--- loncom/homework/cleanxml/post_xml.pm 2016/01/06 16:44:32 1.5
+++ loncom/homework/cleanxml/post_xml.pm 2016/01/20 00:40:39 1.8
@@ -1,7 +1,7 @@
# The LearningOnline Network
# Third step to clean a file.
#
-# $Id: post_xml.pm,v 1.5 2016/01/06 16:44:32 damieng Exp $
+# $Id: post_xml.pm,v 1.8 2016/01/20 00:40:39 damieng Exp $
#
# Copyright Michigan State University Board of Trustees
#
@@ -41,6 +41,8 @@ use Cwd 'abs_path';
use XML::LibXML;
use HTML::TokeParser; # used to parse sty files
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
@@ -50,7 +52,7 @@ my @inline_like_block = ('stringresponse
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 @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 @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',
@@ -552,11 +554,11 @@ sub replace_m {
# Returns the HTML equivalent of LaTeX input, using tth
sub tth {
my ($text) = @_;
- my ($fh, $tmp_path) = tempfile();
- binmode($fh, ':utf8');
- print $fh $text;
- close $fh;
- my $output = `tth -r -w2 -u -y0 < $tmp_path 2>/dev/null`;
+ my $output = &tth::tth($text);
+ my $errorstring = &tth::ttherror();
+ if ($errorstring) {
+ die $errorstring;
+ }
# hopefully the temp file will not be removed before this point (otherwise we should use unlink_on_destroy 0)
$output =~ s/^\s*|\s*$//;
$output =~ s/
<\/div>/
/; # why is tth using such ugly markup for \newline ?
@@ -567,7 +569,7 @@ sub tth {
sub html_to_dom {
my ($text) = @_;
$text = ''.$text.'';
- 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 into a character
my $dom_doc = XML::LibXML->load_xml(string => $textref);
my $root = $dom_doc->documentElement;
@@ -1812,8 +1814,8 @@ sub remove_useless_notsolved {
sub fix_paragraphs_inside {
my ($node, $all_block) = @_;
# 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 @fix_p_if_br_or_p = (@responses,'foil','item','text','label','hintgroup','hintpart','hint','web','windowlink','div','li','dd','td','th','blockquote');
+ 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','solved','notsolved');
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))) {
# if non-empty, add paragraphs where needed between all br and remove br
@@ -2394,8 +2396,29 @@ sub pretty {
my $type = $node->nodeType;
if ($type == XML_ELEMENT_NODE) {
my $name = $node->nodeName;
- if ((string_in_array($all_block, $name) || string_in_array(\@inline_like_block, $name)) &&
- !string_in_array(\@preserve_elements, $name)) {
+ if (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
if (defined $node->firstChild && !string_in_array(\@no_newline_inside, $name)) {
my $first = $node->firstChild;
@@ -2478,26 +2501,6 @@ sub pretty {
if ($text eq '') {
$node->removeChild($node->lastChild);
} 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);
}
}