Base primitive improvements

lib/LaTeXML/Common/Font.pm

@@ -22,7 +22,7 @@ use LaTeXML::Common::Font::Metric;
 use LaTeXML::Common::Font::StandardMetrics;
 use LaTeXML::Common::Color;
 use List::Util qw(min max sum);
-use base qw(LaTeXML::Common::Object);
+use base       qw(LaTeXML::Common::Object);
 
 # Note that this has evolved way beynond just "font",
 # but covers text properties (or even display properties) in general
@@ -62,7 +62,7 @@ my $FLAG_EMPH         = 0x10;
 my %font_family = (
   cmr   => { family => 'serif' },      cmss  => { family => 'sansserif' },
   cmtt  => { family => 'typewriter' }, cmvtt => { family => 'typewriter' },
-  cmti  => { family => 'typewriter', shape => 'italic' },
+  cmt   => { family => 'serif' },    # for cmti "text italic"
   cmfib => { family => 'serif' },      cmfr  => { family => 'serif' },
   cmdh  => { family => 'serif' },      cm    => { family => 'serif' },
   ptm   => { family => 'serif' },      ppl   => { family => 'serif' },

lib/LaTeXML/Core/Document.pm

@@ -21,7 +21,7 @@ use LaTeXML::Common::XML;
 use LaTeXML::Util::Radix;
 use Unicode::Normalize;
 use Scalar::Util qw(blessed);
-use base qw(LaTeXML::Common::Object);
+use base         qw(LaTeXML::Common::Object);
 
 #**********************************************************************
 # These two element names are `leaks' of the document structure into
@@ -250,6 +250,7 @@ sub canAutoClose {
     || (($t == XML_ELEMENT_NODE)                              # otherwise must be element
     && !$node->getAttribute('_noautoclose')                   # without _noautoclose
     && ($node->getAttribute('_autoclose')                     # and either with _autoclose
+      || $node->getAttribute('_autoopened')                   # or was autoopened
                                                               # OR it has autoClose set on tag properties
       || (($props = $STATE->lookupMapping('TAG_PROPERTIES', getNodeQName($self, $node)))
         && $$props{autoClose})));

lib/LaTeXML/Core/State.pm

@@ -123,6 +123,17 @@ sub new {
   $$self{uccode}              = {};
   $$self{delcode}             = {};
   $$self{tracing_definitions} = {};
+  # Initializations that INITEX would have set.
+  $$self{mathcode}{'.'} = [0];
+  for (my $c = ord('0') ; $c <= ord('9') ; $c++) {
+    $$self{mathcode}{ chr($c) } = [0x7000]; }
+  for (my $c = ord('a') ; $c <= ord('z') ; $c++) {
+    my $C = $c + ord('A') - ord('a');
+    $$self{mathcode}{ chr($c) } = [0x7100];
+    $$self{mathcode}{ chr($C) } = [0x7100];
+    $$self{uccode}{ chr($c) }   = [$C];
+    $$self{lccode}{ chr($C) }   = [$c];
+    $$self{sfcode}{ chr($C) }   = [999]; }
   return $self; }
 
 sub assign_internal {

lib/LaTeXML/Engine/Base_ParameterTypes.pool.ltxml

@@ -68,9 +68,13 @@ DefParameterType('Optional', sub {
 DefParameterType('GeneralText', sub {
     my ($gullet) = @_;
     $gullet->unread($gullet->readXToken);    # Force expansion to skip <filler> before required {
-
     return $gullet->readBalanced(0, 0, 1); });
 
