2 # Time-stamp: "2001-05-25 07:50:08 MDT"
6 Locale::Maketext -- framework for localization
13 # ...which inherits from Locale::Maketext
14 my $lh = MyProgram::L10N->get_handle() || die "What language?";
16 # And then any messages your program emits, like:
17 warn $lh->maketext( "Can't open file [_1]: [_2]\n", $f, $! );
22 It is a common feature of applications (whether run directly,
23 or via the Web) for them to be "localized" -- i.e., for them
24 to a present an English interface to an English-speaker, a German
25 interface to a German-speaker, and so on for all languages it's
26 programmed with. Locale::Maketext
27 is a framework for software localization; it provides you with the
28 tools for organizing and accessing the bits of text and text-processing
29 code that you need for producing localized applications.
31 In order to make sense of Maketext and how all its
32 components fit together, you should probably
33 go read L<Locale::Maketext::TPJ13|Locale::Maketext::TPJ13>, and
34 I<then> read the following documentation.
36 You may also want to read over the source for C<File::Findgrep>
37 and its constituent modules -- they are a complete (if small)
38 example application that uses Maketext.
42 The basic design of Locale::Maketext is object-oriented, and
43 Locale::Maketext is an abstract base class, from which you
44 derive a "project class".
45 The project class (with a name like "TkBocciBall::Localize",
46 which you then use in your module) is in turn the base class
47 for all the "language classes" for your project
48 (with names "TkBocciBall::Localize::it",
49 "TkBocciBall::Localize::en",
50 "TkBocciBall::Localize::fr", etc.).
53 a class containing a lexicon of phrases as class data,
54 and possibly also some methods that are of use in interpreting
55 phrases in the lexicon, or otherwise dealing with text in that
58 An object belonging to a language class is called a "language
59 handle"; it's typically a flyweight object.
61 The normal course of action is to call:
63 use TkBocciBall::Localize; # the localization project class
64 $lh = TkBocciBall::Localize->get_handle();
65 # Depending on the user's locale, etc., this will
66 # make a language handle from among the classes available,
67 # and any defaults that you declare.
68 die "Couldn't make a language handle??" unless $lh;
70 From then on, you use the C<maketext> function to access
71 entries in whatever lexicon(s) belong to the language handle
74 print $lh->maketext("You won!"), "\n";
76 ...emits the right text for this language. If the object
77 in C<$lh> belongs to class "TkBocciBall::Localize::fr" and
78 %TkBocciBall::Localize::fr::Lexicon contains C<("You won!"
79 =E<gt> "Tu as gagnE<eacute>!")>, then the above
80 code happily tells the user "Tu as gagnE<eacute>!".
84 Locale::Maketext offers a variety of methods, which fall
85 into three categories:
91 Methods to do with constructing language handles.
95 C<maketext> and other methods to do with accessing %Lexicon data
96 for a given language handle.
100 Methods that you may find it handy to use, from routines of
101 yours that you put in %Lexicon entries.
105 These are covered in the following section.
107 =head2 Construction Methods
109 These are to do with constructing a language handle:
115 $lh = YourProjClass->get_handle( ...langtags... ) || die "lg-handle?";
117 This tries loading classes based on the language-tags you give (like
118 C<("en-US", "sk", "kon", "es-MX", "ja", "i-klingon")>, and for the first class
119 that succeeds, returns YourProjClass::I<language>->new().
121 It runs thru the entire given list of language-tags, and finds no classes
122 for those exact terms, it then tries "superordinate" language classes.
123 So if no "en-US" class (i.e., YourProjClass::en_us)
124 was found, nor classes for anything else in that list, we then try
125 its superordinate, "en" (i.e., YourProjClass::en), and so on thru
126 the other language-tags in the given list: "es".
127 (The other language-tags in our example list:
128 happen to have no superordinates.)
130 If none of those language-tags leads to loadable classes, we then
131 try classes derived from YourProjClass->fallback_languages() and
132 then if nothing comes of that, we use classes named by
133 YourProjClass->fallback_language_classes(). Then in the (probably
134 quite unlikely) event that that fails, we just return undef.
138 $lh = YourProjClass->get_handleB<()> || die "lg-handle?";
140 When C<get_handle> is called with an empty parameter list, magic happens:
142 If C<get_handle> senses that it's running in program that was
143 invoked as a CGI, then it tries to get language-tags out of the
144 environment variable "HTTP_ACCEPT_LANGUAGE", and it pretends that
145 those were the languages passed as parameters to C<get_handle>.
147 Otherwise (i.e., if not a CGI), this tries various OS-specific ways
148 to get the language-tags for the current locale/language, and then
149 pretends that those were the value(s) passed to C<cet_handle>.
151 Currently this OS-specific stuff consists of looking in the environment
152 variables "LANG" and "LANGUAGE"; and on MSWin machines (where those
153 variables are typically unused), this also tries using
154 the module Win32::Locale to get a language-tag for whatever language/locale
155 is currently selected in the "Regional Settings" (or "International"?)
156 Control Panel. I welcome further
157 suggestions for making this do the Right Thing under other operating
158 systems that support localization.
160 If you're using localization in an application that keeps a configuration
161 file, you might consider something like this in your project class:
163 sub get_handle_via_config {
165 my $preferred_language = $Config_settings{'language'};
167 if($preferred_language) {
168 $lh = $class->get_handle($chosen_language)
169 || die "No language handle for \"$chosen_language\" or the like";
171 # Config file missing, maybe?
172 $lh = $class->get_handle()
173 || die "Can't get a language handle";
178 =item $lh = YourProjClass::langname->new();
180 This constructs a language handle. You usually B<don't> call this
181 directly, but instead let C<get_handle> find a language class to C<use>
182 and to then call ->new on.
