Diff for /loncom/homework/cleanxml/post_xml.pm between versions 1.1 and 1.6

version 1.1, 2015/12/03 20:40:31 version 1.6, 2016/01/08 20:32:42
Line 42  use XML::LibXML; Line 42  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 Env qw(RES_DIR); # path of res directory parent (without the / at the end)  
   
 no warnings 'recursion'; # yes, fix_paragraph is using heavy recursion, I know  no warnings 'recursion'; # yes, fix_paragraph is using heavy recursion, I know
   
 # these are constants  # these are constants
 my @block_elements = ('parameter','location','answer','foil','image','polygon','rectangle','text','conceptgroup','itemgroup','item','label','data','function','array','unit','answergroup','functionplotresponse','functionplotruleset','functionplotelements','functionplotcustomrule','essayresponse','hintpart','formulahint','numericalhint','reactionhint','organichint','optionhint','radiobuttonhint','stringhint','customhint','mathhint','formulahintcondition','numericalhintcondition','reactionhintcondition','organichintcondition','optionhintcondition','radiobuttonhintcondition','stringhintcondition','customhintcondition','mathhintcondition','imageresponse','foilgroup','datasubmission','textfield','hiddensubmission','radiobuttonresponse','rankresponse','matchresponse','import','style','script','window','block','library','notsolved','part','postanswerdate','preduedate','problem','problemtype','randomlabel','bgimg','labelgroup','randomlist','solved','while','tex','print','web','gnuplot','curve','Task','IntroParagraph','ClosingParagraph','Question','QuestionText','Setup','Instance','InstanceText','Criteria','CriteriaText','GraderNote','languageblock','translated','lang','instructorcomment','dataresponse','togglebox','standalone','comment','drawimage','allow','displayduedate','displaytitle','responseparam','organicstructure','scriptlib','parserlib','drawoptionlist','spline','backgroundplot','plotobject','plotvector','drawvectorsum','functionplotrule','functionplotvectorrule','functionplotvectorsumrule','axis','key','xtics','ytics','title','xlabel','ylabel','hiddenline','dtm');  my @block_elements = ('parameter','location','answer','foil','image','polygon','rectangle','text','conceptgroup','itemgroup','item','label','data','function','array','unit','answergroup','functionplotresponse','functionplotruleset','functionplotelements','functionplotcustomrule','essayresponse','hintpart','formulahint','numericalhint','reactionhint','organichint','optionhint','radiobuttonhint','stringhint','customhint','mathhint','formulahintcondition','numericalhintcondition','reactionhintcondition','organichintcondition','optionhintcondition','radiobuttonhintcondition','stringhintcondition','customhintcondition','mathhintcondition','imageresponse','foilgroup','datasubmission','textfield','hiddensubmission','radiobuttonresponse','rankresponse','matchresponse','import','style','script','window','block','library','notsolved','part','postanswerdate','preduedate','problem','problemtype','randomlabel','bgimg','labelgroup','randomlist','solved','while','tex','print','web','gnuplot','curve','Task','IntroParagraph','ClosingParagraph','Question','QuestionText','Setup','Instance','InstanceText','Criteria','CriteriaText','GraderNote','languageblock','instructorcomment','dataresponse','togglebox','standalone','comment','drawimage','allow','displayduedate','displaytitle','responseparam','organicstructure','scriptlib','parserlib','drawoptionlist','spline','backgroundplot','plotobject','plotvector','drawvectorsum','functionplotrule','functionplotvectorrule','functionplotvectorsumrule','axis','key','xtics','ytics','title','xlabel','ylabel','hiddenline','dtm');
 my @inline_like_block = ('stringresponse','optionresponse','numericalresponse','formularesponse','mathresponse','organicresponse','reactionresponse','customresponse','externalresponse', 'hint', 'hintgroup'); # inline elements treated like blocks for pretty print and some other things  my @inline_like_block = ('stringresponse','optionresponse','numericalresponse','formularesponse','mathresponse','organicresponse','reactionresponse','customresponse','externalresponse', 'hint', 'hintgroup','translated','lang'); # inline elements treated like blocks for pretty print and some other things
 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 77  my $warnings; # 1 = print warnings Line 75  my $warnings; # 1 = print warnings
   
   
 # Parses the XML document and fixes many things to turn it into a document matching the schema.  # Parses the XML document and fixes many things to turn it into a document matching the schema.
 # Returns the text of the document as a byte string.  # @param {reference} textref - reference to the text of the document
   # @param {string} file_path - path of the file being processed (we only extract the directory path from it)
   # @param {string} res_dir - path of res directory parent (without the / at the end)
   # @param {boolean} warn - 1 to print warnings, 0 otherwise
   # @returns the text of the document as a byte string.
 sub post_xml {  sub post_xml {
   my ($textref, $file_path, $warn) = @_;    my ($textref, $file_path, $res_dir, $warn) = @_;
   $warnings = $warn;    $warnings = $warn;
       
   my $dom_doc = XML::LibXML->load_xml(string => $textref);    my $dom_doc = XML::LibXML->load_xml(string => $textref);
Line 92  sub post_xml { Line 94  sub post_xml {
       
