use this to print out all the variables in a package, for
84 instance. Here is F<dumpvar.pl> from the Perl library:
89 local(*stab) = eval("*${package}::");
90 while (($key,$val) = each(%stab)) {
93 print "\$$key = '$entry'\n";
98 foreach $num ($[ .. $#entry) {
99 print " $num\t'",$entry[$num],"'\n";
104 if ($key ne "${package}::" && defined %entry) {
105 print "\%$key = (\n";
106 foreach $key (sort keys(%entry)) {
107 print " $key\t'",$entry{$key},"'\n";
114 Note that even though the subroutine is compiled in package C<dumpvar>,
115 the name of the subroutine is qualified so that its name is inserted
116 into package C<main>.
118 Assignment to a typeglob performs an aliasing operation, i.e.,
122 causes variables, subroutines and file handles accessible via the
123 identifier C<richard> to also be accessible via the identifier C<dick>. If
124 you only want to alias a particular variable or subroutine, you can
125 assign a reference instead:
129 makes $richard and $dick the same variable, but leaves
130 @richard and @dick as separate arrays. Tricky, eh?
132 This mechanism may be used to pass and return cheap references
133 into or from subroutines if you won't want to copy the whole
137 *some_hash = fn( \%another_hash );
139 local *hashsym = shift;
140 # now use %hashsym normally, and you
141 # will affect the caller's %another_hash
142 my %nhash = (); # do what you want
146 On return, the reference wil overwrite the hash slot in the
147 symbol table specified by the *some_hash typeglob. This
148 is a somewhat tricky way of passing around references cheaply
149 when you won't want to have to remember to dereference variables
152 Another use of symbol tables is for making "constant" scalars.
154 *PI = \3.14159265358979;
156 Now you cannot alter $PI, which is probably a good thing all in all.
158 You can say C<*foo{PACKAGE}> and C<*foo{NAME}> to find out what name and
159 package the *foo symbol table entry comes from. This may be useful
160 in a subroutine which is passed typeglobs as arguments
162 sub identify_typeglob {
164 print 'You gave me ', *{$glob}{PACKAGE}, '::', *{$glob}{NAME}, "\n";
166 identify_typeglob *foo;
167 identify_typeglob *bar::baz;
171 You gave me main::foo
174 The *foo{THING} notation can also be used to obtain references to the
175 individual elements of *foo, see L<perlref>.
177 =head2 Package Constructors and Destructors
179 There are two special subroutine definitions that function as package
180 constructors and destructors. These are the C<BEGIN> and C<END>
181 routines. The C<sub> is optional for these routines.
183 A C<BEGIN> subroutine is executed as soon as possible, that is, the
184 moment it is completely defined, even before the rest of the containing
185 file is parsed. You may have multiple C<BEGIN> blocks within a
186 file--they will execute in order of definition. Because a C<BEGIN>
187 block executes immediately, it can pull in definitions of subroutines
188 and such from other files in time to be visible to the rest of the
191 An C<END> subroutine is executed as late as possible, that is, when the
192 interpreter is being exited, even if it is exiting as a result of a
193 die() function. (But not if it's is being blown out of the water by a
194 signal--you have to trap that yourself (if you can).) You may have
195 multiple C<END> blocks within a file--they will execute in reverse
196 order of definition; that is: last in, first out (LIFO).
198 Inside an C<END> subroutine C<$?> contains the value that the script is
199 going to pass to C<exit()>. You can modify C<$?> to change the exit
200 value of the script. Beware of changing C<$?> by accident (eg, by
201 running something via C<system>).
203 Note that when you use the B<-n> and B<-p> switches to Perl, C<BEGIN>
204 and C<END> work just as they do in B<awk>, as a degenerate case.
208 There is no special class syntax in Perl, but a package may function
209 as a class if it provides subroutines that function as methods. Such a
210 package may also derive some of its methods from another class package
211 by listing the other package name in its @ISA array.
213 For more on this, see L<perlobj>.
217 A module is just a package that is defined in a library file of
218 the same name, and is designed to be reusable. It may do this by
219 providing a mechanism for exporting some of its symbols into the symbol
220 table of any package using it. Or it may function as a class
221 definition and make its semantics available implicitly through method
222 calls on the class and its objects, without explicit exportation of any
223 symbols. Or it can do a little of both.