+DefParameterType('XGeneralText', sub {
+    my ($gullet) = @_;
+    $gullet->unread($gullet->readXToken);    # Force expansion to skip <filler> before required {
+    return $gullet->readBalanced(1, 0, 1); });
+
 DefParameterType('Until', sub {
     my ($gullet, $until) = @_;
     $gullet->readUntil($until); },
@@ -371,26 +375,38 @@ DefParameterType('BalancedParen', sub {
 # It is useful when the content would usually need to have been \protect'd
 # in order to correctly deal with catcodes.
 # BEWARE: This is NOT a shorthand for a simple digested {}!
-DefParameterType('Digested', sub {
-    no warnings 'recursion';
-    my ($gullet) = @_;
-    $gullet->skipSpaces;
-    my $ismath = $STATE->lookupValue('IN_MATH');
-    my @list   = ();
-    my $token;
-    do { $token = $gullet->readXToken(0);
-    } while (defined $token && (($token->getCatcode == CC_SPACE) || $token->equals(T_CS('\relax'))));
-    if    (!defined $token) { }
-    elsif ($token->getCatcode == CC_BEGIN) {
-      Digest($token);
-      push(@list, $STATE->getStomach->digestNextBody()); pop(@list); }    # content w/o the braces
-    else {
-      push(@list, $STATE->getStomach->invokeToken($token)); }
-    @list = grep { ref $_ ne 'LaTeXML::Core::Comment' } @list;
-    List(@list, mode => ($ismath ? 'math' : 'text')); },
+sub readDigested {
+  no warnings 'recursion';
+  my ($gullet) = @_;
+  $gullet->skipSpaces;
+  my $ismath = $STATE->lookupValue('IN_MATH');
+  my @list   = ();
+  my $token;
+  do { $token = $gullet->readXToken(0);
+  } while (defined $token && (($token->getCatcode == CC_SPACE) || $token->equals(T_CS('\relax'))));
+  if    (!defined $token) { }
+  elsif ($token->getCatcode == CC_BEGIN) {
+    Digest($token);
+    push(@list, $STATE->getStomach->digestNextBody()); pop(@list); }    # content w/o the braces
+  else {
+    push(@list, $STATE->getStomach->invokeToken($token)); }
+  @list = grep { ref $_ ne 'LaTeXML::Core::Comment' } @list;
+  return List(@list, mode => ($ismath ? 'math' : 'text')); }
+
+DefParameterType('Digested', \&readDigested,
   undigested => 1,                                           # since _already_ digested.
   reversion  => sub { (T_BEGIN, Revert($_[0]), T_END); });
 
+# Read a Delimiter;
+# Formally a delimiter is either a token, or \delimiter<number> or maybe \radical<number>,
+# but we don't actually restrict to those.
+# Here, we just read a single Digested thing, but avoid reversion braces if possible.
+DefParameterType('TeXDelimiter', \&readDigested,
+  undigested => 1,      # since _already_ digested.
+  reversion  => sub {
+    my @arg = Revert($_[0]);
+    (scalar(@arg) == 1 ? $arg[0] : (T_BEGIN, @arg, T_END)); });
+
 # A variation: Digest until we encounter a given token!
 DefParameterType('DigestUntil', sub {
     my ($gullet, $until) = @_;

lib/LaTeXML/Engine/TeX_FileIO.pool.ltxml

@@ -106,18 +106,23 @@ DefPrimitive('\openout Number SkipSpaces SkipMatch:= SkipSpaces TeXFileName', su
 DefPrimitive('\closeout Number', sub {
     my ($stomach, $port) = @_;
     $port = ToString($port);
+    if ($LaTeXML::DEBUG{write}) {
+      if (my $filename = LookupValue('output_file:' . $port)) {
+        my $handle   = $filename . '_contents';
+        my $contents = LookupValue($handle);
+        Debug("CLOSING $filename with content:\n$contents\n============================="); } }
     AssignValue('output_file:' . $port => undef, 'global');
     return; });
 
-DefPrimitive('\write Number {}', sub {
+DefPrimitive('\write Number XGeneralText', sub {
     my ($stomach, $port, $tokens) = @_;
     $port = ToString($port);
     if (my $filename = LookupValue('output_file:' . $port)) {
       my $handle   = $filename . '_contents';
       my $contents = LookupValue($handle);
-      AssignValue($handle => $contents . UnTeX(Expand($tokens), 1) . "\n", 'global'); }
+      AssignValue($handle => $contents . UnTeX($tokens, 1) . "\n", 'global'); }
     else {
-      Note(UnTeX(Expand($tokens))); }
+      Note(UnTeX($tokens)); }
     return; });
 
 # Since we don't paginate, we're effectively always "shipping out",
@@ -145,7 +150,7 @@ DefMacro('\input TeXFileName', sub {
 #----------------------------------------------------------------------
 # \special          c  sends material to the dvi file for special processing.
 
-DefPrimitive('\special {}', sub {
+DefPrimitive('\special XGeneralText', sub {
     my ($stomach, $arg) = @_;
     my $special_str = ToString($arg);
     # recognize one special graphics inclusion case
@@ -161,7 +166,7 @@ DefPrimitive('\special {}', sub {
       $stomach->getGullet->unread(
         T_CS('\ltx@special@graphics'), @kv, T_BEGIN, T_OTHER($graphic), T_END); }
     else {
-      Info('ignored', 'special', $stomach, 'Unrecognized TeX Special', $arg); }
+      Info('ignored', 'special', $stomach, 'Unrecognized TeX Special' . ToString($arg)); }
     return; });
 
 # adapted from graphicx.sty.ltxml