' . __( 'Sorry, there has been an error.', 'blogger-importer' ) . ' ';
printf( __( 'The export file could not be found at %s. It is likely that this was caused by a permissions problem.', 'blogger-importer' ), esc_html( $file['file'] ) );
echo '
' . __( 'Sorry, there has been an error.', 'blogger-importer' ) . ' ';
echo esc_html( $import_data->get_error_message() ) . '
';
return false;
}
$this->get_authors_from_import( $import_data );
return true;
}
/**
* Retrieve authors from parsed data
*
* @param array $import_data Data returned by a WXR parser
*/
function get_authors_from_import( $import_data ) {
$feed = $this->parse($import_data);
$authors = $feed->get_authors();
foreach ($authors as $author) {
$login = sanitize_user( $author->get_name(), true );
$this->authors[$login] = array(
'author_login' => $author->get_name(),
'author_display_name' => $author->get_name(),
'author_email' => $author->get_email(),
);
}
}
/**
* Display pre-import options, author importing/mapping and option to
* fetch attachments
*/
function import_options() {
$j = 0;
?>
' . esc_html( $author['author_display_name'] );
echo ' ';
$create_users = $this->allow_create_users();
if ( $create_users ) {
_e( 'as a new user:', 'blogger-importer' );
$value = esc_attr( sanitize_user( $author['author_login'], true ) );
echo ' ';
}
_e( 'or assign posts to an existing user:', 'blogger-importer' );
wp_dropdown_users( array( 'name' => "user_map[$n]", 'multi' => true, 'show_option_all' => __( '- Select -', 'blogger-importer' ) ) );
echo '';
}
/**
* Map old author logins to local user IDs based on decisions made
* in import options form. Can map to an existing user, create a new user
* or falls back to the current user in case of error with either of the previous
*/
function get_author_mapping() {
if ( ! isset( $_POST['imported_authors'] ) )
return;
$create_users = $this->allow_create_users();
foreach ( (array) $_POST['imported_authors'] as $i => $old_login ) {
// Multisite adds strtolower to sanitize_user. Need to sanitize here to stop breakage in process_posts.
$santized_old_login = sanitize_user( $old_login, true );
$old_id = isset( $this->authors[$old_login]['author_id'] ) ? intval($this->authors[$old_login]['author_id']) : false;
if ( ! empty( $_POST['user_map'][$i] ) ) {
$user = get_userdata( intval($_POST['user_map'][$i]) );
if ( isset( $user->ID ) ) {
if ( $old_id )
$this->processed_authors[$old_id] = $user->ID;
$this->author_mapping[$santized_old_login] = $user->ID;
}
} else if ( $create_users ) {
if ( ! empty($_POST['user_new'][$i]) ) {
$user_id = wp_create_user( $_POST['user_new'][$i], wp_generate_password() );
} else if ( $this->version != '1.0' ) {
$user_data = array(
'user_login' => $old_login,
'user_pass' => wp_generate_password(),
'user_email' => isset( $this->authors[$old_login]['author_email'] ) ? $this->authors[$old_login]['author_email'] : '',
'display_name' => $this->authors[$old_login]['author_display_name'],
'first_name' => isset( $this->authors[$old_login]['author_first_name'] ) ? $this->authors[$old_login]['author_first_name'] : '',
'last_name' => isset( $this->authors[$old_login]['author_last_name'] ) ? $this->authors[$old_login]['author_last_name'] : '',
);
$user_id = wp_insert_user( $user_data );
}
if ( ! is_wp_error( $user_id ) ) {
if ( $old_id )
$this->processed_authors[$old_id] = $user_id;
$this->author_mapping[$santized_old_login] = $user_id;
} else {
if ( array_key_exists( $old_login, $this->authors ) ) {
printf( __( 'Failed to create new user for %s. Their posts will be attributed to the current user.', 'blogger-importer' ), esc_html( $this->authors[$old_login]['author_display_name'] ) );
} else {
printf( __( 'Failed to create new user. Their posts will be attributed to the current user.', 'blogger-importer' ) );