186 This is called by ->new to initialize newly-constructed language handles.
187 If you define an init method in your class, remember that it's usually
188 considered a good idea to call $lh->SUPER::init in it (presumably at the
189 beginning), so that all classes get a chance to initialize a new object
190 however they see fit.
192 =item YourProjClass->fallback_languages()
194 C<get_handle> appends the return value of this to the end of
195 whatever list of languages you pass C<get_handle>. Unless
196 you override this method, your project class
197 will inherit Locale::Maketext's C<fallback_languages>, which
198 currently returns C<('i-default', 'en', 'en-US')>.
199 ("i-default" is defined in RFC 2277).
201 This method (by having it return the name
202 of a language-tag that has an existing language class)
203 can be used for making sure that
204 C<get_handle> will always manage to construct a language
205 handle (assuming your language classes are in an appropriate
206 @INC directory). Or you can use the next method:
208 =item YourProjClass->fallback_language_classes()
210 C<get_handle> appends the return value of this to the end
211 of the list of classes it will try using. Unless
212 you override this method, your project class
213 will inherit Locale::Maketext's C<fallback_language_classes>,
214 which currently returns an empty list, C<()>.
215 By setting this to some value (namely, the name of a loadable
216 language class), you can be sure that
217 C<get_handle> will always manage to construct a language
222 =head2 The "maketext" Method
224 This is the most important method in Locale::Maketext:
226 $text = $lh->maketext(I<key>, ...parameters for this phrase...);
228 This looks in the %Lexicon of the language handle
229 $lh and all its superclasses, looking
230 for an entry whose key is the string I<key>. Assuming such
231 an entry is found, various things then happen, depending on the
234 If the value is a scalarref, the scalar is dereferenced and returned
235 (and any parameters are ignored).
236 If the value is a coderef, we return &$value($lh, ...parameters...).
237 If the value is a string that I<doesn't> look like it's in Bracket Notation,
238 we return it (after replacing it with a scalarref, in its %Lexicon).
239 If the value I<does> look like it's in Bracket Notation, then we compile
240 it into a sub, replace the string in the %Lexicon with the new coderef,
241 and then we return &$new_sub($lh, ...parameters...).
243 Bracket Notation is discussed in a later section. Note
244 that trying to compile a string into Bracket Notation can throw
245 an exception if the string is not syntactically valid (say, by not
246 balancing brackets right.)
248 Also, calling &$coderef($lh, ...parameters...) can throw any sort of
249 exception (if, say, code in that sub tries to divide by zero). But
250 a very common exception occurs when you have Bracket
251 Notation text that says to call a method "foo", but there is no such
252 method. (E.g., "You have [quaB<tn>,_1,ball]." will throw an exception
253 on trying to call $lh->quaB<tn>($_[1],'ball') -- you presumably meant
254 "quant".) C<maketext> catches these exceptions, but only to make the
255 error message more readable, at which point it rethrows the exception.
257 An exception I<may> be thrown if I<key> is not found in any
258 of $lh's %Lexicon hashes. What happens if a key is not found,
259 is discussed in a later section, "Controlling Lookup Failure".
261 Note that you might find it useful in some cases to override
262 the C<maketext> method with an "after method", if you want to
263 translate encodings, or even scripts:
265 package YrProj::zh_cn; # Chinese with PRC-style glyphs
266 use base ('YrProj::zh_tw'); # Taiwan-style
268 my $self = shift(@_);
269 my $value = $self->maketext(@_);
270 return Chineeze::taiwan2mainland($value);
273 Or you may want to override it with something that traps
274 any exceptions, if that's critical to your program:
277 my($lh, @stuff) = @_;
279 eval { $out = $lh->SUPER::maketext(@stuff) };
280 return $out unless $@;
281 ...otherwise deal with the exception...
284 Other than those two situations, I don't imagine that
285 it's useful to override the C<maketext> method. (If
286 you run into a situation where it is useful, I'd be
287 interested in hearing about it.)
291 =item $lh->fail_with I<or> $lh->fail_with(I<PARAM>)
293 =item $lh->failure_handler_auto
295 These two methods are discussed in the section "Controlling
300 =head2 Utility Methods
302 These are methods that you may find it handy to use, generally
303 from %Lexicon routines of yours (whether expressed as
304 Bracket Notation or not).
308 =item $language->quant($number, $singular)
310 =item $language->quant($number, $singular, $plural)
312 =item $language->quant($number, $singular, $plural, $negative)
314 This is generally meant to be called from inside Bracket Notation
315 (which is discussed later), as in
317 "Your search matched [quant,_1,document]!"
319 It's for I<quantifying> a noun (i.e., saying how much of it there is,
320 while giving the currect form of it). The behavior of this method is
321 handy for English and a few other Western European languages, and you
322 should override it for languages where it's not suitable. You can feel
323 free to read the source, but the current implementation is basically
324 as this pseudocode describes:
326 if $number is 0 and there's a $negative,
329 return "1 $singular";
330 elsif there's a $plural,
331 return "$number $plural";
333 return "$number " . $singular . "s";
335 # ...except that we actually call numf to
336 # stringify $number before returning it.
338 So for English (with Bracket Notation)
339 C<"...[quant,_1,file]..."> is fine (for 0 it returns "0 files",
340 for 1 it returns "1 file", and for more it returns "2 files", etc.)
