# BEGIN BPS TAGGED BLOCK {{{
#
# COPYRIGHT:
-#
-# This software is Copyright (c) 1996-2005 Best Practical Solutions, LLC
+#
+# This software is Copyright (c) 1996-2009 Best Practical Solutions, LLC
# <jesse@bestpractical.com>
#
# (Except where explicitly superseded by other copyright notices)
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
-# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+# 02110-1301 or visit their web page on the internet at
+# http://www.gnu.org/licenses/old-licenses/gpl-2.0.html.
#
#
# CONTRIBUTION SUBMISSION POLICY:
package RT::I18N;
use strict;
+use warnings;
+
use Locale::Maketext 1.04;
use Locale::Maketext::Lexicon 0.25;
use base ('Locale::Maketext::Fuzzy');
-use vars qw( %Lexicon );
-#If we're running on 5.6, we desperately need Encode::compat. But if we're on 5.8, we don't really need it.
-BEGIN { if ($] < 5.007001) {
-require Encode::compat;
-} }
use Encode;
-
use MIME::Entity;
use MIME::Head;
# I decree that this project's first language is English.
-%Lexicon = (
+our %Lexicon = (
'TEST_STRING' => 'Concrete Mixer',
'__Content-Type' => 'text/plain; charset=utf-8',
Initializes the lexicons used for localization.
-=begin testing
-
-use_ok (RT::I18N);
-ok(RT::I18N->Init);
-
-=end testing
=cut
sub Init {
require File::Glob;
+ my @lang = RT->Config->Get('LexiconLanguages');
+ @lang = ('*') unless @lang;
+
+ # load default functions
+ require substr(__FILE__, 0, -3) . '/i_default.pm';
+
# Load language-specific functions
- foreach my $language ( File::Glob::bsd_glob(substr(__FILE__, 0, -3) . "/*.pm")) {
- if ($language =~ /^([-\w\s.\/\\~:]+)$/) {
- require $1;
+ foreach my $file ( File::Glob::bsd_glob(substr(__FILE__, 0, -3) . "/*.pm") ) {
+ unless ( $file =~ /^([-\w\s\.\/\\~:]+)$/ ) {
+ warn("$file is tainted. not loading");
+ next;
}
- else {
- warn("$language is tainted. not loading");
- }
+ $file = $1;
+
+ my ($lang) = ($file =~ /([^\\\/]+?)\.pm$/);
+ next unless grep $_ eq '*' || $_ eq $lang, @lang;
+ require $file;
}
- my @lang = @RT::LexiconLanguages;
- @lang = ('*') unless @lang;
+ my %import;
+ foreach my $l ( @lang ) {
+ $import{$l} = [
+ Gettext => (substr(__FILE__, 0, -3) . "/$l.po"),
+ Gettext => "$RT::LocalLexiconPath/*/$l.po",
+ Gettext => "$RT::LocalLexiconPath/$l.po",
+ ];
+ push @{ $import{$l} }, map {(Gettext => "$_/$l.po")} RT->PluginDirs('po');
+ }
# Acquire all .po files and iterate them into lexicons
- Locale::Maketext::Lexicon->import({
- _decode => 1, map {
- $_ => [
- Gettext => (substr(__FILE__, 0, -3) . "/$_.po"),
- Gettext => "$RT::LocalLexiconPath/*/$_.po",
- Gettext => "$RT::LocalLexiconPath/$_.po",
- ],
- } @lang
- });
+ Locale::Maketext::Lexicon->import({ _decode => 1, %import });
return 1;
}
Returns the encoding of the current lexicon, as yanked out of __ContentType's "charset" field.
