view release on metacpan or  search on metacpan

lib/LaTeXML/Post/UnicodeMath.pm  view on Meta::CPAN

  UTF(0x60)  => "\x{0300}",               # \grave
  "\x{02D9}" => "\x{0307}",               # \dot
  UTF(0xAB)  => "\x{0308}",               # \ddot
  UTF(0xAF)  => "\x{0304}",               # \bar, \overline
  "\x{2192}" => "\x{20D7}",               # \vec
  "\x{02D8}" => "\x{0306}",               # \breve
  "o"        => "\x{030A}",               # \r
  "\x{02DD}" => "\x{030B}",               # \H
);
our %underaccents = (
  UTF(0xB8) => "\x{0327}",                # \c
  '.'       => "\x{0323}",                # dot below
  UTF(0xAF) => "\x{0331}",                # macron below
  "="       => "\x{0361}",                # \t
  ","       => "\x{0361}",                # lfhook
);
# \overbrace, \underbrace, \overleftarrow ??
sub unimath_overaccent {
  my ($op, $base) = @_;
  my $acc  = $op->textContent;
  my $cacc = $acc && $overaccents{$acc};
  my ($ubase, $uprec) = unimath_internal($base);
  $ubase = '(' . $ubase . ')' if length($ubase) > 1;
  return ($ubase . ($cacc ? $cacc : "\x{252C}" . $acc), $PREC_SCRIPTOP); }

sub unimath_underaccent {
  my ($op, $base) = @_;
  my $acc  = $op->textContent;
  my $cacc = $acc && $underaccents{$acc};
  my ($ubase, $uprec) = unimath_internal($base);
  $ubase = '(' . $ubase . ')' if length($ubase) > 1;    # maybe NBSP, too???
  return ($ubase . ($cacc ? $cacc : "\x{252C}" . $acc), $PREC_SCRIPTOP); }

# Handle text contents.
# We probably should pass this back to the same code used in CrossRef.
# But also, there's the question of what to do with nested math?
# Is that even allowable in UnicodeMath?
sub unimath_text {
  my (@nodes) = @_;
  return ('"' . join('', map { (ref $_ ? $_->textContent : $_); } @nodes) . '"', $PREC_SYMBOL); }

sub unimath_error {
  return unimath_text('ERROR ', @_); }

#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
# Tranlators
#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

DefUnicodeMath('Apply:?:?',             \&unimath_prefix);
DefUnicodeMath('Apply:ADDOP:?',         \&unimath_infix);
DefUnicodeMath('Apply:MULOP:?',         \&unimath_infix);
DefUnicodeMath('Apply:BINOP:?',         \&unimath_infix);
DefUnicodeMath('Apply:RELOP:?',         \&unimath_infix);
DefUnicodeMath('Apply:METARELOP:?',     \&unimath_infix);
DefUnicodeMath('Apply:ARROW:?',         \&unimath_infix);
DefUnicodeMath('Apply:COMPOSEOP:?',     \&unimath_infix);
DefUnicodeMath("Apply:DIFFOP:?",        \&unimath_prefix);
DefUnicodeMath('Apply:BIGOP:?',         \&unimath_prefix);
DefUnicodeMath('Apply:INTOP:?',         \&unimath_prefix);
DefUnicodeMath('Apply:SUMOP:?',         \&unimath_prefix);
DefUnicodeMath('Apply:?:formulae',      \&unimath_map);
DefUnicodeMath('Apply:?:multirelation', \&unimath_args);
DefUnicodeMath('Apply:?:limit-from',    \&unimath_prefix);
DefUnicodeMath('Apply:?:annotated',     \&unimath_prefix);

DefUnicodeMath('Apply:FRACOP:?', sub {
    my ($op, $num, $den, @more) = @_;
    my $thickness = $op->getAttribute('thickness');
    if (defined $thickness) {    # Hmm? maybe not even a fraction?
      return ('(' . unimath_nested($num, 0) . UTF(0xA6) . unimath_nested($den, 0) . ')', $PREC_SYMBOL); }
    else {
      return (unimath_nested($num, $PREC_MULOP) . '/' . unimath_nested($den, $PREC_MULOP), 1); } });

DefUnicodeMath('Apply:MODIFIEROP:?',    \&unimath_infix);
DefUnicodeMath('Apply:MIDDLE:?',        \&unimath_infix);
DefUnicodeMath('Apply:SUPERSCRIPTOP:?', \&unimath_sup);
DefUnicodeMath('Apply:SUBSCRIPTOP:?',   \&unimath_sub);
# These could search for candidate combining chars?
DefUnicodeMath('Apply:OVERACCENT:?',  \&unimath_overaccent);
DefUnicodeMath('Apply:UNDERACCENT:?', \&unimath_underaccent);

DefUnicodeMath('Apply:POSTFIX:?', sub {    # Reverse presentation, no @apply
    my ($op) = unimath_internal($_[0]);
    return (unimath_nested($_[1], $PREC_MULOP) . $op, $PREC_MULOP); });

DefUnicodeMath('Apply:?:square-root', sub {
    return ("\x{221A}" . unimath_nested($_[1], $PREC_MULOP), $PREC_MULOP); });
DefUnicodeMath('Apply:?:nth-root', sub {
    # Could convert to \x{221B} for cube, \x{221C} for quartic
    my ($n) = unimath_internal($_[2]);
    my $op = ($n eq '2' ? "\x{221A}"
      : ($n eq '3' ? "\x{221B}"
        : ($n eq '4' ? "\x{221C}"
          : "\\root " . $n . "\\of")));
    return ($op . unimath_nested($_[1], $PREC_MULOP), $PREC_MULOP); });
DefUnicodeMath('Apply:ENCLOSE:?', sub {
    my ($op, $base) = @_;
    return (unimath_nested($base, $PREC_SYMBOL), $PREC_SYMBOL); });

# ================================================================================
# cfrac! Ugh!
DefUnicodeMath('Apply:?:continued-fraction', sub {
    return unimath_error("continued fraction"); });

#================================================================================
1;

__END__

=pod

=head1 NAME

C<LaTeXML::Post::UnicodeMath> - Post-Processing module for converting math to UnicodeMath.

=head1 SYNOPSIS

C<LaTeXML::Post::UnicodeMath> converts math into UnicodeMath.
It should be usable as a primary math format (alone), or as a secondary format.
This module can also be used to convert ltx:XMath expressions into plain Unicode strings
for use in attributes.