/**
* Performs the import of the current webPage object as a folder.
*
* @param WebPage $webPage
* @param WPMetaConfigs $meta
* @param Array|null $html_post_lookup
* @param Array|null $media_lookup
*
* @return null
*/
protected function doImport(WebPage $webPage, WPMetaConfigs $meta, &$html_post_lookup = null, &$media_lookup = null)
{
$updateResult = $meta->updateWPPost();
if (is_wp_error($updateResult)) {
echo '<li>***Unable to create folder ' . $meta->getPostTitle() . ' from ' . $meta->getSourcePath() . '</li>';
} else {
$webPage->setWPID($updateResult);
echo '<li>Folder created from ' . $meta->getPostTitle() . ' into post #' . $updateResult . ' with title ' . $meta->getPostTitle() . '</li>';
}
}
/**
* Function that performs the action of updating the local links on the webpage.
*
* @param WebPage $webPage
* @param HTMLImportStages $stagesSettings
* @param WPMetaConfigs $meta
* @param null $html_post_lookup
*
* @return null
*/
protected function performStage(WebPage $webPage, HTMLImportStages $stagesSettings, WPMetaConfigs &$meta, &$html_post_lookup = null)
{
$body = $meta->getPostContent();
if (!is_null($html_post_lookup)) {
$bodyXML = XMLHelper::getXMLObjectFromString($body);
$filepath = dirname($meta->getSourcePath());
$link_table = array();
// get a list of all the links in the page and iterate through them
$all_links = $bodyXML->xpath('//a[@href]');
// TODO: encapsulate this in a function use XMLHelper::getAllHRefsFromHTML as a start
if ($all_links) {
foreach ($all_links as $link) {
// iterate the link's attributes to find the HREF value
foreach ($link->attributes() as $attribute => $value) {
$path = '' . $value;
if (0 == strcasecmp('href', $attribute)) {
// TODO: handle foo.html#rar
if (!preg_match('/^[a-zA-Z].*:.*/', $path)) {
// TODO: need to handle foo.html without handling http://...
// only handle files that end in .html or .htm
if (preg_match('/\\.([hH][tT][mM][lL]?)$/', $path)) {
// if the file the path links to has been imported already, then it will exist in the lookup and we can update the link, otherwise leave the link alone
$fullpath = $webPage->getFullPath($path);
if ($fullpath) {
if (array_key_exists($fullpath, $html_post_lookup)) {
$link_table[$path] = $fullpath;
}
} else {
echo '<span>***could not update link ' . $path . '</span><br>';
}
}
}
}
}
}
}
// after building a list of all the links to update and what to update them to, we can do a change in the html file as a whole to catch all references
foreach ($link_table as $link => $full_link) {
$post_id = $html_post_lookup[$full_link];
$post_link = get_permalink($post_id);
echo 'Updating ' . $link . ' with ' . $post_link . '<br>';
$search_str = '/(\\b[hH][rR][eE][fF]\\s*=\\s*")([\\b\\.\\/]*' . preg_quote($link, '/') . '\\b)(")/';
$body = preg_replace($search_str, '$1' . preg_quote($post_link, '/') . '$3', $body);
}
$meta->setPostContent($body);
}
}
/**
* Build a LinkedTree tree from the file ordering and the file listing derived from the Flare index files.
* The HTML file information is stored in the payload of the LinkedTree with WebPage object representing the file.
*
* @param Array $fileOrder array based on Toc.js
* @param Array $fileList array based on Toc_Chunk0.js
* @param LinkedTree $parentNode the base node to attach the hierarchy to
*
* @return LinkedTree|null
*/
private function buildTree(array $fileOrder, array $fileList, LinkedTree $parentNode = null)
{
$firstNode = null;
$counter = 0;
foreach ($fileOrder as $item) {
$itemIndex = $item['i'];
$pagePath = $fileList[$itemIndex]['path'];
$pageTitle = $fileList[$itemIndex]['title'];
$node = new WebPage($this->retriever, $pageTitle, $pagePath);
$node->setOrderPosition($itemIndex);
if (!is_null($parentNode)) {
$parentNode->addChild($node);
} else {
$this->trees[] = $node;
}
if ($counter == 0) {
$firstNode = $node;
}
if (array_key_exists('n', $item)) {
$itemChildren = $item['n'];
$this->buildTree($itemChildren, $fileList, $node);
}
$counter++;
}
return $firstNode;
}
/**
* Imports a WebPage object into Wordpress, using the provided HtMLImportSettings, and assigning it to be a child of the post with the id defined in $parent_page_id. $html_post_lookup is used to determine if the page had already been created by this session's import.
