[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 1.12

version 1.6, 2016/01/08 20:32:42 version 1.12, 2017/01/17 20:29:06
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 126  sub post_xml { Line 128  sub post_xml {
       
   remove_useless_notsolved($root);    remove_useless_notsolved($root);
       
     fix_comments($root);
     
   fix_paragraphs_inside($root, \@all_block);    fix_paragraphs_inside($root, \@all_block);
   
   remove_empty_style($root);    remove_empty_style($root);
       
   fix_empty_lc_elements($root);    fix_empty_lc_elements($root);
       
     reduce_empty_p($root);
     
   lowercase_attribute_values($root);    lowercase_attribute_values($root);
       
   replace_numericalresponse_unit_attribute($root);    replace_numericalresponse_unit_attribute($root);
Line 424  sub replace_m { Line 430  sub replace_m {
     my $var_key1 = 'dfhg3df54hg65hg4';      my $var_key1 = 'dfhg3df54hg65hg4';
     my $var_key2 = 'dfhg654d6f5g4h5f';      my $var_key2 = 'dfhg654d6f5g4h5f';
     my $eval = defined $m->getAttribute('eval') && $m->getAttribute('eval') eq 'on';      my $eval = defined $m->getAttribute('eval') && $m->getAttribute('eval') eq 'on';
       my $display = $m->getAttribute('display');
       if (defined $display) {
           if ($display eq '') {
               $display = undef;
           }
           if (lc($display) eq 'jsmath') {
               $display = 'mathjax';
           }
       }
     if ($eval) {      if ($eval) {
       # replace variables        # replace variables
       foreach my $variable (@variables) {        foreach my $variable (@variables) {
Line 461  sub replace_m { Line 476  sub replace_m {
       if ($eval) {        if ($eval) {
         $new_node->setAttribute('eval', 'on');          $new_node->setAttribute('eval', 'on');
       }        }
         if (defined $display) {
           $new_node->setAttribute('display', $display);
         }
       $new_node->appendChild($doc->createTextNode($new_text));        $new_node->appendChild($doc->createTextNode($new_text));
       $m->parentNode->replaceChild($new_node, $m);        $m->parentNode->replaceChild($new_node, $m);
       next;        next;
Line 491  sub replace_m { Line 509  sub replace_m {
     # there are math separators inside, even after hiding variables, or there was no math symbol      # there are math separators inside, even after hiding variables, or there was no math symbol
           
     # hide math parts inside before running tth      # hide math parts inside before running tth
     my $math_key1 = '#ghjgdh5hg45gf';      my $math_key1 = '#5752398247516385';
     my $math_key2 = '#';      my $math_key2 = '#';
     my @maths = ();      my @maths = ();
     my @separators = (['$$','$$'], ['\\(','\\)'], ['\\[','\\]'], ['$','$']);      my @separators = (['$$','$$'], ['\\(','\\)'], ['\\[','\\]'], ['$','$']);
Line 522  sub replace_m { Line 540  sub replace_m {
       $math =~ s/&/&/g;        $math =~ s/&/&/g;
       $math =~ s/</&lt;/g;        $math =~ s/</&lt;/g;
       $math =~ s/>/&gt;/g;        $math =~ s/>/&gt;/g;
         my ($mel, $inside);
       if ($math =~ /^\$\$(.*)\$\$$/s) {        if ($math =~ /^\$\$(.*)\$\$$/s) {
         $math = '<dtm>'.$1.'</dtm>';          $mel = 'dtm';
           $inside = $1;
       } elsif ($math =~ /^\\\[(.*)\\\]$/s) {        } elsif ($math =~ /^\\\[(.*)\\\]$/s) {
         $math = '<dtm>'.$1.'</dtm>';          $mel = 'dtm';
           $inside = $1;
       } elsif ($math =~ /^\\\((.*)\\\)$/s) {        } elsif ($math =~ /^\\\((.*)\\\)$/s) {
         $math = '<tm>'.$1.'</tm>';          $mel = 'tm';
           $inside = $1;
       } elsif ($math =~ /^\$(.*)\$$/s) {        } elsif ($math =~ /^\$(.*)\$$/s) {
         $math = '<tm>'.$1.'</tm>';          $mel = 'tm';
           $inside = $1;
         }
         if (defined $inside) {
           if ($inside =~ /^\s*$/) {
             $math = '';
           } else {
             $math = '<'.$mel;
             if ($eval && $inside =~ /$var_key1/) {
               $math .= ' eval="on"';
             }
             $math .= '>'.$inside.'</'.$mel.'>';
           }
       }        }
       my $replace = $math_key1.($i+1).$math_key2;        my $replace = $math_key1.($i+1).$math_key2;
       $html_text =~ s/$replace/$math/;        $html_text =~ s/$replace/$math/;
Line 552  sub replace_m { Line 586  sub replace_m {
 # Returns the HTML equivalent of LaTeX input, using tth  # Returns the HTML equivalent of LaTeX input, using tth
 sub tth {  sub tth {
   my ($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 ?    $output =~ s/<div class="p"><!----><\/div>/<br\/>/; # why is tth using such ugly markup for \newline ?
Line 567  sub tth { Line 601  sub tth {
 sub html_to_dom {  sub html_to_dom {
   my ($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;
Line 1808  sub remove_useless_notsolved { Line 1842  sub remove_useless_notsolved {
   }    }
 }  }
   