225 For example, to start a normal module called Fred, create
226 a file called Fred.pm and put this at the start of it:
231 @EXPORT = qw(func1 func2);
232 @EXPORT_OK = qw($sally @listabob %harry func3);
234 Then go on to declare and use your variables in functions
235 without any qualifications.
236 See L<Exporter> and the I<Perl Modules File> for details on
237 mechanics and style issues in module creation.
239 Perl modules are included into your program by saying
247 This is exactly equivalent to
249 BEGIN { require "Module.pm"; import Module; }
253 BEGIN { require "Module.pm"; import Module LIST; }
259 is exactly equivalent to
261 BEGIN { require "Module.pm"; }
263 All Perl module files have the extension F<.pm>. C<use> assumes this so
264 that you don't have to spell out "F<Module.pm>" in quotes. This also
265 helps to differentiate new modules from old F<.pl> and F<.ph> files.
266 Module names are also capitalized unless they're functioning as pragmas,
267 "Pragmas" are in effect compiler directives, and are sometimes called
268 "pragmatic modules" (or even "pragmata" if you're a classicist).
270 Because the C<use> statement implies a C<BEGIN> block, the importation
271 of semantics happens at the moment the C<use> statement is compiled,
272 before the rest of the file is compiled. This is how it is able
273 to function as a pragma mechanism, and also how modules are able to
274 declare subroutines that are then visible as list operators for
275 the rest of the current file. This will not work if you use C<require>
276 instead of C<use>. With require you can get into this problem:
278 require Cwd; # make Cwd:: accessible
279 $here = Cwd::getcwd();
281 use Cwd; # import names from Cwd::
284 require Cwd; # make Cwd:: accessible
285 $here = getcwd(); # oops! no main::getcwd()
287 In general C<use Module ();> is recommended over C<require Module;>.
289 Perl packages may be nested inside other package names, so we can have
290 package names containing C<::>. But if we used that package name
291 directly as a filename it would makes for unwieldy or impossible
292 filenames on some systems. Therefore, if a module's name is, say,
293 C<Text::Soundex>, then its definition is actually found in the library
294 file F<Text/Soundex.pm>.
296 Perl modules always have a F<.pm> file, but there may also be dynamically
297 linked executables or autoloaded subroutine definitions associated with
298 the module. If so, these will be entirely transparent to the user of
299 the module. It is the responsibility of the F<.pm> file to load (or
300 arrange to autoload) any additional functionality. The POSIX module
301 happens to do both dynamic loading and autoloading, but the user can
302 just say C<use POSIX> to get it all.
304 For more information on writing extension modules, see L<perlxs>
309 Perl does not enforce private and public parts of its modules as you may
310 have been used to in other languages like C++, Ada, or Modula-17. Perl
311 doesn't have an infatuation with enforced privacy. It would prefer
312 that you stayed out of its living room because you weren't invited, not
313 because it has a shotgun.
315 The module and its user have a contract, part of which is common law,
316 and part of which is "written". Part of the common law contract is
317 that a module doesn't pollute any namespace it wasn't asked to. The
318 written contract for the module (AKA documentation) may make other
319 provisions. But then you know when you C<use RedefineTheWorld> that
320 you're redefining the world and willing to take the consequences.
322 =head1 THE PERL MODULE LIBRARY
324 A number of modules are included the the Perl distribution. These are
325 described below, and all end in F<.pm>. You may also discover files in
326 the library directory that end in either F<.pl> or F<.ph>. These are old
327 libraries supplied so that old programs that use them still run. The
328 F<.pl> files will all eventually be converted into standard modules, and
329 the F<.ph> files made by B<h2ph> will probably end up as extension modules
330 made by B<h2xs>. (Some F<.ph> values may already be available through the
331 POSIX module.) The B<pl2pm> file in the distribution may help in your
332 conversion, but it's just a mechanical process, so is far from bulletproof.
334 =head2 Pragmatic Modules
336 They work somewhat like pragmas in that they tend to affect the compilation of
337 your program, and thus will usually only work well when used within a
338 C<use>, or C<no>. Most of these are locally scoped, so an inner BLOCK
339 may countermand any of these by saying:
344 which lasts until the end of that BLOCK.