*
* @param \html_import\indices\WebPage $webPage
* @param \html_import\admin\HtmlImportSettings $globalSettings
* @param $parent_page_id
* @param $html_post_lookup
*
* @return \html_import\WPMetaConfigs
*/
private function importAnHTML(\html_import\indices\WebPage $webPage, html_import\admin\HtmlImportSettings $globalSettings, $parent_page_id, $html_post_lookup)
{
$title = $webPage->getTitle();
$pageMeta = new \html_import\WPMetaConfigs();
$post_id = null;
// determine if the page has already been imported, search by post title
$post = get_page_by_title(htmlspecialchars($title));
// TODO: bad form, its saved with htmlspecialchars so need to search using that. Need to find a way to not require this knowledge
if (isset($html_post_lookup)) {
// check to see if there's been an import of this page already, if so get its ID from the lookup
if (array_key_exists($webPage->getFullPath(), $html_post_lookup)) {
$post_id = $html_post_lookup[$webPage->getFullPath()];
} else {
// the post wasn't imported during this import, but a post already exists with its title. Use it.
if (!is_null($post)) {
$post_id = $post->ID;
echo '<li>Page with title ' . $title . ' and ID ' . $post_id . ' already exists, now tagged to be overwritten.</li>';
}
}
}
$pageMeta->buildConfig($globalSettings, $webPage, $post_id, $parent_page_id);
if (!is_null($title)) {
$pageMeta->setPostTitle($title);
}
return $pageMeta;
}
/**
* Performs the stage action of uploading media files and updating the WebPage accordingly.
*
* @param WebPage $webPage
* @param HTMLImportStages $stagesSettings
* @param WPMetaConfigs $meta
* @param null $media_lookup
*
* @return null
*/
protected function performStage(WebPage $webPage, HTMLImportStages $stagesSettings, WPMetaConfigs &$meta, &$media_lookup = null)
{
$post_id = $meta->getPostId();
$body = $meta->getPostContent();
if (is_null($body) || strcmp('', $body) == 0) {
echo '** the body for post ' . $post_id . ' was empty, no media to import.';
return;
}
$media_table = array();
$file_as_xml_obj = XMLHelper::getXMLObjectFromString($body);
// import img srcs
$all_imgs = $file_as_xml_obj->xpath('//img[@src]');
if ($all_imgs) {
foreach ($all_imgs as $img) {
foreach ($img->attributes() as $attribute => $value) {
$path = '' . $value;
if (0 == strcasecmp('src', $attribute)) {
// TODO: this is duplicated below, refactor it out
if (!preg_match('/^[a-zA-Z].*:.*/', $path)) {
// if it's local
if (!is_null($media_lookup) && !array_key_exists($path, $media_table)) {
$fullpath = $webPage->getFullPath($path);
if (array_key_exists($fullpath, $media_lookup)) {
$attach_id = $media_lookup[$fullpath];
require_once ABSPATH . 'wp-admin/includes/image.php';
$attach_data = wp_get_attachment_metadata($attach_id);
wp_update_attachment_metadata($attach_id, $attach_data);
$media_table[$path] = $fullpath;
} else {
$filename = basename($fullpath);
$upload = wp_upload_bits($filename, null, $webPage->getLinkContents($path));
if ($upload['error']) {
echo '<li>***Unable to upload media file ' . $filename . '</li>';
} else {
echo '<li>' . $filename . ' media file uploaded.</li>';
$wp_filetype = wp_check_filetype(basename($upload['file']), null);