If it can't find anything, it returns 'ISO-8859-1'
-=begin testing
-
-ok(my $chinese = RT::I18N->get_handle('zh_tw'));
-ok(UNIVERSAL::can($chinese, 'maketext'));
-ok($chinese->maketext('__Content-Type') =~ /utf-8/i, "Found the utf-8 charset for traditional chinese in the string ".$chinese->maketext('__Content-Type'));
-ok($chinese->encoding eq 'utf-8', "The encoding is 'utf-8' -".$chinese->encoding);
-
-ok(my $en = RT::I18N->get_handle('en'));
-ok(UNIVERSAL::can($en, 'maketext'));
-ok($en->encoding eq 'utf-8', "The encoding ".$en->encoding." is 'utf-8'");
-
-=end testing
=cut
=head2 SetMIMEEntityToUTF8 $entity
-An utility method which will try to convert entity body into utf8.
+An utility function which will try to convert entity body into utf8.
It's now a wrap-up of SetMIMEEntityToEncoding($entity, 'utf-8').
=cut
# }}}
+# {{{ IsTextualContentType
+
+=head2 IsTextualContentType $type
+
+An utility function that determines whether $type is I<textual>, meaning
+that it can sensibly be converted to Unicode text.
+
+Currently, it returns true iff $type matches this regular expression
+(case-insensitively):
+
+ ^(?:text/(?:plain|html)|message/rfc822)\b
+
+# }}}
+
+=cut
+
+sub IsTextualContentType {
+ my $type = shift;
+ ($type =~ m{^(?:text/(?:plain|html)|message/rfc822)\b}i) ? 1 : 0;
+}
+
# {{{ SetMIMEEntityToEncoding
=head2 SetMIMEEntityToEncoding $entity, $encoding
-An utility method which will try to convert entity body into specified
+An utility function which will try to convert entity body into specified
charset encoding (encoded as octets, *not* unicode-strings). It will
iterate all the entities in $entity, and try to convert each one into
specified charset if whose Content-Type is 'text/plain'.
-This method doesn't return anything meaningful.
+This function doesn't return anything meaningful.
=cut
sub SetMIMEEntityToEncoding {
my ( $entity, $enc, $preserve_words ) = ( shift, shift, shift );
- #if ( $entity->is_multipart ) {
- #$RT::Logger->crit("This entity is a multipart " . $entity->head->as_string);
- SetMIMEEntityToEncoding( $_, $enc, $preserve_words ) foreach $entity->parts;
- #}
+ # do the same for parts first of all
+ SetMIMEEntityToEncoding( $_, $enc, $preserve_words ) foreach $entity->parts;
my $charset = _FindOrGuessCharset($entity) or return;
# one and only normalization
}
# If this is a textual entity, we'd need to preserve its original encoding
- $head->add( "X-RT-Original-Encoding" => $charset )
- if $head->mime_attr('content-type.charset') or $head->mime_type =~ /^text/;
-
+ $head->replace( "X-RT-Original-Encoding" => $charset )
+ if $head->mime_attr('content-type.charset') or IsTextualContentType($head->mime_type);
- return unless ( $head->mime_type =~ qr{^(text/plain|message/rfc822)$}i );
-
+ return unless IsTextualContentType($head->mime_type);
my $body = $entity->bodyhandle;
# {{{ Convert the body
eval {
- $RT::Logger->debug("Converting '$charset' to '$enc' for ". $head->mime_type . " - ". $head->get('subject'));
+ $RT::Logger->debug("Converting '$charset' to '$enc' for ". $head->mime_type . " - ". ($head->get('subject') || 'Subjectless message'));
