--- loncom/localize/lonlocal.pm 2003/09/16 22:40:25 1.3 +++ loncom/localize/lonlocal.pm 2010/12/06 00:35:19 1.60.4.2 @@ -1,7 +1,7 @@ # The LearningOnline Network with CAPA # Localization routines # -# $Id: lonlocal.pm,v 1.3 2003/09/16 22:40:25 www Exp $ +# $Id: lonlocal.pm,v 1.60.4.2 2010/12/06 00:35:19 raeburn Exp $ # # Copyright Michigan State University Board of Trustees # @@ -28,40 +28,556 @@ ###################################################################### ###################################################################### +=pod + +=head1 NAME + +Apache::lonlocal - provides localization services + +=head1 SYNOPSIS + +lonlocal provides localization services for LON-CAPA programmers based +on Locale::Maketext. See +C +for more information on Maketext. + +=head1 OVERVIEWX + +As of LON-CAPA 1.1, we've started to localize LON-CAPA using the +Locale::Maketext module. Internationalization is the bulk of the work +right now (pre-1.1); localizing can be done anytime, and involves +little or no programming. + +The internationalization process involves putting a wrapper around +on-screen user messages and menus and turning them into keys, +which the MaketextX library translates into the desired +language output using a look-up table ("lexicon").X + +As keys we are currently using the plain English messages, and +Maketext is configured to replace the message by its own key if no +translation is found. This makes it easy to phase in the +internationalization without disturbing the screen output. + +Internationalization is somewhat tedious and effectively impossible +for a non-fluent speaker to perform, but is fairly easy to create +translations, requiring no programming skill. As a result, this is one +area where you can really help LON-CAPA out, even if you aren't a +programmer, and we'd really appreciate it. + +=head1 How To Localize Handlers For Programmers + +Into the "use" section of a module, we need to insert + + use Apache::lonlocal; + +Note that there are B, we B to pollute our +namespace. + +Inside might be something like this + + sub message { + my $status=shift; + my $message='Status unknown'; + if ($status eq 'WON') { + $message='You have won.'; + } elsif ($status eq 'LOST') { + $message='You are a total looser.'; + } + return $message; + } + ... + $r->print('

Gamble your Homework Points

'); + ... + $r->print(<Rules: + No purchase necessary. Illegal where not allowed. + ENDMSG + +We have to now wrap the subroutine &mt()X ("maketext") around our +messages, but not around markup, etc. We also want minimal disturbance. +The first two examples are easy: + + sub message { + my $status=shift; + my $message='Status unknown'; + if ($status eq 'WON') { + $message='You have won.'; + } elsif ($status eq 'LOST') { + $message='You are a total looser.'; + } + return &mt($message); + } + ... + $r->print('

'.&mt('Gamble your Homework Points').'

'); + +The last one is a bummer, since you cannot call subroutines inside of +(< 'Rules', 'disclaimer' => + 'No purchase necessary. Illegal where not allowed.'); + $r->print(<$lt{'header'}: + $lt{'disclaimer'} + ENDMSG + +As a programmer, your job is done here. If everything worked, you +should see no changes on the screen. + +=head1 How To Localize LON-CAPA for Translators + +As a translator, you need to provide the lexicon for the keys, which in +this case is the plain text message. The lexicons sit in +loncom/localize/localize, with the language code as filename, for +example de.pm for the German translation. The file then simply looks +like this: + + 'You have won.' + => 'Sie haben gewonnen.', + + 'You are a total looser.' + => 'Sie sind der totale Verlierer.', + + 'Rules' + => 'Regeln', + + 'No purchase necessary. Illegal where not allowed.' + => 'Es ist erlaubt, einfach zu verlieren, und das ist Ihre Schuld.' + + +Comments may be added with the # symbol, which outside