342 But for "directory", you'd want C<"[quant,_1,direcory,directories]">
343 so that our elementary C<quant> method doesn't think that the
344 plural of "directory" is "directorys". And you might find that the
345 output may sound better if you specify a negative form, as in:
347 "[quant,_1,file,files,No files] matched your query.\n"
349 Remember to keep in mind verb agreement (or adjectives too, in
350 other languages), as in:
352 "[quant,_1,document] were matched.\n"
354 Because if _1 is one, you get "1 document B<were> matched".
355 An acceptable hack here is to do something like this:
357 "[quant,_1,document was, documents were] matched.\n"
359 =item $language->numf($number)
361 This returns the given number formatted nicely according to
362 this language's conventions. Maketext's default method is
363 mostly to just take the normal string form of the number
364 (applying sprintf "%G" for only very large numbers), and then
365 to add commas as necessary. (Except that
366 we apply C<tr/,./.,/> if $language->{'numf_comma'} is true;
367 that's a bit of a hack that's useful for languages that express
368 two million as "2.000.000" and not as "2,000,000").
370 If you want anything fancier, consider overriding this with something
371 that uses L<Number::Format|Number::Format>, or does something else
374 Note that numf is called by quant for stringifying all quantifying
377 =item $language->sprintf($format, @items)
379 This is just a wrapper around Perl's normal C<sprintf> function.
380 It's provided so that you can use "sprintf" in Bracket Notation:
382 "Couldn't access datanode [sprintf,%10x=~[%s~],_1,_2]!\n"
386 Couldn't access datanode Stuff=[thangamabob]!
388 =item $language->language_tag()
390 Currently this just takes the last bit of C<ref($language)>, turns
391 underscores to dashes, and returns it. So if $language is
392 an object of class Hee::HOO::Haw::en_us, $language->language_tag()
393 returns "en-us". (Yes, the usual representation for that language
394 tag is "en-US", but case is I<never> considered meaningful in
395 language-tag comparison.)
397 You may override this as you like; Maketext doesn't use it for
400 =item $language->encoding()
402 Currently this isn't used for anything, but it's provided
403 (with default value of
404 C<(ref($language) && $language-E<gt>{'encoding'})) or "iso-8859-1">
405 ) as a sort of suggestion that it may be useful/necessary to
406 associate encodings with your language handles (whether on a
407 per-class or even per-handle basis.)
411 =head2 Language Handle Attributes and Internals
413 A language handle is a flyweight object -- i.e., it doesn't (necessarily)
414 carry any data of interest, other than just being a member of
415 whatever class it belongs to.
417 A language handle is implemented as a blessed hash. Subclasses of yours
418 can store whatever data you want in the hash. Currently the only hash
419 entry used by any crucial Maketext method is "fail", so feel free to
420 use anything else as you like.
422 B<Remember: Don't be afraid to read the Maketext source if there's
423 any point on which this documentation is unclear.> This documentation
424 is vastly longer than the module source itself.
430 =head1 LANGUAGE CLASS HIERARCHIES
432 These are Locale::Maketext's assumptions about the class
433 hierarchy formed by all your language classes:
439 You must have a project base class, which you load, and
440 which you then use as the first argument in
441 the call to YourProjClass->get_handle(...). It should derive
442 (whether directly or indirectly) from Locale::Maketext.
443 It B<doesn't matter> how you name this class, altho assuming this
444 is the localization component of your Super Mega Program,
445 good names for your project class might be
446 SuperMegaProgram::Localization, SuperMegaProgram::L10N,
447 SuperMegaProgram::I18N, SuperMegaProgram::International,
448 or even SuperMegaProgram::Languages or SuperMegaProgram::Messages.
452 Language classes are what YourProjClass->get_handle will try to load.
453 It will look for them by taking each language-tag (B<skipping> it
454 if it doesn't look like a language-tag or locale-tag!), turning it to
455 all lowercase, turning and dashes to underscores, and appending it
456 to YourProjClass . "::". So this:
458 $lh = YourProjClass->get_handle(
459 'en-US', 'fr', 'kon', 'i-klingon', 'i-klingon-romanized'
462 will try loading the classes
463 YourProjClass::en_us (note lowercase!), YourProjClass::fr,
465 YourProjClass::i_klingon
466 and YourProjClass::i_klingon_romanized. (And it'll stop at the
467 first one that actually loads.)
471 I assume that each language class derives (directly or indirectly)
472 from your project class, and also defines its @ISA, its %Lexicon,
473 or both. But I anticipate no dire consequences if these assumptions
478 Language classes may derive from other language classes (altho they
479 should have "use I<Thatclassname>" or "use base qw(I<...classes...>)").
480 They may derive from the project
481 class. They may derive from some other class altogether. Or via
482 multiple inheritance, it may derive from any mixture of these.
486 I foresee no problems with having multiple inheritance in
487 your hierarchy of language classes. (As usual, however, Perl will
488 complain bitterly if you have a cycle in the hierarchy: i.e., if
489 any class is its own ancestor.)
493 =head1 ENTRIES IN EACH LEXICON
495 A typical %Lexicon entry is meant to signify a phrase,
496 taking some number (0 or more) of parameters. An entry
497 is meant to be accessed by via
498 a string I<key> in $lh->maketext(I<key>, ...parameters...),
499 which should return a string that is generally meant for
500 be used for "output" to the user -- regardless of whether
501 this actually means printing to STDOUT, writing to a file,
502 or putting into a GUI widget.
