view release on metacpan or  search on metacpan

lib/Pod/Simple.pm  view on Meta::CPAN

have the POD FORMATTING ERRORS section note the localtime, and the
version of Pod::Simple.

option to delete all E<shy>s?
option to scream if under-0x20 literals are found in the input, or
under-E<32> E codes are found in the tree. And ditto \x7f-\x9f

Option to turn highbit characters into their compromised form? (applies
to E parsing too)

TODO: BOM/encoding things.

TODO: ascii-compat things in the XML classes?

lib/Pod/Simple/BlackBox.pm  view on Meta::CPAN

# If this perl doesn't have the Deprecated property, there's only one code
# point in it that we need be concerned with.
my $deprecated_re = my_qr('\p{IsDeprecated}', "\x{149}");
$deprecated_re = qr/\x{149}/ unless $deprecated_re;

my $utf8_bom;
if ( do { no integer; "$]" >= 5.007_003 }) {
  $utf8_bom = "\x{FEFF}";
  utf8::encode($utf8_bom);
} else {
  $utf8_bom = "\xEF\xBB\xBF";   # No EBCDIC BOM detection for early Perls.
}

# This is used so that the 'content_seen' method doesn't return true on a
# file that just happens to have a line that matches /^=[a-zA-z]/.  Only if
# there is a valid =foo line will we return that content was seen.
my $seen_legal_directive = 0;

#@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

sub parse_line { shift->parse_lines(@_) } # alias

lib/Pod/Simple/BlackBox.pm  view on Meta::CPAN

    if( $self->{'line_count'}++ ) {
      ($line = $source_line) =~ tr/\n\r//d;
       # If we don't have two vars, we'll end up with that there
       # tr/// modding the (potentially read-only) original source line!

    } else {
      DEBUG > 2 and print STDERR "First line: [$source_line]\n";

      if( ($line = $source_line) =~ s/^$utf8_bom//s ) {
        DEBUG and print STDERR "UTF-8 BOM seen.  Faking a '=encoding utf8'.\n";
        $self->_handle_encoding_line( "=encoding utf8" );
        delete $self->{'_processed_encoding'};
        $line =~ tr/\n\r//d;

      } elsif( $line =~ s/^\xFE\xFF//s ) {
        DEBUG and print STDERR "Big-endian UTF-16 BOM seen.  Aborting parsing.\n";
        $self->scream(
          $self->{'line_count'},
          "UTF16-BE Byte Encoding Mark found; but Pod::Simple v$Pod::Simple::VERSION doesn't implement UTF16 yet."
        );
        splice @_;
        push @_, undef;
        next;

        # TODO: implement somehow?

      } elsif( $line =~ s/^\xFF\xFE//s ) {
        DEBUG and print STDERR "Little-endian UTF-16 BOM seen.  Aborting parsing.\n";
        $self->scream(
          $self->{'line_count'},
          "UTF16-LE Byte Encoding Mark found; but Pod::Simple v$Pod::Simple::VERSION doesn't implement UTF16 yet."
        );
        splice @_;
        push @_, undef;
        next;

        # TODO: implement somehow?

      } else {
        DEBUG > 2 and print STDERR "First line is BOM-less.\n";
        ($line = $source_line) =~ tr/\n\r//d;
      }
    }

    if(!$self->{'parse_characters'} && !$self->{'encoding'}
      && ($self->{'in_pod'} || $line =~ /^=/s)
      && $line =~ /$non_ascii_re/
    ) {

      my $encoding;