   # Use <pre> for multi-line comments without elements.
   sub fix_comments {
     my ($root) = @_;
     my $doc = $root->ownerDocument;
     my @comments = $root->getElementsByTagName('comment');
     foreach my $comment (@comments) {
       my $first = $comment->firstChild;
       if (defined $first) {
         if ($first->nodeType == XML_TEXT_NODE && $first->nodeValue =~ /\n/ &&
             !defined $first->nextSibling) {
           my $pre = $doc->createElement('pre');
           $comment->removeChild($first);
           $comment->appendChild($pre);
           $pre->appendChild($first);
         }
       }
     }
   }
   
 # adds a paragraph inside if needed and calls fix_paragraph for all paragraphs (including new ones)  # adds a paragraph inside if needed and calls fix_paragraph for all paragraphs (including new ones)
 sub fix_paragraphs_inside {  sub fix_paragraphs_inside {
   my ($node, $all_block) = @_;    my ($node, $all_block) = @_;
   # blocks in which paragrahs will be added:    # blocks in which paragrahs will be added:
   my @blocks_with_p = ('loncapa','library','problem','part','problemtype','window','block','while','postanswerdate','preduedate','solved','notsolved','languageblock','instructorcomment','togglebox','standalone','body','form');    my @blocks_with_p = ('loncapa','library','problem','part','problemtype','window','block','while','postanswerdate','preduedate','languageblock','instructorcomment','togglebox','standalone','body','form');
   my @fix_p_if_br_or_p = (@responses,'foil','item','text','label','hintgroup','hintpart','hint','web','windowlink','div','li','dd','td','th','blockquote');    my @fix_p_if_br_or_p = (@responses,'foil','item','text','label','hintgroup','hintpart','hint','web','windowlink','div','li','dd','td','th','blockquote','solved','notsolved');
   if ((string_in_array(\@blocks_with_p, $node->nodeName) && paragraph_needed($node)) ||    if ((string_in_array(\@blocks_with_p, $node->nodeName) && paragraph_needed($node)) ||
       (string_in_array(\@fix_p_if_br_or_p, $node->nodeName) && paragraph_inside($node))) {        (string_in_array(\@fix_p_if_br_or_p, $node->nodeName) && paragraph_inside($node))) {
     # if non-empty, add paragraphs where needed between all br and remove br      # if non-empty, add paragraphs where needed between all br and remove br
Line 1833  sub fix_paragraphs_inside { Line 1886  sub fix_paragraphs_inside {
           push(@new_children, $doc->createElement('p'));            push(@new_children, $doc->createElement('p'));
         }          }
         $p = undef;          $p = undef;
           # ignore the next node if it is a br (the paragraph default margin will take as much space)
           # (ignoring whitespace)
           while (defined $next && $next->nodeType == XML_TEXT_NODE && $next->nodeValue =~ /^[ \t\f\n\r]*$/) {
             my $next2 = $next->nextSibling;
             $node->removeChild($next);
             $next = $next2;
           }
           if (defined $next && $next->nodeType == XML_ELEMENT_NODE && $next->nodeName eq 'br') {
             my $next2 = $next->nextSibling;
             $node->removeChild($next);
             $next = $next2;
           }