346 Unlike the pragrmas that effect the C<$^H> hints variable, the C<use
347 vars> and C<use subs> declarations are not BLOCK-scoped. They allow
348 you to pre-declare a variables or subroutines within a particular
349 <I>file</I> rather than just a block. Such declarations are effective
350 for the entire file for which they were declared. You cannot rescind
351 them with C<no vars> or C<no subs>.
353 The following pragmas are defined (and have their own documentation).
359 force verbose warning diagnostics
363 compute arithmetic in integer instead of double
367 request less of something from the compiler
371 manipulate @INC at compile time
375 restrict unsafe operations when compiling
379 package for overloading perl operations
383 enable simple signal handling
387 restrict unsafe constructs
395 predeclare global variable names
399 =head2 Standard Modules
401 Standard, bundled modules are all expected to behave in a well-defined
402 manner with respect to namespace pollution because they use the
403 Exporter module. See their own documentation for details.
409 provide framework for multiple DBMs
413 load functions only on demand
417 split a package for autoloading
421 benchmark running times of code
425 warn of errors (from perspective of caller)
429 access Perl configuration information
433 get pathname of current working directory
437 access to Berkeley DB
439 =item Devel::SelfStubber
441 generate stubs for a SelfLoading module
445 supply object methods for directory handles
449 Dynamically load C libraries into Perl code
453 use nice English (or awk) names for ugly punctuation variables
457 import environment variables
461 implements default import method for modules
463 =item ExtUtils::Embed
465 Utilities for embedding Perl in C/C++ applications
467 =item ExtUtils::Install
469 install files from here to there
471 =item ExtUtils::Liblist
473 determine libraries to use and how to use them
475 =item ExtUtils::MakeMaker
477 create an extension Makefile
479 =item ExtUtils::Manifest
481 utilities to write and check a MANIFEST file
483 =item ExtUtils::Miniperl
485 write the C code for perlmain.c
487 =item ExtUtils::Mkbootstrap
489 make a bootstrap file for use by DynaLoader
491 =item ExtUtils::Mksymlists
493 write linker options files for dynamic extension
495 =item ExtUtils::MM_OS2
497 methods to override UN*X behaviour in ExtUtils::MakeMaker
499 =item ExtUtils::MM_Unix
501 methods used by ExtUtils::MakeMaker
503 =item ExtUtils::MM_VMS
505 methods to override UN*X behaviour in ExtUtils::MakeMaker
507 =item ExtUtils::testlib
509 add blib/* directories to @INC
513 replace functions with equivalents which succeed or die
517 load the C Fcntl.h defines
521 parse file specifications
525 keep more files open than the system permits
527 =item File::CheckTree
529 run many filetest checks on a tree
533 Copy files or filehandles
541 supply object methods for filehandles
545 create or remove a series of directories
549 locate directory of original perl script
553 access to the gdbm library.
557 extended processing of command line options
561 process single-character switches with switch clustering
565 compare 8-bit scalar data according to the current locale
569 load various IO modules
573 supply object methods for filehandles
577 supply object methods for I/O handles
581 supply object methods for pipes
585 supply seek based methods for I/O objects
589 OO interface to the select system call
593 object interface to socket communications
597 open a process for both reading and writing
601 open a process for reading, writing, and error handling
605 arbitrary length float math package
609 arbitrary size integer math package
613 complex numbers and associated mathematical functions
617 tied access to ndbm files
621 check a host for upness
625 disable named opcodes when compiling perl code
629 convert POD data to formatted ASCII text
633 interface to IEEE Std 1003.1
637 compile and execute code in restricted compartments
641 tied access to sdbm files
645 search for key in dictionary file
649 save and restore selected file handle
653 load functions only on demand
657 run shell commands transparently within perl
661 load the C socket.h defines and structure manipulators
665 manipulate Perl symbols and their names
669 try every conceivable way to get hostname
673 interface to the UNIX syslog(3) calls
677 Perl termcap interface
681 word completion module
685 interface to various readline packages.
689 run perl standard test scripts with statistics
693 create an abbreviation table from a list
695 =item Text::ParseWords
697 parse text into an array of tokens
701 implementation of the Soundex Algorithm as Described by Knuth
705 expand and unexpand tabs per the unix expand(1) and unexpand(1)
709 line wrapping to form simple paragraphs
713 base class definitions for tied hashes
717 base class definitions for tied scalars
719 =item Tie::SubstrHash
721 fixed-table-size, fixed-key-length hashing
725 efficiently compute time from local and GMT time
729 base class for ALL classes (blessed references)
733 To find out I<all> the modules installed on your system, including
734 those without documentation or outside the standard release, do this:
736 find `perl -e 'print "@INC"'` -name '*.pm' -print
738 They should all have their own documentation installed and accessible via
739 your system man(1) command. If that fails, try the I<perldoc> program.