504 While the key must be a string value (since that's a basic
505 restriction that Perl places on hash keys), the value in
506 the lexicon can currenly be of several types:
507 a defined scalar, scalarref, or coderef. The use of these is
508 explained above, in the section 'The "maketext" Method', and
509 Bracket Notation for strings is discussed in the next section.
511 While you can use arbitrary unique IDs for lexicon keys
512 (like "_min_larger_max_error"), it is often
513 useful for if an entry's key is itself a valid value, like
514 this example error message:
516 "Minimum ([_1]) is larger than maximum ([_2])!\n",
518 Compare this code that uses an arbitrary ID...
520 die $lh->maketext( "_min_larger_max_error", $min, $max )
523 ...to this code that uses a key-as-value:
526 "Minimum ([_1]) is larger than maximum ([_2])!\n",
530 The second is, in short, more readable. In particular, it's obvious
531 that the number of parameters you're feeding to that phrase (two) is
532 the number of parameters that it I<wants> to be fed. (Since you see
533 _1 and a _2 being used in the key there.)
535 Also, once a project is otherwise
536 complete and you start to localize it, you can scrape together
537 all the various keys you use, and pass it to a translator; and then
538 the translator's work will go faster if what he's presented is this:
540 "Minimum ([_1]) is larger than maximum ([_2])!\n",
541 => "", # fill in something here, Jacques!
543 rather than this more cryptic mess:
545 "_min_larger_max_error"
546 => "", # fill in something here, Jacques
548 I think that keys as lexicon values makes the completed lexicon
549 entries more readable:
551 "Minimum ([_1]) is larger than maximum ([_2])!\n",
552 => "Le minimum ([_1]) est plus grand que le maximum ([_2])!\n",
554 Also, having valid values as keys becomes very useful if you set
555 up an _AUTO lexicon. _AUTO lexicons are discussed in a later
558 I almost always use keys that are themselves
559 valid lexicon values. One notable exception is when the value is
560 quite long. For example, to get the screenful of data that
561 a command-line program might returns when given an unknown switch,
562 I often just use a key "_USAGE_MESSAGE". At that point I then go
563 and immediately to define that lexicon entry in the
564 ProjectClass::L10N::en lexicon (since English is always my "project
567 '_USAGE_MESSAGE' => <<'EOSTUFF',
568 ...long long message...
571 and then I can use it as:
573 getopt('oDI', \%opts) or die $lh->maketext('_USAGE_MESSAGE');
576 note that each class's C<%Lexicon> inherits-and-extends
577 the lexicons in its superclasses. This is not because these are
578 special hashes I<per se>, but because you access them via the
579 C<maketext> method, which looks for entries across all the
580 C<%Lexicon>'s in a language class I<and> all its ancestor classes.
581 (This is because the idea of "class data" isn't directly implemented
582 in Perl, but is instead left to individual class-systems to implement
585 Note that you may have things stored in a lexicon
586 besides just phrases for output: for example, if your program
587 takes input from the keyboard, asking a "(Y/N)" question,
588 you probably need to know what equivalent of "Y[es]/N[o]" is
589 in whatever language. You probably also need to know what
590 the equivalents of the answers "y" and "n" are. You can
591 store that information in the lexicon (say, under the keys
592 "~answer_y" and "~answer_n", and the long forms as
593 "~answer_yes" and "~answer_no", where "~" is just an ad-hoc
594 character meant to indicate to programmers/translators that
595 these are not phrases for output).
597 Or instead of storing this in the language class's lexicon,
598 you can (and, in some cases, really should) represent the same bit
599 of knowledge as code is a method in the language class. (That
600 leaves a tidy distinction between the lexicon as the things we
601 know how to I<say>, and the rest of the things in the lexicon class
602 as things that we know how to I<do>.) Consider
603 this example of a processor for responses to French "oui/non"
607 return undef unless defined $_[1] and length $_[1];
608 my $answer = lc $_[1]; # smash case
609 return 1 if $answer eq 'o' or $answer eq 'oui';
610 return 0 if $answer eq 'n' or $answer eq 'non';
614 ...which you'd then call in a construct like this:
617 until(defined $response) {
618 print $lh->maketext("Open the pod bay door (y/n)? ");
619 $response = $lh->y_or_n( get_input_from_keyboard_somehow() );
621 if($response) { $pod_bay_door->open() }
622 else { $pod_bay_door->leave_closed() }
624 Other data worth storing in a lexicon might be things like
625 filenames for language-targetted resources:
629 => "/styles/en_us/main_splash.png",
630 "_main_splash_imagemap"
631 => "/styles/en_us/main_splash.incl",
632 "_general_graphics_path"
635 => "/styles/en_us/hey_there.wav",
639 => "right_arrow.png",
640 # In some other languages, left equals
641 # BACKwards, and right is FOREwards.
644 You might want to do the same thing for expressing key bindings
645 or the like (since hardwiring "q" as the binding for the function
646 that quits a screen/menu/program is useful only if your language
647 happens to associate "q" with "quit"!)
649 =head1 BRACKET NOTATION
651 Bracket Notation is a crucial feature of Locale::Maketext. I mean
652 Bracket Notation to provide a replacement for sprintf formatting.
653 Everything you do with Bracket Notation could be done with a sub block,
654 but bracket notation is meant to be much more concise.
656 Bracket Notation is a like a miniature "template" system (in the sense
657 of L<Text::Template|Text::Template>, not in the sense of C++ templates),
658 where normal text is passed thru basically as is, but text is special
659 regions is specially interpreted. In Bracket Notation, you use brackets
660 ("[...]" -- not "{...}"!) to note sections that are specially interpreted.
662 For example, here all the areas that are taken literally are underlined with
663 a "^", and all the in-bracket special regions are underlined with an X:
665 "Minimum ([_1]) is larger than maximum ([_2])!\n",
666 ^^^^^^^^^ XX ^^^^^^^^^^^^^^^^^^^^^^^^^^ XX ^^^^
668 When that string is compiled from bracket notation into a real Perl sub,
669 it's basically turned into:
677 ") is larger than maximum (",
681 # to be called by $lh->maketext(KEY, params...)