$attachment = array('guid' => $upload['file'], 'post_mime_type' => $wp_filetype['type'], 'post_title' => preg_replace('/\\.[^.]+$/', '', basename($upload['file'])), 'post_content' => '', 'post_status' => 'inherit');
$attach_id = wp_insert_attachment($attachment, $upload['file'], $post_id);
require_once ABSPATH . 'wp-admin/includes/image.php';
$attach_data = wp_generate_attachment_metadata($attach_id, $upload['file']);
wp_update_attachment_metadata($attach_id, $attach_data);
$media_lookup[$fullpath] = $attach_id;
$media_table[$path] = $fullpath;
echo '<li>' . $filename . ' attached to post ' . $post_id . '</li>';
}
}
}
}
}
}
}
}
// linked media
$all_links = $file_as_xml_obj->xpath('//a[@href]');
// TODO: encapsulate this in a function
if ($all_links) {
foreach ($all_links as $link) {
foreach ($link->attributes() as $attribute => $value) {
$path = '' . $value;
if (0 == strcasecmp('href', $attribute)) {
if (!preg_match('/^[a-zA-Z].*:.*/', $path)) {
if (preg_match('/\\.(png|bmp|jpg|jpeg|gif|pdf|doc|docx|mp3|ogg|wav)$/', strtolower($path))) {
// media png,bmp,jpg,jpeg,gif,pdf,doc,docx,mp3,ogg,wav
if (!is_null($media_lookup)) {
/*if ( $path[0] != '/' ) {
$fullpath = realpath( dirname( $meta->getSourcePath() ) . '/' . $path );
} else {
$fullpath = $path;
}*/
$fullpath = $webPage->getFullPath($path);
if (array_key_exists($fullpath, $media_lookup)) {
$attach_id = $media_lookup[$fullpath];
require_once ABSPATH . 'wp-admin/includes/image.php';
$attach_data = wp_get_attachment_metadata($attach_id);
wp_update_attachment_metadata($attach_id, $attach_data);
$media_table[$path] = $fullpath;
} else {
$filename = basename($fullpath);
$upload = wp_upload_bits($filename, null, $webPage->getLinkContents($path));
if ($upload['error']) {
echo '<li>***Unable to upload media file ' . $filename . '</li>';
} else {
echo '<li>' . $filename . ' media file uploaded.</li>';
$wp_filetype = wp_check_filetype(basename($upload['file']), null);
$attachment = array('guid' => $upload['file'], 'post_mime_type' => $wp_filetype['type'], 'post_title' => preg_replace('/\\.[^.]+$/', '', basename($upload['file'])), 'post_content' => '', 'post_status' => 'inherit');
$attach_id = wp_insert_attachment($attachment, $upload['file'], $post_id);
require_once ABSPATH . 'wp-admin/includes/image.php';
$attach_data = wp_generate_attachment_metadata($attach_id, $upload['file']);
wp_update_attachment_metadata($attach_id, $attach_data);
$media_lookup[$fullpath] = $attach_id;
$media_table[$path] = $fullpath;
echo '<li>' . $filename . ' attached to post ' . $post_id . '</li>';
}
}
}
}
}
}
}
}
}
foreach ($media_table as $media_item => $full_media_path) {
$media_id = $media_lookup[$full_media_path];
$media_url = wp_get_attachment_url($media_id);
$search_str = '/(\\b[iI][mM][gG]\\s*[^>]*\\s+[sS][rR][cC]\\s*=\\s*")([\\b\\/\\.]*' . preg_quote($media_item, '/') . '\\b)(")/';
$body = preg_replace($search_str, '$1' . preg_quote($media_url, '/') . '$3', $body);
// img src
$body = preg_replace('/(\\b[hH][rR][eE][fF]\\s*=\\s*")(\\b' . preg_quote($media_item, '/') . '\\b)(")/', '$1' . preg_quote($media_url, '/') . '$3', $body);
// a href
}
$meta->setPostContent($body);
echo '<li>Post ' . $post_id . ' updated with correct image links.</li>';
}