741 =head2 Extension Modules
743 Extension modules are written in C (or a mix of Perl and C) and get
744 dynamically loaded into Perl if and when you need them. Supported
745 extension modules include the Socket, Fcntl, and POSIX modules.
747 Many popular C extension modules do not come bundled (at least, not
748 completely) due to their size, volatility, or simply lack of time for
749 adequate testing and configuration across the multitude of platforms on
750 which Perl was beta-tested. You are encouraged to look for them in
751 archie(1L), the Perl FAQ or Meta-FAQ, the WWW page, and even with their
752 authors before randomly posting asking for their present condition and
757 CPAN stands for the Comprehensive Perl Archive Network. This is a globally
758 replicated collection of all known Perl materials, including hundreds
759 of unbundled modules. Here are the major categories of modules:
764 Language Extensions and Documentation Tools
770 Operating System Interfaces
773 Networking, Device Control (modems) and InterProcess Communication
776 Data Types and Data Type Utilities
785 Interfaces to / Emulations of Other Programming Languages
788 File Names, File Systems and File Locking (see also File Handles)
791 String Processing, Language Text Processing, Parsing and Searching
794 Option, Argument, Parameter and Configuration File Processing
797 Internationalization and Locale
800 Authentication, Security and Encryption
803 World Wide Web, HTML, HTTP, CGI, MIME
806 Server and Daemon Utilities
809 Archiving and Compression
812 Images, Pixmap and Bitmap Manipulation, Drawing and Graphing
818 Control Flow Utilities (callbacks and exceptions etc)
821 File Handle and Input/Output Stream Utilities
824 Miscellaneous Modules
828 The registered CPAN sites as of this writing include the following.
829 You should try to choose one close to you:
834 ftp://ftp.sterling.com/programming/languages/perl/
837 ftp://ftp.sedl.org/pub/mirrors/CPAN/
840 ftp://ftp.uoknor.edu/mirrors/CPAN/
843 ftp://ftp.delphi.com/pub/mirrors/packages/perl/CPAN/
846 ftp://uiarchive.cso.uiuc.edu/pub/lang/perl/CPAN/
849 ftp://ftp.cis.ufl.edu/pub/perl/CPAN/
852 ftp://ftp.switch.ch/mirror/CPAN/
855 ftp://ftp.sunet.se/pub/lang/perl/CPAN/
858 ftp://ftp.ci.uminho.pt/pub/lang/perl/
861 ftp://ftp.cs.ruu.nl/pub/PERL/CPAN/
864 ftp://ftp.demon.co.uk/pub/mirrors/perl/CPAN/
867 ftp://ftp.rz.ruhr-uni-bochum.de/pub/programming/languages/perl/CPAN/
870 ftp://ftp.leo.org/pub/comp/programming/languages/perl/CPAN/
873 ftp://ftp.pasteur.fr/pub/computing/unix/perl/CPAN/
876 ftp://ftp.ibp.fr/pub/perl/CPAN/
879 ftp://ftp.funet.fi/pub/languages/perl/CPAN/
882 ftp://ftp.tekotago.ac.nz/pub/perl/CPAN/
885 ftp://ftp.mame.mu.oz.au/pub/perl/CPAN/
888 ftp://coombs.anu.edu.au/pub/perl/
891 ftp://dongpo.math.ncu.edu.tw/perl/CPAN/
894 ftp://ftp.lab.kdd.co.jp/lang/perl/CPAN/
897 ftp://ftp.is.co.za/programming/perl/CPAN/
901 For an up-to-date listing of CPAN sites,
902 see F<http://www.perl.com/perl/CPAN> or F<ftp://ftp.perl.com/perl/>.
904 =head1 Modules: Creation, Use and Abuse
906 (The following section is borrowed directly from Tim Bunce's modules
907 file, available at your nearest CPAN site.)
