[BACK]Return to post_xml.pm CVS log [TXT] [DIR] Up to [LON-CAPA] / loncom / homework / cleanxml

Diff for /loncom/homework/cleanxml/post_xml.pm between versions 1.6 and 1.7

version 1.6, 2016/01/08 20:32:42 version 1.7, 2016/01/14 16:28:06
Line 41  use Cwd 'abs_path'; Line 41  use Cwd 'abs_path';
 use XML::LibXML;  use XML::LibXML;
 use HTML::TokeParser; # used to parse sty files  use HTML::TokeParser; # used to parse sty files
 use Tie::IxHash; # for ordered hashes  use Tie::IxHash; # for ordered hashes
   use tth;
   use Apache::html_to_xml;
   
 no warnings 'recursion'; # yes, fix_paragraph is using heavy recursion, I know  no warnings 'recursion'; # yes, fix_paragraph is using heavy recursion, I know
   
Line 552  sub replace_m { Line 554  sub replace_m {
 # Returns the HTML equivalent of LaTeX input, using tth  # Returns the HTML equivalent of LaTeX input, using tth
 sub tth {  sub tth {
   my ($text) = @_;    my ($text) = @_;
   my ($fh, $tmp_path) = tempfile();    my $output = &tth::tth($text);
   binmode($fh, ':utf8');    my $errorstring = &tth::ttherror();
   print $fh $text;    if ($errorstring) {
   close $fh;      die $errorstring;
   my $output = `tth -r -w2 -u -y0 < $tmp_path 2>/dev/null`;    }
   # hopefully the temp file will not be removed before this point (otherwise we should use unlink_on_destroy 0)    # hopefully the temp file will not be removed before this point (otherwise we should use unlink_on_destroy 0)
   $output =~ s/^\s*|\s*$//;    $output =~ s/^\s*|\s*$//;
   $output =~ s/<div class="p"><!----><\/div>/<br\/>/; # why is tth using such ugly markup for \newline ?    $output =~ s/<div class="p"><!----><\/div>/<br\/>/; # why is tth using such ugly markup for \newline ?
Line 567  sub tth { Line 569  sub tth {
 sub html_to_dom {  sub html_to_dom {
   my ($text) = @_;    my ($text) = @_;
   $text = '<root>'.$text.'</root>';    $text = '<root>'.$text.'</root>';
   my $textref = html_to_xml::html_to_xml(\$text);    my $textref = Apache::html_to_xml::html_to_xml(\$text);
   utf8::upgrade($$textref); # otherwise the XML parser fails when the HTML parser turns &nbsp; into a character    utf8::upgrade($$textref); # otherwise the XML parser fails when the HTML parser turns &nbsp; into a character
   my $dom_doc = XML::LibXML->load_xml(string => $textref);    my $dom_doc = XML::LibXML->load_xml(string => $textref);
   my $root = $dom_doc->documentElement;    my $root = $dom_doc->documentElement;
Removed from v.1.6  
changed lines
  Added in v.1.7

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>