texinfo/tp/Texinfo/Convert Converter.pm NodeNam...

[Patrice Dumas]

CVSROOT:        /sources/texinfo
Module name:    texinfo
Changes by:     Patrice Dumas <pertusus>        11/08/11 21:07:16

Modified files:
        tp/Texinfo/Convert: Converter.pm NodeNameNormalization.pm 
                            Text.pm Unicode.pm 

Log message:
        Don't pass a fallback function to unicode_accent.
        Stop processing a stack as unicode as soon as it cannot be represented 
as
        unicode.
        Allready handle upper casing in eight_bit_accents and unicode_accents.

CVSWeb URLs:
http://cvs.savannah.gnu.org/viewcvs/texinfo/tp/Texinfo/Convert/Converter.pm?cvsroot=texinfo&r1=1.23&r2=1.24
http://cvs.savannah.gnu.org/viewcvs/texinfo/tp/Texinfo/Convert/NodeNameNormalization.pm?cvsroot=texinfo&r1=1.10&r2=1.11
http://cvs.savannah.gnu.org/viewcvs/texinfo/tp/Texinfo/Convert/Text.pm?cvsroot=texinfo&r1=1.48&r2=1.49
http://cvs.savannah.gnu.org/viewcvs/texinfo/tp/Texinfo/Convert/Unicode.pm?cvsroot=texinfo&r1=1.12&r2=1.13

Patches:
Index: Converter.pm
===================================================================
RCS file: /sources/texinfo/texinfo/tp/Texinfo/Convert/Converter.pm,v
retrieving revision 1.23
retrieving revision 1.24
diff -u -b -r1.23 -r1.24
--- Converter.pm        2 Aug 2011 21:01:34 -0000       1.23
+++ Converter.pm        11 Aug 2011 21:07:16 -0000      1.24
@@ -595,10 +595,11 @@
   return xml_accent($text, $command, 1);
 }
 
-sub xml_accents($$)
+sub xml_accents($$;$)
 { 
   my $self = shift;
   my $accent = shift;
+  my $in_upper_case = shift;
   my $format_accents;
   if ($self->get_conf('USE_NUMERIC_ENTITY')) {
     $format_accents = \&xml_accent_numeric_entities;

Index: NodeNameNormalization.pm
===================================================================
RCS file: /sources/texinfo/texinfo/tp/Texinfo/Convert/NodeNameNormalization.pm,v
retrieving revision 1.10
retrieving revision 1.11
diff -u -b -r1.10 -r1.11
--- NodeNameNormalization.pm    4 Aug 2011 23:22:27 -0000       1.10
+++ NodeNameNormalization.pm    11 Aug 2011 21:07:16 -0000      1.11
@@ -255,8 +255,15 @@
     # commands with braces
     } elsif ($accent_commands{$root->{'cmdname'}}) {
       return '' if (!$root->{'args'});
-      my $accented_char = 
Texinfo::Convert::Unicode::unicode_accent(_convert($root->{'args'}->[0]), 
-                         $root, \&Texinfo::Convert::Text::ascii_accent);
+      my $accent_text = _convert($root->{'args'}->[0]);
+      my $accented_char 
+        = Texinfo::Convert::Unicode::unicode_accent($accent_text, 
+                         #$root, \&Texinfo::Convert::Text::ascii_accent);
+                                                    $root);
+      if (!defined($accented_char)) {
+        $accented_char = Texinfo::Convert::Text::ascii_accent($accent_text,
+                                                              $root);
+      }
       if ($in_sc) {
         return uc ($accented_char);
       } else {

Index: Text.pm
===================================================================
RCS file: /sources/texinfo/texinfo/tp/Texinfo/Convert/Text.pm,v
retrieving revision 1.48
retrieving revision 1.49
diff -u -b -r1.48 -r1.49
--- Text.pm     4 Aug 2011 23:22:28 -0000       1.48
+++ Text.pm     11 Aug 2011 21:07:16 -0000      1.49
@@ -641,11 +641,12 @@
   return ($eight_bit, $codepoint);
 }
 
-sub eight_bit_accents($$$)
+sub eight_bit_accents($$$;$)
 {
   my $current = shift;
   my $encoding = shift;
   my $convert_accent = shift;
+  my $in_upper_case = shift;
 
   my $debug;
   #$debug = 1;
@@ -664,19 +665,36 @@
   my @results_stack;
 
   while (1) {
-    $current_result 
-      = Texinfo::Convert::Unicode::unicode_accent($current_result, $accent,
-                                                    $convert_accent);
-    push @results_stack, [$current_result, $accent];
+    my $unicode_formatted_accent
+      = Texinfo::Convert::Unicode::unicode_accent($current_result, $accent);
+    if (!defined($unicode_formatted_accent)) {
+      last;
+    }
+    $current_result = $unicode_formatted_accent;
+    $unicode_formatted_accent = uc($unicode_formatted_accent)
+      if ($in_upper_case);
+    push @results_stack, [$unicode_formatted_accent, $accent];
     last if ($accent eq $current);
     $accent = $accent->{'parent'}->{'parent'};
   }
 