909 Perl 5 implements a class using a package, but the presence of a
910 package doesn't imply the presence of a class. A package is just a
911 namespace. A class is a package that provides subroutines that can be
912 used as methods. A method is just a subroutine that expects, as its
913 first argument, either the name of a package (for "static" methods),
914 or a reference to something (for "virtual" methods).
916 A module is a file that (by convention) provides a class of the same
917 name (sans the .pm), plus an import method in that class that can be
918 called to fetch exported symbols. This module may implement some of
919 its methods by loading dynamic C or C++ objects, but that should be
920 totally transparent to the user of the module. Likewise, the module
921 might set up an AUTOLOAD function to slurp in subroutine definitions on
922 demand, but this is also transparent. Only the .pm file is required to
925 =head2 Guidelines for Module Creation
929 =item Do similar modules already exist in some form?
931 If so, please try to reuse the existing modules either in whole or
932 by inheriting useful features into a new class. If this is not
933 practical try to get together with the module authors to work on
934 extending or enhancing the functionality of the existing modules.
935 A perfect example is the plethora of packages in perl4 for dealing
936 with command line options.
938 If you are writing a module to expand an already existing set of
939 modules, please coordinate with the author of the package. It
940 helps if you follow the same naming scheme and module interaction
941 scheme as the original author.
943 =item Try to design the new module to be easy to extend and reuse.
945 Use blessed references. Use the two argument form of bless to bless
946 into the class name given as the first parameter of the constructor,
951 return bless {}, $class;
954 or even this if you'd like it to be used as either a static
959 my $class = ref($self) || $self;
960 return bless {}, $class;
963 Pass arrays as references so more parameters can be added later
964 (it's also faster). Convert functions into methods where
965 appropriate. Split large methods into smaller more flexible ones.
966 Inherit methods from other modules if appropriate.
968 Avoid class name tests like: C<die "Invalid" unless ref $ref eq 'FOO'>.
969 Generally you can delete the "C<eq 'FOO'>" part with no harm at all.
970 Let the objects look after themselves! Generally, avoid hardwired
971 class names as far as possible.
973 Avoid C<$r-E<gt>Class::func()> where using C<@ISA=qw(... Class ...)> and
974 C<$r-E<gt>func()> would work (see L<perlbot> for more details).
976 Use autosplit so little used or newly added functions won't be a
977 burden to programs which don't use them. Add test functions to
978 the module after __END__ either using AutoSplit or by saying:
980 eval join('',<main::DATA>) || die $@ unless caller();
982 Does your module pass the 'empty sub-class' test? If you say
983 "C<@SUBCLASS::ISA = qw(YOURCLASS);>" your applications should be able
984 to use SUBCLASS in exactly the same way as YOURCLASS. For example,
985 does your application still work if you change: C<$obj = new YOURCLASS;>
986 into: C<$obj = new SUBCLASS;> ?
988 Avoid keeping any state information in your packages. It makes it
989 difficult for multiple other packages to use yours. Keep state
990 information in objects.
992 Always use B<-w>. Try to C<use strict;> (or C<use strict qw(...);>).
993 Remember that you can add C<no strict qw(...);> to individual blocks
994 of code which need less strictness. Always use B<-w>. Always use B<-w>!
995 Follow the guidelines in the perlstyle(1) manual.
997 =item Some simple style guidelines
999 The perlstyle manual supplied with perl has many helpful points.
1001 Coding style is a matter of personal taste. Many people evolve their
1002 style over several years as they learn what helps them write and
1003 maintain good code. Here's one set of assorted suggestions that
1004 seem to be widely used by experienced developers:
1006 Use underscores to separate words. It is generally easier to read
1007 $var_names_like_this than $VarNamesLikeThis, especially for
1008 non-native speakers of English. It's also a simple rule that works
1009 consistently with VAR_NAMES_LIKE_THIS.
1011 Package/Module names are an exception to this rule. Perl informally
1012 reserves lowercase module names for 'pragma' modules like integer
1013 and strict. Other modules normally begin with a capital letter and
1014 use mixed case with no underscores (need to be short and portable).
1016 You may find it helpful to use letter case to indicate the scope
1017 or nature of a variable. For example:
1019 $ALL_CAPS_HERE constants only (beware clashes with perl vars)
1020 $Some_Caps_Here package-wide global/static
1021 $no_caps_here function scope my() or local() variables
1023 Function and method names seem to work best as all lowercase.
1024 E.g., C<$obj-E<gt>as_string()>.
1026 You can use a leading underscore to indicate that a variable or
1027 function should not be used outside the package that defined it.
