version 1.8, 2016/01/20 00:40:39
|
version 1.11, 2016/11/10 21:53:56
|
Line 128 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 1810 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) = @_; |
Line 1835 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 2011 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 2247 sub fix_empty_lc_elements {
|
Line 2296 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 { |