# NOTE:: see the comments at the end of the sub.
Encode::_utf8_off( $lines[$_] ) foreach ( 0 .. $#lines );
my $str = shift;
my $enc = shift;
-
- @_ = $str =~ m/([^=]*)=\?([^?]+)\?([QqBb])\?([^?]+)\?=([^=]*)/g;
-
+ @_ = $str =~ m/(.*?)=\?([^?]+)\?([QqBb])\?([^?]+)\?=([^=]*)/gcs;
return ($str) unless (@_);
+ # add everything that hasn't matched to the end of the latest
+ # string in array this happen when we have 'key="=?encoded?="; key="plain"'
+ $_[-1] .= substr($str, pos $str);
+
$str = "";
while (@_) {
my ($prefix, $charset, $encoding, $enc_str, $trailing) =
- (shift, shift, shift, shift, shift);
+ (shift, shift, lc shift, shift, shift);
$trailing =~ s/\s?\t?$//; # Observed from Outlook Express
- if ($encoding eq 'Q' or $encoding eq 'q') {
+ if ( $encoding eq 'q' ) {
use MIME::QuotedPrint;
$enc_str =~ tr/_/ /; # Observed from Outlook Express
$enc_str = decode_qp($enc_str);
- } elsif ($encoding eq 'B' or $encoding eq 'b') {
+ } elsif ( $encoding eq 'b' ) {
use MIME::Base64;
$enc_str = decode_base64($enc_str);
} else {
- $RT::Logger->warning("RT::I18N::DecodeMIMEWordsToCharset got a " .
- "strange encoding: $encoding.");
+ $RT::Logger->warning("Incorrect encoding '$encoding' in '$str', "
+ ."only Q(uoted-printable) and B(ase64) are supported");
}
# now we have got a decoded subject, try to convert into the encoding
# until this is fixed, we must escape any string containing a comma or semicolon
# this is only a bandaid
- $enc_str = qq{"$enc_str"} if ($enc_str =~ /[,;]/);
+ # Some _other_ MUAs encode quotes _already_, and double quotes
+ # confuse us a lot, so only quote it if it isn't quoted
+ # already.
+ $enc_str = qq{"$enc_str"} if $enc_str =~ /[,;]/ and $enc_str !~ /^".*"$/;
+
$str .= $prefix . $enc_str . $trailing;
}
+ # We might have \n without trailing whitespace, which will result in
+ # invalid headers.
+ $str =~ s/\n//g;
+
return ($str)
}
my $head_only = shift;
my $head = $entity->head;
- if ($head->mime_attr("content-type.charset")) {
- return $head->mime_attr("content-type.charset");
+ if ( my $charset = $head->mime_attr("content-type.charset") ) {
+ return $charset;
}
if ( !$head_only and $head->mime_type =~ m{^text/}) {
sub _GuessCharset {
my $fallback = 'iso-8859-1';
- my $charset;
- if ( @RT::EmailInputEncodings and eval { require Encode::Guess; 1 } ) {
- Encode::Guess->set_suspects(@RT::EmailInputEncodings);
+ # if $_[0] is null/empty, we don't guess its encoding
+ return $fallback unless defined $_[0] && length $_[0];
+
+ my $charset;
+ my @encodings = RT->Config->Get('EmailInputEncodings');
+ if ( @encodings and eval { require Encode::Guess; 1 } ) {
+ Encode::Guess->set_suspects( @encodings );
my $decoder = Encode::Guess->guess( $_[0] );
+ if ( defined($decoder) ) {
if ( ref $decoder ) {
$charset = $decoder->name;
$RT::Logger->debug("Guessed encoding: $charset");
my %matched = map { $_ => 1 } split(/ or /, $1);
return 'utf-8' if $matched{'utf8'}; # one and only normalization
- foreach my $suspect (@RT::EmailInputEncodings) {
+ foreach my $suspect (RT->Config->Get('EmailInputEncodings')) {
next unless $matched{$suspect};
$RT::Logger->debug("Encode::Guess ambiguous ($decoder); using $suspect");
$charset = $suspect;
else {
$RT::Logger->warning("Encode::Guess failed: $decoder; fallback to $fallback");
}
+ }
+ else {
+ $RT::Logger->warning("Encode::Guess failed: decoder is undefined; fallback to $fallback");
+ }
}
- else {
- $RT::Logger->warning("Cannot Encode::Guess; fallback to $fallback");
+ elsif ( @encodings && $@ ) {
+ $RT::Logger->error("You have set EmailInputEncodings, but we couldn't load Encode::Guess: $@");
+ } else {
+ $RT::Logger->warning("No EmailInputEncodings set, fallback to $fallback");
}
- return($charset || $fallback);
+ return ($charset || $fallback);
}
# }}}
projects / freeside.git / blobdiff