683 In other words, text outside bracket groups is turned into string
684 literals. Text in brackets is rather more complex, and currently follows
691 Bracket groups that are empty, or which consist only of whitespace,
692 are ignored. (Examples: "[]", "[ ]", or a [ and a ] with returns
693 and/or tabs and/or spaces between them.
695 Otherwise, each group is taken to be a comma-separated group of items,
696 and each item is interpreted as follows:
700 An item that is "_I<digits>" or "_-I<digits>" is interpreted as
701 $_[I<value>]. I.e., "_1" is becomes with $_[1], and "_-3" is interpreted
702 as $_[-3] (in which case @_ should have at least three elements in it).
703 Note that $_[0] is the language handle, and is typically not named
708 An item "_*" is interpreted to mean "all of @_ except $_[0]".
709 I.e., C<@_[1..$#_]>. Note that this is an empty list in the case
710 of calls like $lh->maketext(I<key>) where there are no
711 parameters (except $_[0], the language handle).
715 Otherwise, each item is interpreted as a string literal.
719 The group as a whole is interpreted as follows:
725 If the first item in a bracket group looks like a method name,
726 then that group is interpreted like this:
728 $lh->that_method_name(
729 ...rest of items in this group...
734 If the first item in a bracket group is empty-string, or "_*"
735 or "_I<digits>" or "_-I<digits>", then that group is interpreted
736 as just the interpolation of all its items:
739 ...rest of items in this group...
742 Examples: "[_1]" and "[,_1]", which are synonymous; and
743 "[,ID-(,_4,-,_2,)]", which compiles as
744 C<join "", "ID-(", $_[4], "-", $_[2], ")">.
748 Otherwise this bracket group is invalid. For example, in the group
749 "[!@#,whatever]", the first item C<"!@#"> is neither empty-string,
750 "_I<number>", "_-I<number>", "_*", nor a valid method name; and so
751 Locale::Maketext will throw an exception of you try compiling an
752 expression containing this bracket group.
756 Note, incidentally, that items in each group are comma-separated,
757 not C</\s*,\s*/>-separated. That is, you might expect that this
760 "Hoohah [foo, _1 , bar ,baz]!"
762 would compile to this:
768 $lh->foo( $_[1], "bar", "baz"),
772 But it actually compiles as this:
778 $lh->foo(" _1 ", " bar ", "baz"), #!!!
782 In the notation discussed so far, the characters "[" and "]" are given
783 special meaning, for opening and closing bracket groups, and "," has
784 a special meaning inside bracket groups, where it separates items in the
785 group. This begs the question of how you'd express a literal "[" or
786 "]" in a Bracket Notation string, and how you'd express a literal
787 comma inside a bracket group. For this purpose I've adopted "~" (tilde)
788 as an escape character: "~[" means a literal '[' character anywhere
789 in Bracket Notation (i.e., regardless of whether you're in a bracket
790 group or not), and ditto for "~]" meaning a literal ']', and "~," meaning
791 a literal comma. (Altho "," means a literal comma outside of
792 bracket groups -- it's only inside bracket groups that commas are special.)
794 And on the off chance you need a literal tilde in a bracket expression,
795 you get it with "~~".
797 Currently, an unescaped "~" before a character
798 other than a bracket or a comma is taken to mean just a "~" and that
799 charecter. I.e., "~X" means the same as "~~X" -- i.e., one literal tilde,
800 and then one literal "X". However, by using "~X", you are assuming that
801 no future version of Maketext will use "~X" as a magic escape sequence.
802 In practice this is not a great problem, since first off you can just
803 write "~~X" and not worry about it; second off, I doubt I'll add lots
804 of new magic characters to bracket notation; and third off, you
805 aren't likely to want literal "~" characters in your messages anyway,
806 since it's not a character with wide use in natural language text.
808 Brackets must be balanced -- every openbracket must have
809 one matching closebracket, and vice versa. So these are all B<invalid>:
811 "I ate [quant,_1,rhubarb pie."
812 "I ate [quant,_1,rhubarb pie[."
813 "I ate quant,_1,rhubarb pie]."
814 "I ate quant,_1,rhubarb pie[."
816 Currently, bracket groups do not nest. That is, you B<cannot> say:
818 "Foo [bar,baz,[quux,quuux]]\n";
820 If you need a notation that's that powerful, use normal Perl:
828 $lh->bar('baz', $lh->quux('quuux')),
834 Or write the "bar" method so you don't need to pass it the
835 output from calling quux.
837 I do not anticipate that you will need (or particularly want)
838 to nest bracket groups, but you are welcome to email me with
839 convincing (real-life) arguments to the contrary.
843 If maketext goes to look in an individual %Lexicon for an entry
844 for I<key> (where I<key> does not start with an underscore), and
845 sees none, B<but does see> an entry of "_AUTO" => I<some_true_value>,
846 then we actually define $Lexicon{I<key>} = I<key> right then and there,
847 and then use that value as if it had been there all
848 along. This happens before we even look in any superclass %Lexicons!
850 (This is meant to be somewhat like the AUTOLOAD mechanism in
851 Perl's function call system -- or, looked at another way,
852 like the L<AutoLoader|AutoLoader> module.)
854 I can picture all sorts of circumstances where you just
855 do not want lookup to be able to fail (since failing
856 normally means that maketext throws a C<die>, altho
857 see the next section for greater control over that). But
858 here's one circumstance where _AUTO lexicons are meant to
859 be I<especially> useful:
861 As you're writing an application, you decide as you go what messages
862 you need to emit. Normally you'd go to write this:
865 go_process_file($filename)
867 print "Couldn't find file \"$filename\"!\n";
870 but since you anticipate localizing this, you write:
872 use ThisProject::I18N;
873 my $lh = ThisProject::I18N->get_handle();
874 # For the moment, assume that things are set up so
875 # that we load class ThisProject::I18N::en
876 # and that that's the class that $lh belongs to.