of a string +(the things with the apostrophe surrounding them, which are the +keys and translations) will cause the translation routines to +ignore the rest of the line. + +This is a relatively easy task, and any help is appreciated. + +Maketext can do a whole lot more, see +C +but for most purposes, we do not have to mess with that. + +=cut + package Apache::lonlocal; use strict; use Apache::localize; -use Apache::File; +use locale; +use POSIX qw(locale_h strftime); +use DateTime(); +use DateTime::TimeZone; +use DateTime::Locale; require Exporter; our @ISA = qw (Exporter); -our @EXPORT = qw(mt); +our @EXPORT = qw(mt mtn ns mt_user); + +my %mtcache=(); # ========================================================= The language handle -use vars qw($lh); +use vars qw($lh $current_language); # ===================================================== The "MakeText" function sub mt (@) { - unless ($ENV{'environment.translator'}) { - return $lh->maketext(@_); +# open(LOG,'>>/home/www/loncapa/loncom/localize/localize/newphrases.txt'); +# print LOG (@_[0]."\n"); +# close(LOG); + if ($lh) { + if ($_[0] eq '') { + if (wantarray) { + return @_; + } else { + return $_[0]; + } + } else { + if ($#_>0) { return $lh->maketext(@_); } + if ($mtcache{$current_language.':'.$_[0]}) { + return $mtcache{$current_language.':'.$_[0]}; + } + my $translation=$lh->maketext(@_); + $mtcache{$current_language.':'.$_[0]}=$translation; + return $translation; + } + } else { + if (wantarray) { + return @_; + } else { + return $_[0]; + } + } +} + +sub mt_user { + my ($user_lh,@what) = @_; + if ($user_lh) { + if ($what[0] eq '') { + if (wantarray) { + return @what; + } else { + return $what[0]; + } + } else { + return $user_lh->maketext(@what); + } } else { - return ''.$lh->maketext(@_).''; + if (wantarray) { + return @what; + } else { + return $what[0]; + } + } +} + +# ============================================================== What language? + +sub current_language { + if ($lh) { + my $lang=$lh->maketext('language_code'); + return ($lang eq 'language_code'?'en':$lang); } + return 'en'; +} + +sub preferred_languages { + my $defdom = &Apache::lonnet::default_login_domain(); + foreach my $domtype ($Apache::lonnet::env{'user.domain'},$Apache::lonnet::env{'request.role.domain'},$defdom) { + next if ($domtype eq ''); + if ($domtype =~ /^\w+ci/) { + return ($domtype); + } elsif ($domtype =~ /^(\w+c)itest$/) { + return ($1.'t'); + } else { + return ('gci'); + } + } + my @languages=(); + if (($Apache::lonnet::env{'request.role.adv'}) && ($Apache::lonnet::env{'form.languages'})) { + @languages=(@languages,split(/\s*(\,|\;|\:)\s*/,$Apache::lonnet::env{'form.languages'})); + } + if ($Apache::lonnet::env{'course.'.$Apache::lonnet::env{'request.course.id'}.'.languages'}) { + @languages=(@languages,split(/\s*(\,|\;|\:)\s*/, + $Apache::lonnet::env{'course.'.$Apache::lonnet::env{'request.course.id'}.'.languages'})); + } + + if ($Apache::lonnet::env{'environment.languages'}) { + @languages=(@languages, + split(/\s*(\,|\;|\:)\s*/,$Apache::lonnet::env{'environment.languages'})); + } + my $browser=$ENV{'HTTP_ACCEPT_LANGUAGE'}; + if ($browser) { + my @browser = + map { (split(/\s*;\s*/,$_))[0] } (split(/\s*,\s*/,$browser)); + push(@languages,@browser); + } + + foreach my $domtype ($Apache::lonnet::env{'user.domain'},$Apache::lonnet::env{'request.role.domain'},$defdom) { + if ($domtype ne '') { + my %domdefs = &Apache::lonnet::get_domain_defaults($domtype); + if ($domdefs{'lang_def'} ne '') { + push(@languages,$domdefs{'lang_def'}); + } + } + } + return &get_genlanguages(@languages); +} + +sub get_genlanguages { + my (@languages) = @_; +# turn "en-ca" into "en-ca,en" + my @genlanguages; + foreach my $lang (@languages) { + unless ($lang=~/\w/) { next; } + push(@genlanguages,$lang); + if ($lang=~/(\-|\_)/) { + push(@genlanguages,(split(/(\-|\_)/,$lang))[0]); + } + } + #uniqueify