function &process($html, &$pipeline)
{
// Run the XML parser on the XHTML we've prepared
$dom_tree = TreeBuilder::build($html);
// Check if parser returned valid document
if (is_null($dom_tree)) {
readfile(HTML2PS_DIR . '/templates/cannot_parse.html');
error_log(sprintf("Cannot parse document: %s", $pipeline->get_base_url()));
die;
}
/**
* Detect the base URI for this document.
*
* According to the HTML 4.01 p. 12.4.1:
* User agents must calculate the base URI according to the following precedences (highest priority to lowest):
*
* 1. The base URI is set by the BASE element.
* 2. The base URI is given by meta data discovered during a protocol interaction, such as an HTTP header (see [RFC2616]).
* 3. By default, the base URI is that of the current document. Not all HTML documents have a base URI (e.g., a valid HTML document may appear in an email and may not be designated by a URI). Such HTML documents are considered erroneous if they contain relative URIs and rely on a default base URI.
*/
/**
* Check if BASE element present; use its first occurrence
*/
$this->_scan_base($dom_tree, $pipeline);
/**
* @todo fall back to the protocol metadata
*/
/**
* Parse STYLE / LINK nodes containing CSS references and definitions
* This should be done here, as the document body may include STYLE node
* (this violates HTML standard, but is rather often appears in Web)
*/
scan_styles($dom_tree, $pipeline);
// Temporary hack: convert CSS rule array to CSS object
global $g_css;
global $g_css_obj;
$g_css_obj = new CSSObject();
foreach ($g_css as $rule) {
$g_css_obj->add_rule($rule, $pipeline);
}
$body = traverse_dom_tree_pdf($dom_tree);
$box =& create_pdf_box($body, $pipeline);
return $box;
}
function scan_styles($root, &$pipeline)
{
switch ($root->node_type()) {
case XML_ELEMENT_NODE:
if ($root->tagname() === 'style') {
// Parse <style ...> ... </style> nodes
//
parse_style_node($root, $pipeline);
} elseif ($root->tagname() === 'link') {
// Parse <link rel="stylesheet" ...> nodes
//
$rel = strtolower($root->get_attribute("rel"));
$type = strtolower($root->get_attribute("type"));
if ($root->has_attribute("media")) {
$media = explode(",", $root->get_attribute("media"));
} else {
$media = array();
}
if ($rel == "stylesheet" && ($type == "text/css" || $type == "") && (count($media) == 0 || is_allowed_media($media))) {
$src = $root->get_attribute("href");
if ($src) {
css_import($src, $pipeline);
}
}
}
// Note that we continue processing here!
// Note that we continue processing here!
case XML_DOCUMENT_NODE:
// Scan all child nodes
$child = $root->first_child();
while ($child) {
scan_styles($child, $pipeline);
$child = $child->next_sibling();
}
break;
}
}
function FrameBox(&$root, &$pipeline)
{
// Inherit 'border' CSS value from parent (FRAMESET tag), if current FRAME
// has no FRAMEBORDER attribute, and FRAMESET has one
$parent = $root->parent();
if (!$root->has_attribute('frameborder') && $parent->has_attribute('frameborder')) {
pop_border();
push_border(get_border());
}
$this->GenericContainerBox($root);
// If NO src attribute specified, just return.
if (!$root->has_attribute('src')) {
return;
}
// Determine the fullly qualified URL of the frame content
$src = $root->get_attribute('src');
$url = $pipeline->guess_url($src);
$data = $pipeline->fetch($url);
/**
* If framed page could not be fetched return immediately
*/
if (is_null($data)) {
return;
}
/**
* Render only iframes containing HTML only
*
* Note that content-type header may contain additional information after the ';' sign
*/
$content_type = $data->get_additional_data('Content-Type');
$content_type_array = explode(';', $content_type);
if ($content_type_array[0] != "text/html") {
return;
}
$html = $data->get_content();
// Remove control symbols if any
$html = preg_replace('/[\\x00-\\x07]/', "", $html);
$converter = Converter::create();
$html = $converter->to_utf8($html, $data->detect_encoding());
$html = html2xhtml($html);
$tree = TreeBuilder::build($html);
// Save current stylesheet, as each frame may load its own stylesheets
//
global $g_css;
$old_css = $g_css;
global $g_css_obj;
$old_obj = $g_css_obj;
scan_styles($tree, $pipeline);
// Temporary hack: convert CSS rule array to CSS object
$g_css_obj = new CSSObject();
foreach ($g_css as $rule) {
$g_css_obj->add_rule($rule, $pipeline);
}
// TODO: stinks. Rewrite
//
$frame_root = traverse_dom_tree_pdf($tree);
$box_child =& create_pdf_box($frame_root, $pipeline);
$this->add_child($box_child);
// Restore old stylesheet
//
$g_css = $old_css;
$g_css_obj = $old_obj;
$pipeline->pop_base_url();
}