879 go_process_file($filename)
882 "Couldn't find file \"[_1]\"!\n", $filename
886 Now, right after you've just written the above lines, you'd
887 normally have to go open the file
888 ThisProject/I18N/en.pm, and immediately add an entry:
890 "Couldn't find file \"[_1]\"!\n"
891 => "Couldn't find file \"[_1]\"!\n",
893 But I consider that somewhat of a distraction from the work
894 of getting the main code working -- to say nothing of the fact
895 that I often have to play with the program a few times before
896 I can decide exactly what wording I want in the messages (which
897 in this case would require me to go changing three lines of code:
898 the call to maketext with that key, and then the two lines in
899 ThisProject/I18N/en.pm).
901 However, if you set "_AUTO => 1" in the %Lexicon in,
902 ThisProject/I18N/en.pm (assuming that English (en) is
903 the language that all your programmers will be using for this
904 project's internal message keys), then you don't ever have to
905 go adding lines like this
907 "Couldn't find file \"[_1]\"!\n"
908 => "Couldn't find file \"[_1]\"!\n",
910 to ThisProject/I18N/en.pm, because if _AUTO is true there,
911 then just looking for an entry with the key "Couldn't find
912 file \"[_1]\"!\n" in that lexicon will cause it to be added,
915 Note that the reason that keys that start with "_"
916 are immune to _AUTO isn't anything generally magical about
917 the underscore character -- I just wanted a way to have most
918 lexicon keys be autoable, except for possibly a few, and I
919 arbitrarily decided to use a leading underscore as a signal
920 to distinguish those few.
922 =head1 CONTROLLING LOOKUP FAILURE
924 If you call $lh->maketext(I<key>, ...parameters...),
925 and there's no entry I<key> in $lh's class's %Lexicon, nor
926 in the superclass %Lexicon hash, I<and> if we can't auto-make
927 I<key> (because either it starts with a "_", or because none
928 of its lexicons have C<_AUTO =E<gt> 1,>), then we have
929 failed to find a normal way to maketext I<key>. What then
930 happens in these failure conditions, depends on the $lh object
933 If the language handle has no "fail" attribute, maketext
934 will simply throw an exception (i.e., it calls C<die>, mentioning
935 the I<key> whose lookup failed, and naming the line number where
936 the calling $lh->maketext(I<key>,...) was.
938 If the language handle has a "fail" attribute whose value is a
939 coderef, then $lh->maketext(I<key>,...params...) gives up and calls:
941 return &{$that_subref}($lh, $key, @params);
943 Otherwise, the "fail" attribute's value should be a string denoting
944 a method name, so that $lh->maketext(I<key>,...params...) can
947 return $lh->$that_method_name($phrase, @params);
949 The "fail" attribute can be accessed with the C<fail_with> method:
952 $lh->fail_with( \&failure_handler );
954 # Set to a method name:
955 $lh->fail_with( 'failure_method' );
957 # Set to nothing (i.e., so failure throws a plain exception)
958 $lh->fail_with( undef );
961 $handler = $lh->fail_with();
963 Now, as to what you may want to do with these handlers: Maybe you'd
964 want to log what key failed for what class, and then die. Maybe
965 you don't like C<die> and instead you want to send the error message
966 to STDOUT (or wherever) and then merely C<exit()>.
968 Or maybe you don't want to C<die> at all! Maybe you could use a
971 # Make all lookups fall back onto an English value,
972 # but after we log it for later fingerpointing.
973 my $lh_backup = ThisProject->get_handle('en');
974 open(LEX_FAIL_LOG, ">>wherever/lex.log") || die "GNAARGH $!";
976 my($failing_lh, $key, $params) = @_;
977 print LEX_FAIL_LOG scalar(localtime), "\t",
978 ref($failing_lh), "\t", $key, "\n";
979 return $lh_backup->maketext($key,@params);
982 Some users have expressed that they think this whole mechanism of
983 having a "fail" attribute at all, seems a rather pointless complication.
984 But I want Locale::Maketext to be usable for software projects of I<any>
985 scale and type; and different software projects have different ideas
986 of what the right thing is to do in failure conditions. I could simply
987 say that failure always throws an exception, and that if you want to be
988 careful, you'll just have to wrap every call to $lh->maketext in an
989 S<eval { }>. However, I want programmers to reserve the right (via
990 the "fail" attribute) to treat lookup failure as something other than
991 an exception of the same level of severity as a config file being
992 unreadable, or some essential resource being inaccessable.
994 One possibly useful value for the "fail" attribute is the method name
995 "failure_handler_auto". This is a method defined in class
996 Locale::Maketext itself. You set it with:
998 $lh->fail_with('failure_handler_auto');
1000 Then when you call $lh->maketext(I<key>, ...parameters...) and
1001 there's no I<key> in any of those lexicons, maketext gives up with
1003 return $lh->failure_handler_auto($key, @params);
1005 But failure_handler_auto, instead of dying or anything, compiles
1006 $key, caching it in $lh->{'failure_lex'}{$key} = $complied,
1007 and then calls the compiled value, and returns that. (I.e., if
1008 $key looks like bracket notation, $compiled is a sub, and we return
1009 &{$compiled}(@params); but if $key is just a plain string, we just
1012 The effect of using "failure_auto_handler"
1013 is like an AUTO lexicon, except that it 1) compiles $key even if
1014 it starts with "_", and 2) you have a record in the new hashref
1015 $lh->{'failure_lex'} of all the keys that have failed for
1016 this object. This should avoid your program dying -- as long
1017 as your keys aren't actually invalid as bracket code, and as
1018 long as they don't try calling methods that don't exist.