1029 =item Select what to export.
1031 Do NOT export method names!
1033 Do NOT export anything else by default without a good reason!
1035 Exports pollute the namespace of the module user. If you must
1036 export try to use @EXPORT_OK in preference to @EXPORT and avoid
1037 short or common names to reduce the risk of name clashes.
1039 Generally anything not exported is still accessible from outside the
1040 module using the ModuleName::item_name (or C<$blessed_ref-E<gt>method>)
1041 syntax. By convention you can use a leading underscore on names to
1042 informally indicate that they are 'internal' and not for public use.
1044 (It is actually possible to get private functions by saying:
1045 C<my $subref = sub { ... }; &$subref;>. But there's no way to call that
1046 directly as a method, since a method must have a name in the symbol
1049 As a general rule, if the module is trying to be object oriented
1050 then export nothing. If it's just a collection of functions then
1051 @EXPORT_OK anything but use @EXPORT with caution.
1053 =item Select a name for the module.
1055 This name should be as descriptive, accurate and complete as
1056 possible. Avoid any risk of ambiguity. Always try to use two or
1057 more whole words. Generally the name should reflect what is special
1058 about what the module does rather than how it does it. Please use
1059 nested module names to informally group or categorise a module.
1060 A module should have a very good reason not to have a nested name.
1061 Module names should begin with a capital letter.
1063 Having 57 modules all called Sort will not make life easy for anyone
1064 (though having 23 called Sort::Quick is only marginally better :-).
1065 Imagine someone trying to install your module alongside many others.
1066 If in any doubt ask for suggestions in comp.lang.perl.misc.
1068 If you are developing a suite of related modules/classes it's good
1069 practice to use nested classes with a common prefix as this will
1070 avoid namespace clashes. For example: Xyz::Control, Xyz::View,
1071 Xyz::Model etc. Use the modules in this list as a naming guide.
1073 If adding a new module to a set, follow the original author's
1074 standards for naming modules and the interface to methods in
1077 To be portable each component of a module name should be limited to
1078 11 characters. If it might be used on DOS then try to ensure each is
1079 unique in the first 8 characters. Nested modules make this easier.
1081 =item Have you got it right?
1083 How do you know that you've made the right decisions? Have you
1084 picked an interface design that will cause problems later? Have
1085 you picked the most appropriate name? Do you have any questions?
1087 The best way to know for sure, and pick up many helpful suggestions,
1088 is to ask someone who knows. Comp.lang.perl.misc is read by just about
1089 all the people who develop modules and it's the best place to ask.
1091 All you need to do is post a short summary of the module, its
1092 purpose and interfaces. A few lines on each of the main methods is
1093 probably enough. (If you post the whole module it might be ignored
1094 by busy people - generally the very people you want to read it!)
1096 Don't worry about posting if you can't say when the module will be
1097 ready - just say so in the message. It might be worth inviting
1098 others to help you, they may be able to complete it for you!
1100 =item README and other Additional Files.
1102 It's well known that software developers usually fully document the
1103 software they write. If, however, the world is in urgent need of
1104 your software and there is not enough time to write the full
1105 documentation please at least provide a README file containing:
1110 A description of the module/package/extension etc.
1113 A copyright notice - see below.
1116 Prerequisites - what else you may need to have.
1119 How to build it - possible changes to Makefile.PL etc.
1125 Recent changes in this release, especially incompatibilities
1128 Changes / enhancements you plan to make in the future.
1132 If the README file seems to be getting too large you may wish to
1133 split out some of the sections into separate files: INSTALL,
1138 =item Adding a Copyright Notice.
1140 How you choose to licence your work is a personal decision.
1141 The general mechanism is to assert your Copyright and then make
1142 a declaration of how others may copy/use/modify your work.
1144 Perl, for example, is supplied with two types of license: The GNU
1145 GPL and The Artistic License (see the files README, Copying and
1146 Artistic). Larry has good reasons for NOT just using the GNU GPL.
1148 My personal recommendation, out of respect for Larry, Perl and the
1149 perl community at large is to simply state something like:
1151 Copyright (c) 1995 Your Name. All rights reserved.
1152 This program is free software; you can redistribute it and/or
1153 modify it under the same terms as Perl itself.