   fix_attribute_case($root);    fix_attribute_case($root);
       
   my $fix_by_hand = replace_m($root);    replace_m($root);
       
   my @all_block = (@block_elements, @block_html);    my @all_block = (@block_elements, @block_html);
   add_sty_blocks($file_path, $root, \@all_block); # must come before the subs using @all_block    add_sty_blocks($file_path, $res_dir, $root, \@all_block); # must come before the subs using @all_block
   
   fix_block_styles($root, \@all_block);    fix_block_styles($root, \@all_block);
   $root->normalize();    $root->normalize();
Line 140  sub post_xml { Line 142  sub post_xml {
   
   replace_tm_dtm($root);    replace_tm_dtm($root);
       
   if ($fix_by_hand) {  
     die "The file has been converted but it should be fixed by hand.";  
   }  
   return $dom_doc->toString(); # byte string !    return $dom_doc->toString(); # byte string !
 }  }
   
Line 151  sub fix_structure { Line 150  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 $element (@elements) {
Line 353  sub fix_attribute_case { Line 352  sub fix_attribute_case {
 # Replaces m by HTML, tm and/or dtm (which will be replaced by <m> later, but they are useful  # Replaces m by HTML, tm and/or dtm (which will be replaced by <m> later, but they are useful
 #   to know if the element is a block element or not).  #   to know if the element is a block element or not).
 # m might contain non-math LaTeX, while tm and dtm may only contain math.  # m might contain non-math LaTeX, while tm and dtm may only contain math.
 # Returns 1 if the file should be fixed by hand, 0 otherwise.  
 sub replace_m {  sub replace_m {
   my ($root) = @_;    my ($root) = @_;
   my $doc = $root->ownerDocument;    my $doc = $root->ownerDocument;
   my $fix_by_hand = 0;  
   # search for variable declarations    # search for variable declarations
   my @variables = ();    my @variables = ();
   my @scripts = $root->getElementsByTagName('script');    my @scripts = $root->getElementsByTagName('script');
Line 406  sub replace_m { Line 403  sub replace_m {
           if ($warnings) {            if ($warnings) {
             print "WARNING: <m> is used in a script, it should be converted by hand\n";              print "WARNING: <m> is used in a script, it should be converted by hand\n";
           }            }
           $fix_by_hand = 1;  
         }          }
       }        }
     }      }
Line 421  sub replace_m { Line 417  sub replace_m {
       if ($warnings) {        if ($warnings) {
         print "WARNING: m value is not simple text\n";          print "WARNING: m value is not simple text\n";
       }        }
       $fix_by_hand = 1;  
       next;        next;
     }      }
     my $text = $m->firstChild->nodeValue;      my $text = $m->firstChild->nodeValue;
Line 552  sub replace_m { Line 547  sub replace_m {
     $m->parentNode->replaceChild($fragment, $m);      $m->parentNode->replaceChild($fragment, $m);
           
   }    }
   return $fix_by_hand;  
 }  }
   
 # Returns the HTML equivalent of LaTeX input, using tth  # Returns the HTML equivalent of LaTeX input, using tth
Line 591  sub html_to_dom { Line 585  sub html_to_dom {
 # Use the linked sty files to guess which newly defined elements should be considered blocks.  # Use the linked sty files to guess which newly defined elements should be considered blocks.
 # Also adds to @containing_styles the sty elements that contain styles.  # Also adds to @containing_styles the sty elements that contain styles.
 # @param {string} fn - the file path (we only extract the directory path from it)  # @param {string} fn - the file path (we only extract the directory path from it)
   # @param {string} res_dir - path of res directory parent (without the / at the end)
   # @param {Element} root - the root element
 sub add_sty_blocks {  sub add_sty_blocks {
   my ($fn, $root, $all_block) = @_;    my ($fn, $res_dir, $root, $all_block) = @_;
   my $doc = $root->ownerDocument;    my $doc = $root->ownerDocument;
   my @parserlibs = $doc->getElementsByTagName('parserlib');    my @parserlibs = $doc->getElementsByTagName('parserlib');
   my @libs = ();    my @libs = ();
Line 608  sub add_sty_blocks { Line 604  sub add_sty_blocks {
   my ($name, $path, $suffix) = fileparse($fn);    my ($name, $path, $suffix) = fileparse($fn);
   foreach my $sty (@libs) {    foreach my $sty (@libs) {
     if (substr($sty, 0, 1) eq '/') {      if (substr($sty, 0, 1) eq '/') {
       $sty = $RES_DIR.$sty;        $sty = $res_dir.$sty;
     } else {      } else {
       $sty = $path.$sty;        $sty = $path.$sty;
     }      }
Line 1816  sub remove_useless_notsolved { Line 1812  sub remove_useless_notsolved {
 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','translated','lang','instructorcomment','togglebox','standalone','form');    my @blocks_with_p = ('loncapa','library','problem','part','problemtype','window','block','while','postanswerdate','preduedate','solved','notsolved','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');
   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))) {
Line 2398  sub pretty { Line 2394  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 = $node->firstChild;
Line 2482  sub pretty { Line 2499  sub pretty {
         if ($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.1.1  
changed lines
  Added in v.1.6


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