1020 "failure_auto_handler" may not be exactly what you want, but I
1021 hope it at least shows you that maketext failure can be mitigated
1022 in any number of very flexible ways. If you can formalize exactly
1023 what you want, you should be able to express that as a failure
1024 handler. You can even make it default for every object of a given
1025 class, by setting it in that class's init:
1028 my $lh = $_[0]; # a newborn handle
1030 $lh->fail_with('my_clever_failure_handler');
1033 sub my_clever_failure_handler {
1034 ...you clever things here...
1037 =head1 HOW TO USE MAKETEXT
1039 Here is a brief checklist on how to use Maketext to localize
1046 Decide what system you'll use for lexicon keys. If you insist,
1047 you can use opaque IDs (if you're nostalgic for C<catgets>),
1048 but I have better suggestions in the
1049 section "Entries in Each Lexicon", above. Assuming you opt for
1050 meaningful keys that double as values (like "Minimum ([_1]) is
1051 larger than maximum ([_2])!\n"), you'll have to settle on what
1052 language those should be in. For the sake of argument, I'll
1053 call this English, specifically American English, "en-US".
1057 Create a class for your localization project. This is
1058 the name of the class that you'll use in the idiom:
1061 my $lh = Projname::L10N->get_handle(...) || die "Language?";
1063 Assuming your call your class Projname::L10N, create a class
1064 consisting minimally of:
1066 package Projname::L10N;
1067 use base qw(Locale::Maketext);
1068 ...any methods you might want all your languages to share...
1070 # And, assuming you want the base class to be an _AUTO lexicon,
1071 # as is discussed a few sections up:
1077 Create a class for the language your internal keys are in. Name
1078 the class after the language-tag for that language, in lowercase,
1079 with dashes changed to underscores. Assuming your project's first
1080 language is US English, you should call this Projname::L10N::en_us.
1081 It should consist minimally of:
1083 package Projname::L10N::en_us;
1084 use base qw(Projname::L10N);
1090 (For the rest of this section, I'll assume that this "first
1091 language class" of Projname::L10N::en_us has
1096 Go and write your program. Everywhere in your program where
1099 print "Foobar $thing stuff\n";
1101 instead do it thru maketext, using no variable interpolation in
1104 print $lh->maketext("Foobar [_1] stuff\n", $thing);
1106 If you get tired of constantly saying C<print $lh-E<gt>maketext>,
1107 consider making a functional wrapper for it, like so:
1111 $lh = Projname::L10N->get_handle(...) || die "Language?";
1112 sub pmt (@) { print( $lh->maketext(@_)) }
1113 # "pmt" is short for "Print MakeText"
1115 # so if maketext fails, we see made the call to pmt
1117 Besides whole phrases meant for output, anything language-dependent
1118 should be put into the class Projname::L10N::en_us,
1119 whether as methods, or as lexicon entries -- this is discussed
1120 in the section "Entries in Each Lexicon", above.
1124 Once the program is otherwise done, and once its localization for
1125 the first language works right (via the data and methods in
1126 Projname::L10N::en_us), you can get together the data for translation.
1127 If your first language lexicon isn't an _AUTO lexicon, then you already
1128 have all the messages explicitly in the lexicon (or else you'd be
1129 getting exceptions thrown when you call $lh->maketext to get
1130 messages that aren't in there). But if you were (advisedly) lazy and are
1131 using an _AUTO lexicon, then you've got to make a list of all the phrases
1132 that you've so far been letting _AUTO generate for you. There are very
1133 many ways to assemble such a list. The most straightforward is to simply
1134 grep the source for every occurrence of "maketext" (or calls
1135 to wrappers around it, like the above C<pmt> function), and to log the
1140 You may at this point want to consider whether the your base class
1141 (Projname::L10N) that all lexicons inherit from (Projname::L10N::en,
1142 Projname::L10N::es, etc.) should be an _AUTO lexicon. It may be true
1143 that in theory, all needed messages will be in each language class;
1144 but in the presumably unlikely or "impossible" case of lookup failure,
1145 you should consider whether your program should throw an exception,
1146 emit text in English (or whatever your project's first language is),
1147 or some more complex solution as described in the section
1148 "Controlling Lookup Failure", above.
1152 Submit all messages/phrases/etc. to translators.
1154 (You may, in fact, want to start with localizing to I<one> other language
1155 at first, if you're not sure that you've property abstracted the
1156 language-dependent parts of your code.)
1158 Translators may request clarification of the situation in which a
1159 particular phrase is found. For example, in English we are entirely happy
1160 saying "I<n> files found", regardless of whether we mean "I looked for files,
1161 and found I<n> of them" or the rather distinct situation of "I looked for
1162 something else (like lines in files), and along the way I saw I<n>
1163 files." This may involve rethinking things that you thought quite clear:
1164 should "Edit" on a toolbar be a noun ("editing") or a verb ("to edit")? Is
1165 there already a conventionalized way to express that menu option, separate
1166 from the target language's normal word for "to edit"?
1168 In all cases where the very common phenomenon of quantification
1169 (saying "I<N> files", for B<any> value of N)
1170 is involved, each translator should make clear what dependencies the
1171 number causes in the sentence. In many cases, dependency is
1172 limited to words adjacent to the number, in places where you might
1173 expect them ("I found the-?PLURAL I<N>
1174 empty-?PLURAL directory-?PLURAL"), but in some cases there are
1175 unexpected dependencies ("I found-?PLURAL ..."!) as well as long-distance
1176 dependencies "The I<N> directory-?PLURAL could not be deleted-?PLURAL"!).
