NAME
Data::Localize - Alternate Data Localization API
SYNOPSIS
use Data::Localize;
my $loc = Data::Localize->new();
$loc->add_localizer(
class => "Namespace", # Locale::Maketext-style .pm files
namespaces => [ "MyApp::I18N" ]
);
$loc->add_localizer(
class => "Gettext",
path => "/path/to/localization/data/*.po"
);
$loc->set_languages();
# or explicitly set one
# $loc->set_languages('en', 'ja' );
# looks under $self->languages, and checks if there are any
# localizers that can handle the job
$loc->localize( 'Hellow, [_1]!', 'John Doe' );
# You can enable "auto", which will be your last resort fallback.
# The key you give to the localize method will be used as the lexicon
$self->auto(1);DESCRIPTION
Data::Localize is an object oriented approach to localization, aimed to be an alternate choice for Locale::Maketext, Locale::Maketext::Lexicon, and Locale::Maketext::Simple.
RATIONALE
Functionality-wise, Locale::Maketext does what it advertises to do. Here's a few reasons why you might or might not choose Data::Localize over Locale::Maketext-based localizers:
Object-Oriented
Data::Localize is completely object-oriented. YMMV.
Faster
On some my benchmarks, Data::Localize is faster than Locale::Maketext by 50~80%. (But see PERFORMANCE)
Scalable For Large Amount Of Lexicons
Whereas Locale::Maketext generally stores the lexicons in memory, Data::Localize allows you to store this data in alternate storage. By default Data::Localize comes with a BerkeleyDB backend.
BASIC WORKING
STRUCTURE
Data::Localize is a wrapper around various Data::Localize::Localizer implementers (localizers). So if you don't specify any localizers, Data::Localize will do... nothing (unless you specify auto).
Localizers are the objects that do the actual localization. Localizers must register themselves to the Data::Localize parent, noting which languages it can handle (which usually is determined by the presence of data files like en.po, ja.po, etc). A special language ID of '*' is used to accept fallback cases. Localizers registered to handle '*' will be tried after all other language possibilities have been exhausted.
If the particular localizer cannot deal with the requested string, then it simply returns nothing.
AUTO-GENERATING LEXICONS
Locale::Maketext allows you to supply an "_AUTO" key in the lexicon hash, which allows you to pass a non-existing key to the localize() method, and use it as the actual lexicon, if no other applicable lexicons exists.
Locale::Maketext attaches this to the lexicon hash itself, but Data::Localizer differs in that it attaches to the Data::Localizer object itself, so you don't have to place _AUTO everywhere.
# here, we're deliberately not setting any localizers
my $loc = Data::Localize->new(auto => 1);
# previous auto => 1 will force Data::Localize to fallback to
# using the key ('Hello, [_1]') as the localization token.
print $loc->localize('Hello, [_1]', 'John Doe'), "\n";UTF8
All data is expected to be in decoded utf8. You must "use utf8" or decode them to Perl's internal representation for all values passed to Data::Localizer. We won't try to be smart for you. USE UTF8!
- Using Explicit decode()
-
use Encode q(decode decode_utf8); use Data::Localizer; my $loc = Data::Localize->new(...); $loc->localize( $key, decode( 'iso-2022-jp', $value ) ); # if $value is encoded utf8... # $loc->localize( $key, decode_utf8( $value ) ); - Using utf8
-
"use utf8" is simpler, but do note that it will affect ALL your literal strings in the current scope
use utf8; $loc->localize( $key, "some-utf8-key-here" );
USING ALTERNATE STORAGE
By default all lexicons are stored on memory, but if you're building an app with thousands and thousands of long messages, this might not be the ideal solution. In such cases, you can change where the lexicons get stored
my $loc = Data::Localize->new();
$loc->add_localizer(
class => 'Gettext',
path => '/path/to/data/*.po'
storage_class => 'BerkeleyDB',
storage_args => {
dir => '/path/to/really/fast/device'
}
);This would cause Data::Localize to put all the lexicon data in several BerkeleyDB files under /path/to/really/fast/device
Note that this approach would buy you no gain if you use Data::Localize::Namespace, as that approach by default expects everything to be in memory.
DEBUGGING
DEBUG
To enable debug tracing, either set DATA_LOCALIZE_DEBUG environment variable,
DATA_LOCALIZE_DEBUG=1 ./yourscript.plor explicitly define a function before loading Data::Localize:
BEGIN {
*Data::Localize::DEBUG = sub () { 1 };
}
use Data::Localize;METHODS
add_localizer
Adds a new localizer. You may either pass a localizer object, or arguments to your localizer's constructor:
$loc->add_localizer( YourLocalizer->new );
$loc->add_localizer(
class => "Namespace",
namespaces => [ 'Blah' ]
);localize
Localize the given string ID, using provided variables.
$localized_string = $loc->localize( $id, @args );detect_languages
Detects the current set of languages to use. If used in an CGI environment, will attempt to detect the language of choice from headers. See I18N::LanguageTags::Detect for details.
detect_languages_from_header
Detects the language from the given header value, or from HTTP_ACCEPT_LANGUAGES environment variable
localizers
Return a arrayref of localizers
add_localizer_map
Used internally.
set_localizer_map
Used internally.
find_localizers
Finds a localizer by its attribute. Currently only supports isa
my @locs = $loc->find_localizers(isa => 'Data::Localize::Gettext');set_languages
If used without any arguments, calls detect_languages() and sets the current language set to the result of detect_languages().
languages
Gets the current list of languages
add_fallback_languages
fallback_languages
count_localizers()
Return the number of localizers available
get_localizer_from_lang($lang)
Get appropriate localizer for language $lang
grep_localizers(\&sub)
Filter localizers
PERFORMANCE
tl;dr: Use one that fits your needs
Using explicit get_handle for every request
This benchmark assumes that you're fetching the lexicon anew for every request. This allows you to switch languages for every request
Benchmark run with Mac OS X (10.8.2) perl 5.16.1
Running benchmarks with
Locale::Maketext: 1.23
Data::Localize: 0.00023
Rate D::L(Namespace) L::M D::L(Gettext) D::L(Gettext+BDB)
D::L(Namespace) 5051/s -- -65% -73% -73%
L::M 14423/s 186% -- -24% -24%
D::L(Gettext) 18868/s 274% 31% -- -1%
D::L(Gettext+BDB) 18987/s 276% 32% 1% --Using cached lexicon objects for all
This benchmark assumes that you're fetching the lexicon once for a particular language, and you keep it in memory for reuse. This does NOT allow you to switch languages for every request.
Benchmark run with Mac OS X (10.8.2) perl 5.16.1
Running benchmarks with
Locale::Maketext: 1.23
Data::Localize: 0.00023
Rate D::L(Namespace) D::L(Gettext+BDB) D::L(Gettext) L::M
D::L(Namespace) 6023/s -- -65% -69% -96%
D::L(Gettext+BDB) 17202/s 186% -- -12% -87%
D::L(Gettext) 19548/s 225% 14% -- -86%
L::M 135993/s 2158% 691% 596% --TODO
Gettext style localization files -- Make it possible to decode them
CONTRIBUTORS
Dave Rolsky
AUTHOR
Daisuke Maki <daisuke@endeworks.jp>
COPYRIGHT
- The "MIT" License
-
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.