the languages list + my %count; + @genlanguages = map { $count{$_}++ == 0 ? $_ : () } @genlanguages; + return @genlanguages; +} + +# ============================================================== What encoding? + +sub current_encoding { + my $default='UTF-8'; +# UTF-8 character encoding needed for the whole LON-CAPA system +# (interface language and homework problem content) +# See Bugzilla 5702 vs. 2189 and 4067 +# if ($Apache::lonnet::env{'browser.os'} eq 'win' && +# $Apache::lonnet::env{'browser.type'} eq 'explorer') { +# $default='ISO-8859-1'; +# } + if ($lh) { + my $enc=$lh->maketext('char_encoding'); + return ($enc eq 'char_encoding'?$default:$enc); + } else { + return $default; + } +} + +# =============================================================== Which locale? +# Refer to locale -a +# +sub current_locale { + if ($lh) { + my $enc=$lh->maketext('lang_locale'); + return ($enc eq 'lang_locale'?'':$enc); + } else { + return undef; + } +} + +# ============================================================== Translate hash + +sub texthash { + my %hash=@_; + foreach (keys %hash) { + $hash{$_}=&mt($hash{$_}); + } + return %hash; } # ========= Get a handle (do not invoke in vain, leave this to access handlers) sub get_language_handle { - $lh=Apache::localize->get_handle(&Apache::loncommon::preferred_languages); + my $r=shift; + if ($r) { + my $headers=$r->headers_in; + $ENV{'HTTP_ACCEPT_LANGUAGE'}=$headers->{'Accept-language'}; + } + my @languages=&preferred_languages(); + $ENV{'HTTP_ACCEPT_LANGUAGE'}=''; + $lh=Apache::localize->get_handle(@languages); + $current_language=¤t_language(); + if ($r) { + $r->content_languages([¤t_language()]); + } +### setlocale(LC_ALL,¤t_locale); +} + +# ========================================================== Localize localtime +sub gettimezone { + my ($timezone) = @_; + if ($timezone ne '') { + if (!DateTime::TimeZone->is_valid_name($timezone)) { + $timezone = 'local'; + } + return $timezone; + } + my $cid = $Apache::lonnet::env{'request.course.id'}; + if ($cid ne '') { + if ($Apache::lonnet::env{'course.'.$cid.'.timezone'}) { + $timezone = $Apache::lonnet::env{'course.'.$cid.'.timezone'}; + } else { + my $cdom = $Apache::lonnet::env{'course.'.$cid.'.domain'}; + if ($cdom ne '') { + my %domdefaults = &Apache::lonnet::get_domain_defaults($cdom); + if ($domdefaults{'timezone_def'} ne '') { + $timezone = $domdefaults{'timezone_def'}; + } + } + } + } elsif ($Apache::lonnet::env{'request.role.domain'} ne '') { + my %uroledomdefs = + &Apache::lonnet::get_domain_defaults($Apache::lonnet::env{'request.role.domain'}); + if ($uroledomdefs{'timezone_def'} ne '') { + $timezone = $uroledomdefs{'timezone_def'}; + } + } elsif ($Apache::lonnet::env{'user.domain'} ne '') { + my %udomdefaults = + &Apache::lonnet::get_domain_defaults($Apache::lonnet::env{'user.domain'}); + if ($udomdefaults{'timezone_def'} ne '') { + $timezone = $udomdefaults{'timezone_def'}; + } + } + if ($timezone ne '') { + if (DateTime::TimeZone->is_valid_name($timezone)) { + return $timezone; + } + } + return 'local'; +} + +sub locallocaltime { + my ($thistime,$timezone) = @_; + + if (!defined($thistime) || $thistime eq '') { + return &mt('Never'); + } + if (($thistime < 0) || ($thistime eq 'NaN')) { + &Apache::lonnet::logthis("Unexpected time (negative or NaN) '$thistime' passed to lonlocal::locallocaltime"); + return &mt('Never'); + } + if ($thistime !~ /^\d+$/) { + &Apache::lonnet::logthis("Unexpected non-numeric time '$thistime' passed to lonlocal::locallocaltime"); + return &mt('Never'); + } + + my $dt = DateTime->from_epoch(epoch => $thistime) + ->set_time_zone(gettimezone($timezone)); + + # TimeZone tries to determine the 'local' timezone from $ENV{TZ} if this + # fails it searches through various system files. Under certain + # circumstances this is an extremly expensive operation. + # So after the first run we store the timezone in $ENV{TZ} to significantly + # speed up future lookups. + $ENV{TZ} = $dt->time_zone()->name() + if (! $ENV{TZ} && gettimezone($timezone) eq 'local'); + + if ((¤t_language=~/^en/) || (!