Diff for /loncom/homework/cleanxml/post_xml.pm between versions 1.4 and 1.11

version 1.4, 2015/12/18 22:07:42 version 1.11, 2016/11/10 21:53:56
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 50  my @inline_like_block = ('stringresponse Line 52  my @inline_like_block = ('stringresponse
 my @responses = ('stringresponse','optionresponse','numericalresponse','formularesponse','mathresponse','organicresponse','reactionresponse','customresponse','externalresponse','essayresponse','radiobuttonresponse','matchresponse','rankresponse','imageresponse','functionplotresponse');  my @responses = ('stringresponse','optionresponse','numericalresponse','formularesponse','mathresponse','organicresponse','reactionresponse','customresponse','externalresponse','essayresponse','radiobuttonresponse','matchresponse','rankresponse','imageresponse','functionplotresponse');
 my @block_html = ('html','head','body','section','h1','h2','h3','h4','h5','h6','div','p','ul','ol','li','table','tbody','tr','td','th','dl','dt','dd','pre','noscript','hr','address','blockquote','object','applet','embed','map','form','fieldset','iframe','center','frameset');  my @block_html = ('html','head','body','section','h1','h2','h3','h4','h5','h6','div','p','ul','ol','li','table','tbody','tr','td','th','dl','dt','dd','pre','noscript','hr','address','blockquote','object','applet','embed','map','form','fieldset','iframe','center','frameset');
 my @no_newline_inside = ('import','parserlib','scriptlib','data','function','label','xlabel','ylabel','tic','text','rectangle','image','title','h1','h2','h3','h4','h5','h6','li','td','p');  my @no_newline_inside = ('import','parserlib','scriptlib','data','function','label','xlabel','ylabel','tic','text','rectangle','image','title','h1','h2','h3','h4','h5','h6','li','td','p');
 my @preserve_elements = ('script','answer','pre');  my @preserve_elements = ('script','answer','pre','style');
 my @accepting_style = ('section','h1','h2','h3','h4','h5','h6','div','p','li','td','th','dt','dd','pre','blockquote');  my @accepting_style = ('section','h1','h2','h3','h4','h5','h6','div','p','li','td','th','dt','dd','pre','blockquote');
 my @latex_math = ('\alpha', '\theta', '\omicron', '\tau', '\beta', '\vartheta', '\pi', '\upsilon', '\gamma', '\gamma', '\varpi', '\phi', '\delta', '\kappa', '\rho', '\varphi', '\epsilon', '\lambda', '\varrho', '\chi', '\varepsilon', '\mu', '\sigma', '\psi', '\zeta', '\nu', '\varsigma', '\omega', '\eta', '\xi',  my @latex_math = ('\alpha', '\theta', '\omicron', '\tau', '\beta', '\vartheta', '\pi', '\upsilon', '\gamma', '\gamma', '\varpi', '\phi', '\delta', '\kappa', '\rho', '\varphi', '\epsilon', '\lambda', '\varrho', '\chi', '\varepsilon', '\mu', '\sigma', '\psi', '\zeta', '\nu', '\varsigma', '\omega', '\eta', '\xi',
   '\Gamma', '\Lambda', '\Sigma', '\Psi', '\Delta', '\Xi', '\Upsilon', '\Omega', '\Theta', '\Pi', '\Phi',    '\Gamma', '\Lambda', '\Sigma', '\Psi', '\Delta', '\Xi', '\Upsilon', '\Omega', '\Theta', '\Pi', '\Phi',
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 150  sub fix_structure { Line 156  sub fix_structure {
   # the root element has already been added in pre_xml    # the root element has already been added in pre_xml
   my $root = $doc->documentElement;    my $root = $doc->documentElement;
   # inside the root, replace html, problem and library elements by their content    # inside the root, replace html, problem and library elements by their content
   my @toreplace = ('html','problem','library');    my @toreplace = ('html','problem','library','Task');
   foreach my $name (@toreplace) {    foreach my $name (@toreplace) {
     my @elements = $root->getElementsByTagName($name);      my @elements = $root->getElementsByTagName($name);
     foreach my $element (@elements) {      foreach my $element (@elements) {
Line 552  sub replace_m { Line 558  sub replace_m {
 # Returns the HTML equivalent of LaTeX input, using tth  # Returns the HTML equivalent of LaTeX input, using tth
 sub tth {  sub tth {