1155 This statement should at least appear in the README file. You may
1156 also wish to include it in a Copying file and your source files.
1157 Remember to include the other words in addition to the Copyright.
1159 =item Give the module a version/issue/release number.
1161 To be fully compatible with the Exporter and MakeMaker modules you
1162 should store your module's version number in a non-my package
1163 variable called $VERSION. This should be a valid floating point
1164 number with at least two digits after the decimal (ie hundredths,
1165 e.g, C<$VERSION = "0.01">). Don't use a "1.3.2" style version.
1166 See Exporter.pm in Perl5.001m or later for details.
1168 It may be handy to add a function or method to retrieve the number.
1169 Use the number in announcements and archive file names when
1170 releasing the module (ModuleName-1.02.tar.Z).
1171 See perldoc ExtUtils::MakeMaker.pm for details.
1173 =item How to release and distribute a module.
1175 It's good idea to post an announcement of the availability of your
1176 module (or the module itself if small) to the comp.lang.perl.announce
1177 Usenet newsgroup. This will at least ensure very wide once-off
1180 If possible you should place the module into a major ftp archive and
1181 include details of it's location in your announcement.
1183 Some notes about ftp archives: Please use a long descriptive file
1184 name which includes the version number. Most incoming directories
1185 will not be readable/listable, i.e., you won't be able to see your
1186 file after uploading it. Remember to send your email notification
1187 message as soon as possible after uploading else your file may get
1188 deleted automatically. Allow time for the file to be processed
1189 and/or check the file has been processed before announcing its
1192 FTP Archives for Perl Modules:
1194 Follow the instructions and links on
1196 http://franz.ww.tu-berlin.de/modulelist
1198 or upload to one of these sites:
1200 ftp://franz.ww.tu-berlin.de/incoming
1201 ftp://ftp.cis.ufl.edu/incoming
1203 and notify upload@franz.ww.tu-berlin.de.
1205 By using the WWW interface you can ask the Upload Server to mirror
1206 your modules from your ftp or WWW site into your own directory on
1209 Please remember to send me an updated entry for the Module list!
1211 =item Take care when changing a released module.
1213 Always strive to remain compatible with previous released versions
1214 (see 2.2 above) Otherwise try to add a mechanism to revert to the
1215 old behaviour if people rely on it. Document incompatible changes.
1221 =head2 Guidelines for Converting Perl 4 Library Scripts into Modules
1225 =item There is no requirement to convert anything.
1227 If it ain't broke, don't fix it! Perl 4 library scripts should
1228 continue to work with no problems. You may need to make some minor
1229 changes (like escaping non-array @'s in double quoted strings) but
1230 there is no need to convert a .pl file into a Module for just that.
1232 =item Consider the implications.
1234 All the perl applications which make use of the script will need to
1235 be changed (slightly) if the script is converted into a module. Is
1236 it worth it unless you plan to make other changes at the same time?
1238 =item Make the most of the opportunity.
1240 If you are going to convert the script to a module you can use the
1241 opportunity to redesign the interface. The 'Guidelines for Module
1242 Creation' above include many of the issues you should consider.
1244 =item The pl2pm utility will get you started.
1246 This utility will read *.pl files (given as parameters) and write
1247 corresponding *.pm files. The pl2pm utilities does the following:
1252 Adds the standard Module prologue lines
1255 Converts package specifiers from ' to ::
1258 Converts die(...) to croak(...)
1261 Several other minor changes
1265 Being a mechanical process pl2pm is not bullet proof. The converted
1266 code will need careful checking, especially any package statements.
1267 Don't delete the original .pl file till the new .pm one works!
1271 =head2 Guidelines for Reusing Application Code
1275 =item Complete applications rarely belong in the Perl Module Library.
1277 =item Many applications contain some perl code which could be reused.
1279 Help save the world! Share your code in a form that makes it easy
1282 =item Break-out the reusable code into one or more separate module files.
1284 =item Take the opportunity to reconsider and redesign the interfaces.
1286 =item In some cases the 'application' can then be reduced to a small
1288 fragment of code built on top of the reusable modules. In these cases
1289 the application could invoked as:
1291 perl -e 'use Module::Name; method(@ARGV)' ...
1293 perl -mModule::Name ... (in perl5.002)