CSS-Inliner

 view release on metacpan or  search on metacpan

lib/CSS/Inliner.pm  view on Meta::CPAN

  unless ($params && $$params{url}) {
    croak 'You must pass in hash params that contain a url argument';
  }

  # fetch and retrieve the remote content
  my ($content,$baseref,$ctcharset) = $self->_fetch_url({ url => $$params{url} });

  my $charset = $self->detect_charset({ content => $content, charset => $$params{charset}, ctcharset => $ctcharset });

  my $decoded_html;
  if ($charset) {
    $decoded_html = $self->decode_characters({ content => $content, charset => $charset });
  }
  else {
    # no good hints found, do the best we can

    if ($self->_fixlatin()) {
      Encoding::FixLatin->import('fix_latin');
      $decoded_html = fix_latin($content);
    }
    else {
      $decoded_html = $self->decode_characters({ content => $content, charset => 'ascii' });
    }
  }

  my $html = $self->_absolutize_references({ content => $decoded_html, baseref => $baseref });

  $self->read({ html => $html, charset => $charset });

  return();
}

=head2 read_file

Opens and reads an HTML file that supposedly contains both HTML and a
style declaration, properly tags the data with the proper charset

lib/CSS/Inliner.pm  view on Meta::CPAN

  unless ($params && $$params{filename}) {
    croak "You must pass in hash params that contain a filename argument";
  }

  open FILE, "<", $$params{filename} or die $!;
  my $content = do { local( $/ ) ; <FILE> };

  my $charset = $self->detect_charset({ content => $content, charset => $$params{charset} });

  my $decoded_html;
  if ($charset) {
    $decoded_html = $self->decode_characters({ content => $content, charset => $charset });
  }
  else {
    # no good hints found, do the best we can

    if ($self->_fixlatin()) {
      Encoding::FixLatin->import('fix_latin');
      $decoded_html = fix_latin($content);
    }
    else {
      $decoded_html = $self->decode_characters({ content => $content, charset => 'ascii' });
    }
  }

  $self->read({ html => $decoded_html, charset => $charset });

  return();
}

=head2 read

Reads passed html data and parses it.  The intermediate data is stored in
class variables.

The <style> block is ripped out of the html here, and stored

lib/CSS/Inliner.pm  view on Meta::CPAN

    croak "You must pass content for content character decoding";
  }

  unless ($params && $$params{charset}) {
    croak "You must pass the charset type of the content to decode";
  }

  my $content = $$params{content};
  my $charset = $$params{charset};

  my $decoded_html;
  eval {
    $decoded_html = decode($charset,$content);
  };

  if (!$decoded_html) {
    croak('Error decoding content with character set "'.$$params{charset}.'"');
  }

  return $decoded_html;
}

=head2 inlinify

Processes the html data that was entered through either 'read' or
'read_file', returns a scalar that contains a composite chunk of html
that has inline styles instead of a top level <style> declaration.

=cut

lib/CSS/Inliner.pm  view on Meta::CPAN

    }
  }
}

sub _extract_meta_charset {
  my ($self,$params) = @_;

  $self->_check_object();

  # we are going to parse an html document as ascii that is not necessarily ascii - silence the warning
  local $SIG{__WARN__} = sub { my $warning = shift; warn $warning unless $warning =~ /^Parsing of undecoded UTF-8/ };

  # parse document and pull out key header elements
  my $extract_tree = new CSS::Inliner::TreeBuilder();
  $self->_configure_tree({ tree => $extract_tree });

  $extract_tree->parse_content($$params{content});

  my $head = $extract_tree->look_down("_tag", "head"); # there should only be one

  my $meta_charset;