   my ($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 ?    $output =~ s/<div class="p"><!----><\/div>/<br\/>/; # why is tth using such ugly markup for \newline ?
Line 567  sub tth { Line 573  sub tth {
 sub html_to_dom {  sub html_to_dom {
   my ($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 1814  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','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 1858  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 2046  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 2246  sub fix_empty_lc_elements { Line 2297  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 {
   my ($root) = @_;    my ($root) = @_;
Line 2394  sub pretty { Line 2472  sub pretty {
   my $type = $node->nodeType;    my $type = $node->nodeType;
   if ($type == XML_ELEMENT_NODE) {    if ($type == XML_ELEMENT_NODE) {
     my $name = $node->nodeName;      my $name = $node->nodeName;
     if ((string_in_array($all_block, $name) || string_in_array(\@inline_like_block, $name)) &&      if (string_in_array(\@preserve_elements, $name)) {
         !string_in_array(\@preserve_elements, $name)) {        # remove newlines at the beginning and the end of preserve elements
         if (defined $node->firstChild && ($node->firstChild->nodeType == XML_TEXT_NODE ||
             $node->firstChild->nodeType == XML_CDATA_SECTION_NODE)) {
           my $text = $node->firstChild->nodeValue;
           $text =~ s/^\n+//;
           if ($text eq '') {
             $node->removeChild($node->firstChild);
           } else {
             $node->firstChild->setData($text);
           }
         }
         if (defined $node->lastChild && ($node->lastChild->nodeType == XML_TEXT_NODE ||
             $node->lastChild->nodeType == XML_CDATA_SECTION_NODE)) {
           my $text = $node->lastChild->nodeValue;
           $text =~ s/\n+$//;
           if ($text eq '') {
             $node->removeChild($node->lastChild);
           } else {
             $node->lastChild->setData($text);
           }
         }
       } elsif (string_in_array($all_block, $name) || string_in_array(\@inline_like_block, $name)) {
       # make sure there is a newline at the beginning and at the end if there is anything inside        # make sure there is a newline at the beginning and at the end if there is anything inside
       if (defined $node->firstChild && !string_in_array(\@no_newline_inside, $name)) {        if (defined $node->firstChild && !string_in_array(\@no_newline_inside, $name)) {
         my $first = $node->firstChild;          my $first = $node->firstChild;
Line 2478  sub pretty { Line 2577  sub pretty {
         if ($text eq '') {          if ($text eq '') {
           $node->removeChild($node->lastChild);            $node->removeChild($node->lastChild);
         } else {          } else {
           $node->lastChild->setData($text);  
         }  
       }  
     } elsif (string_in_array(\@preserve_elements, $name)) {  
       # collapse newlines at the beginning and the end of scripts  
       if (defined $node->firstChild && $node->firstChild->nodeType == XML_TEXT_NODE) {  
         my $text = $node->firstChild->nodeValue;  
         $text =~ s/^\n( *\n)+/\n/;  
         if ($text eq '') {  
           $node->removeChild($node->firstChild);  
         } else {  
           $node->firstChild->setData($text);  
         }  
       }  
       if (defined $node->lastChild && $node->lastChild->nodeType == XML_TEXT_NODE) {  
         my $text = $node->lastChild->nodeValue;  
         $text =~ s/\n( *\n)+$/\n/;  
         if ($text eq '') {  
           $node->removeChild($node->lastChild);  
         } else {  
           $node->lastChild->setData($text);            $node->lastChild->setData($text);
         }          }
       }        }

Removed from v.1.4  
changed lines
  Added in v.1.11


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