$lh)) { + + return $dt->strftime("%a %b %e %I:%M:%S %P %Y (%Z)"); + } else { + my $format=$lh->maketext('date_locale'); + if ($format eq 'date_locale') { + return $dt->strftime("%a %b %e %I:%M:%S %P %Y (%Z)"); + } + my $time_zone = $dt->time_zone_short_name(); + my $seconds = $dt->second(); + my $minutes = $dt->minute(); + my $twentyfour = $dt->hour(); + my $day = $dt->day_of_month(); + my $mon = $dt->month()-1; + my $year = $dt->year(); + my $wday = $dt->wday(); + if ($wday==7) { $wday=0; } + my $month =(split(/\,/,$lh->maketext('date_months')))[$mon]; + my $weekday=(split(/\,/,$lh->maketext('date_days')))[$wday]; + if ($seconds<10) { + $seconds='0'.$seconds; + } + if ($minutes<10) { + $minutes='0'.$minutes; + } + my $twelve=$twentyfour; + my $ampm; + if ($twelve>12) { + $twelve-=12; + $ampm=$lh->maketext('date_pm'); + } else { + $ampm=$lh->maketext('date_am'); + } + foreach ('seconds','minutes','twentyfour','twelve','day','year', + 'month','weekday','ampm') { + $format=~s/\$$_/eval('$'.$_)/gse; + } + return $format." ($time_zone)"; + } +} + +sub getdatelocale { + my ($datelocale,$locale_obj); + if ($Apache::lonnet::env{'course.'.$Apache::lonnet::env{'request.course.id'}.'.datelocale'}) { + $datelocale = $Apache::lonnet::env{'course.'.$Apache::lonnet::env{'request.course.id'}.'.datelocale'}; + } elsif ($Apache::lonnet::env{'request.course.id'} ne '') { + my $cdom = $Apache::lonnet::env{'course.'.$Apache::lonnet::env{'request.course.id'}.'.domain'}; + if ($cdom ne '') { + my %domdefaults = &Apache::lonnet::get_domain_defaults($cdom); + if ($domdefaults{'datelocale_def'} ne '') { + $datelocale = $domdefaults{'datelocale_def'}; + } + } + } elsif ($Apache::lonnet::env{'user.domain'} ne '') { + my %udomdefaults = &Apache::lonnet::get_domain_defaults($Apache::lonnet::env{'user.domain'}); + if ($udomdefaults{'datelocale_def'} ne '') { + $datelocale = $udomdefaults{'datelocale_def'}; + } + } + if ($datelocale ne '') { + eval { + $locale_obj = DateTime::Locale->load($datelocale); + }; + if (!$@) { + if ($locale_obj->id() eq $datelocale) { + return $locale_obj; + } + } + } + return $locale_obj; +} + +=pod + +=item * normalize_string + +Normalize string (reduce fragility in the lexicon files) + +This normalizes a string to reduce fragility in the lexicon files of +huge messages (such as are used by the helper), and allow useful +formatting: reduce all consecutive whitespace to a single space, +and remove all HTML + +=cut + +sub normalize_string { + my $s = shift; + $s =~ s/\s+/ /g; + $s =~ s/<[^>]+>//g; + # Pop off beginning or ending spaces, which aren't good + $s =~ s/^\s+//; + $s =~ s/\s+$//; + return $s; +} + +=pod + +=item * ns + +alias for normalize_string; recommend using it only in the lexicon + +=cut + +sub ns { + return normalize_string(@_); +} + +=pod + +=item * mtn + +mtn: call the mt function and the normalization function easily. +Returns original non-normalized string if there was no translation + +=cut + +sub mtn (@) { + my @args = @_; # don't want to modify caller's string; if we + # didn't care about that we could set $_[0] + # directly + $args[0] = normalize_string($args[0]); + my $translation = &mt(@args); + if ($translation ne $args[0]) { + return $translation; + } else { + return $_[0]; + } +} + +# ---------------------------------------------------- Replace MT{...} in files + +sub transstatic { + my $strptr=shift; + $$strptr=~s/MT\{([^\}]*)\}/&mt($1)/gse; +} + +=pod + +=item * mt_escape + +mt_escape takes a string reference and escape the [] in there so mt +will leave them as is and not try to expand them + +=cut + +sub mt_escape { + my ($str_ref) = @_; + $$str_ref =~s/~/~~/g; + $$str_ref =~s/([\[\]])/~$1/g; } 1;