+  if ($accent ne $current) {
+    while (1) {
+      push @results_stack, [undef, $accent];
+      last if ($accent eq $current);
+      $accent = $accent->{'parent'}->{'parent'};
+    }
+  }
+
   if ($debug) {
-    print STDERR "stack: ".join('|',@$stack)."\nPARTIAL_RESULATS_STACK:\n";
+    print STDERR "stack: ".join('|',@$stack)."\nPARTIAL_RESULTS_STACK:\n";
     foreach my $partial_result (@results_stack) {
+      if (defined($partial_result->[0])) {
       print STDERR "   -> ".Encode::encode('utf8', $partial_result->[0])
                             ."|$partial_result->[1]->{'cmdname'}\n";
+      } else {
+        print STDERR "   -> NO UTF8 |$partial_result->[1]->{'cmdname'}\n";
+      }
     }
   }
 
@@ -691,6 +709,7 @@
   my $eight_bit_command_index = -1;
   foreach my $partial_result (@results_stack) {
     my $char = $partial_result->[0];
+    last if (!defined($char));
 
     my ($new_eight_bit, $new_codepoint) = _eight_bit_and_unicode_point($char,
                                                            $encoding_map_name);
@@ -773,16 +792,31 @@
 }
 
 # format a stack of accents as unicode
-sub unicode_accents ($$)
+sub unicode_accents ($$;$)
 {
   my $current = shift;
-  my $format_accents = shift;
+  my $format_accent = shift;
+  my $in_upper_case = shift;
   my ($result, $innermost_accent, $stack) = _find_innermost_accent($current,
           'utf-8');
+  my @stack_accent_commands = reverse(@$stack);
 
-  foreach my $accent_command (reverse(@$stack)) {
-    $result = Texinfo::Convert::Unicode::unicode_accent($result, 
-       {'cmdname' => $accent_command}, $format_accents);
+  while (@stack_accent_commands) {
+    my $accent_command = shift @stack_accent_commands;
+    my $formatted_result
+     = Texinfo::Convert::Unicode::unicode_accent($result, 
+       {'cmdname' => $accent_command});
+    if (!defined($formatted_result)) {
+      push @stack_accent_commands, $accent_command;
+    } else {
+      $result = $formatted_result;
+    }
+  }
+  $result = uc ($result) if ($in_upper_case);
+  while (@stack_accent_commands) {
+    my $accent_command = shift @stack_accent_commands;
+    $result = &$format_accent($result, 
+       {'cmdname' => $accent_command}, $in_upper_case);
   }
   return $result;
 }

Index: Unicode.pm
===================================================================
RCS file: /sources/texinfo/texinfo/tp/Texinfo/Convert/Unicode.pm,v
retrieving revision 1.12
retrieving revision 1.13
diff -u -b -r1.12 -r1.13
--- Unicode.pm  31 Jul 2011 23:21:49 -0000      1.12
+++ Unicode.pm  11 Aug 2011 21:07:16 -0000      1.13
@@ -803,14 +803,15 @@
 );
 
 
-sub unicode_accent($$$)
+sub unicode_accent($$)
 {
   my $text = shift;
   my $command = shift;
-  my $fallback_convert_accent = shift;
 
   my $accent = $command->{'cmdname'};
 
+  my $result;
+
   # special handling of @dotless{i}.
   # \x{0131}\x{0308} for @dotless{i} @" doesn't lead to NFC 00ef.
   # so it is set to a real dotless i only if not in an accent command.
@@ -819,15 +820,19 @@
                          or !$command->{'parent'}->{'parent'}
                          or !$command->{'parent'}->{'parent'}->{'cmdname'}
                          or 
!$unicode_accented_letters{$command->{'parent'}->{'parent'}->{'cmdname'}})) {
-      return "\x{0131}";
-    }
-    #return "\x{}" if ($text eq 'j'); # dotless j not known i unicode !
-    return $text;
+      $result = "\x{0131}";
+    } else {
+      $result = $text;
+    }
+    return $result;
   }
 
-  return Unicode::Normalize::NFC($text . 
chr(hex($unicode_diacritics{$accent})))
-    if (defined($unicode_diacritics{$accent}));
-  return &$fallback_convert_accent($text, $command);
+  if (defined($unicode_diacritics{$accent})) {
+    $result = Unicode::Normalize::NFC($text . 
chr(hex($unicode_diacritics{$accent})));
+    return $result;
+  } else {
+    return undef;
+  }
 }
 
 sub unicode_text($$$$)