       } elsif ($child->nodeType == XML_ELEMENT_NODE && string_in_array(\@inline_like_block, $child->nodeName)) {        } elsif ($child->nodeType == XML_ELEMENT_NODE && string_in_array(\@inline_like_block, $child->nodeName)) {
         # inline_like_block: use the paragraph if there is one, otherwise do not create one          # inline_like_block: use the paragraph if there is one, otherwise do not create one
         if (defined $p) {          if (defined $p) {
Line 2009  sub fix_paragraph { Line 2074  sub fix_paragraph {
               if (!defined $left || !$left_needs_p) {                if (!defined $left || !$left_needs_p) {
                 $replacement->appendChild($middle);                  $replacement->appendChild($middle);
               }                }
                 # ignore the next node if it is a br (the paragraph default margin will take as much space)
                 my $first_right;
                 if (defined $right) {
                   $first_right = $right->firstChild;
                   # ignore non-nbsp whitespace
                   while (defined $first_right && $first_right->nodeType == XML_TEXT_NODE &&
                       $first_right->nodeValue =~ /^[ \t\f\n\r]*$/) {
                     $first_right = $first_right->nextSibling;
                   }
                 }
                 if (defined $first_right && $first_right->nodeType == XML_ELEMENT_NODE &&
                     $first_right->nodeName eq 'br') {
                   $right->removeChild($first_right);
                 }
             } else {              } else {
               fix_paragraphs_inside($n, $all_block);                fix_paragraphs_inside($n, $all_block);
               $replacement->appendChild($n);                $replacement->appendChild($n);
Line 2245  sub fix_empty_lc_elements { Line 2324  sub fix_empty_lc_elements {
     }      }
   }    }
 }  }
   
   # remove consecutive empty paragraphs (they will not show anyway)
   sub reduce_empty_p {
     my ($node) = @_;
     my $next;
     for (my $child=$node->firstChild; defined $child; $child=$next) {
       $next = $child->nextSibling;
       while (defined $next && $next->nodeType == XML_TEXT_NODE && $next->nodeValue =~ /^[ \t\f\n\r]*$/) {
         $next = $next->nextSibling;
       }
       if ($child->nodeType == XML_ELEMENT_NODE && $child->nodeName eq 'p' && defined $next &&
           $next->nodeType == XML_ELEMENT_NODE && $next->nodeName eq 'p') {
         my $first = $child->firstChild;
         if (!defined $first || (!defined $first->nextSibling &&
             $first->nodeType == XML_TEXT_NODE && $first->nodeValue =~ /^[ \t\f\n\r]*$/)) {
           $first = $next->firstChild;
           if (!defined $first || (!defined $first->nextSibling &&
               $first->nodeType == XML_TEXT_NODE && $first->nodeValue =~ /^[ \t\f\n\r]*$/)) {
             $node->removeChild($child);
           }
         }
       }
       if ($child->nodeType == XML_ELEMENT_NODE) {
         reduce_empty_p($child);
       }
     }
   }
   
 # turn some attribute values into lowercase when they should be  # turn some attribute values into lowercase when they should be
 sub lowercase_attribute_values {  sub lowercase_attribute_values {
Removed from v.1.6  
changed lines
  Added in v.1.12

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