1178 Remind the translators to consider the case where N is 0:
1179 "0 files found" isn't exactly natural-sounding in any language, but it
1180 may be unacceptable in many -- or it may condition special
1181 kinds of agreement (similar to English "I didN'T find ANY files").
1183 Remember to ask your translators about numeral formatting in their
1184 language, so that you can override the C<numf> method as
1185 appropriate. Typical variables in number formatting are: what to
1186 use as a decimal point (comma? period?); what to use as a thousands
1187 separator (space? nonbreakinng space? comma? period? small
1188 middot? prime? apostrophe?); and even whether the so-called "thousands
1189 separator" is actually for every third digit -- I've heard reports of
1190 two hundred thousand being expressable as "2,00,000" for some Indian
1191 (Subcontinental) languages, besides the less surprising "S<200 000>",
1192 "200.000", "200,000", and "200'000". Also, using a set of numeral
1193 glyphs other than the usual ASCII "0"-"9" might be appreciated, as via
1194 C<tr/0-9/\x{0966}-\x{096F}/> for getting digits in Devanagari script
1195 (for Hindi, Konkani, others).
1197 The basic C<quant> method that Locale::Maketext provides should be
1198 good for many languages. For some languages, it might be useful
1199 to modify it (or its constituent C<numerate> method)
1200 to take a plural form in the two-argument call to C<quant>
1201 (as in "[quant,_1,files]") if
1202 it's all-around easier to infer the singular form from the plural, than
1203 to infer the plural form from the singular.
1205 But for other languages (as is discussed at length
1206 in L<Locale::Maketext::TPJ13|Locale::Maketext::TPJ13>), simple
1207 C<quant>/C<numerify> is not enough. For the particularly problematic
1208 Slavic languages, what you may need is a method which you provide
1209 with the number, the citation form of the noun to quantify, and
1210 the case and gender that the sentence's syntax projects onto that
1211 noun slot. The method would then be responsible for determining
1212 what grammatical number that numeral projects onto its noun phrase,
1213 and what case and gender it may override the normal case and gender
1214 with; and then it would look up the noun in a lexicon providing
1215 all needed inflected forms.
1219 You may also wish to discuss with the translators the question of
1220 how to relate different subforms of the same language tag,
1221 considering how this reacts with C<get_handle>'s treatment of
1222 these. For example, if a user accepts interfaces in "en, fr", and
1223 you have interfaces available in "en-US" and "fr", what should
1224 they get? You may wish to resolve this by establishing that "en"
1225 and "en-US" are effectively synonymous, by having one class
1226 zero-derive from the other.
1228 For some languages this issue may never come up (Danish is rarely
1229 expressed as "da-DK", but instead is just "da"). And for other
1230 languages, the whole concept of a "generic" form may verge on
1231 being uselessly vague, particularly for interfaces involving voice
1232 media in forms of Arabic or Chinese.
1236 Once you've localized your program/site/etc. for all desired
1237 languages, be sure to show the result (whether live, or via
1238 screenshots) to the translators. Once they approve, make every
1239 effort to have it then checked by at least one other speaker of
1240 that language. This holds true even when (or especially when) the
1241 translation is done by one of your own programmers. Some
1242 kinds of systems may be harder to find testers for than others,
1243 depending on the amount of domain-specific jargon and concepts
1244 involved -- it's easier to find people who can tell you whether
1245 they approve of your translation for "delete this message" in an
1246 email-via-Web interface, than to find people who can give you
1247 an informed opinion on your translation for "attribute value"
1248 in an XML query tool's interface.
1254 I recommend reading all of these:
1256 L<Locale::Maketext::TPJ13|Locale::Maketext::TPJ13> -- my I<The Perl
1257 Journal> article about Maketext. It explains many important concepts
1258 underlying Locale::Maketext's design, and some insight into why
1259 Maketext is better than the plain old approach of just having
1260 message catalogs that are just databases of sprintf formats.
1262 L<File::Findgrep|File::Findgrep> is a sample application/module
1263 that uses Locale::Maketext to localize its messages.
1265 L<I18N::LangTags|I18N::LangTags>.
1267 L<Win32::Locale|Win32::Locale>.
1269 RFC 3066, I<Tags for the Identification of Languages>,
1270 as at http://sunsite.dk/RFC/rfc/rfc3066.html
1272 RFC 2277, I<IETF Policy on Character Sets and Languages>
1273 is at http://sunsite.dk/RFC/rfc/rfc2277.html -- much of it is
1274 just things of interest to protocol designers, but it explains
1275 some basic concepts, like the distinction between locales and
1278 The manual for GNU C<gettext>. The gettext dist is available in
1279 C<ftp://prep.ai.mit.edu/pub/gnu/> -- get
1280 a recent gettext tarball and look in its "doc/" directory, there's
1281 an easily browsable HTML version in there. The
1282 gettext documentation asks lots of questions worth thinking
1283 about, even if some of their answers are sometimes wonky,
1284 particularly where they start talking about pluralization.
1286 The Locale/Maketext.pm source. Obverse that the module is much
1287 shorter than its documentation!
1289 =head1 COPYRIGHT AND DISCLAIMER
1291 Copyright (c) 1999-2001 Sean M. Burke. All rights reserved.
1293 This library is free software; you can redistribute it and/or modify
1294 it under the same terms as Perl itself.
1296 This program is distributed in the hope that it will be useful, but
1297 without any warranty; without even the implied warranty of
1298 merchantability or fitness for a particular purpose.
1302 Sean M. Burke C<sburke@cpan.org>