}
if ( defined('IMPORT_DEBUG') && IMPORT_DEBUG )
echo ' ' . $user_id->get_error_message();
echo ' ';
}
}
// failsafe: if the user_id was invalid, default to the current user
if ( ! isset( $this->author_mapping[$santized_old_login] ) ) {
if ( $old_id )
$this->processed_authors[$old_id] = (int) get_current_user_id();
$this->author_mapping[$santized_old_login] = (int) get_current_user_id();
}
}
}
/**
* Create new posts based on import information
*/
function process_posts() {
$feed = $this->import_data;
foreach ( $feed->get_items() as $item ) {
// check that it is actually a post first
//
$is_post = false;
$cats = $item->get_categories();
foreach ( $cats as $cat ) {
if ( $cat == 'http://schemas.google.com/blogger/2008/kind#post' ) {
$is_post = true;
break;
}
}
// only import posts for now
if ( ! $is_post ) {
continue;
}
$blogentry = new BloggerEntry();
$blogentry->blogurl = $this->host;
$blogentry->id = $item->get_id();
$blogentry->published = $item->get_published();
$blogentry->updated = $item->get_updated();
$blogentry->isDraft = $item->get_draft_status();
$blogentry->title = $item->get_title();
$blogentry->content = $item->get_content();
$blogentry->geotags = $item->get_geotags();
// map the post author
$blogentry->bloggerauthor = sanitize_user( $item->get_author()->get_name(), true );
if ( isset( $this->author_mapping[$blogentry->bloggerauthor] ) )
$blogentry->author = $this->author_mapping[$blogentry->bloggerauthor];
else
$blogentry->author = (int) get_current_user_id();
$blogentry->links = $item->get_links(array('replies', 'edit', 'self', 'alternate'));
$blogentry->parselinks();
foreach ( $cats as $cat ) {
if ( false === strpos( $cat, 'http://schemas.google.com') ) {
$blogentry->categories[] = $cat;
}
}
// Checks for duplicates
$post_id = $blogentry->post_exists();
if ( $post_id != 0 ) {
$this->posts_skipped++;
} else {
//Unique new post so import it
$post_id = $blogentry->import();
$this->posts_done++;
}
}
}
/**
* Create new comments based on import information
*/
function process_comments() {
$feed = $this->import_data;
foreach ( $feed->get_items() as $item ) {
// check that it is actually a comment first
//
$is_comment = false;
$cats = $item->get_categories();
foreach ( $cats as $cat ) {
if ( $cat == 'http://schemas.google.com/blogger/2008/kind#comment' ) {
$is_comment = true;
break;
}
}
// we only import comments here
if ( ! $is_comment ) {
continue;
}
$commententry = new CommentEntry();
$commententry->id = $item->get_id();
$commententry->updated = $item->get_updated();
$commententry->content = $item->get_content();
$commententry->author = $item->get_author()->get_name();
$commententry->authoruri = $item->get_author()->get_link();
$commententry->authoremail = $item->get_author()->get_email();
$replyto = $item->get_item_tags('http://purl.org/syndication/thread/1.0','in-reply-to');
$commententry->source = $replyto[0]['attribs']['']['source'];
$commententry->source = $item->get_source();
$parts = parse_url($commententry->source);
$commententry->old_post_permalink = $parts['path']; //Will be something like this '/feeds/417730729915399755/posts/default/8397846992898424746'
$bloggerentry = new BloggerEntry();
$commententry->post_ID = $bloggerentry->get_post_by_oldID($commententry->old_post_permalink);
//Get the links
$commententry->links = $item->get_links(array('edit', 'self', 'alternate', 'related'));
$commententry->parselinks();
// Nested comment?
if ( isset($commententry->related) ) {
$commententry->parentcommentid = $commententry->get_comment_by_oldID($commententry->related);
}
//Perhaps could log errors here?
if ($commententry->post_ID != 0) {
// Checks for duplicates
if ($comment_id = $commententry->exists()) {
$this->comments_skipped++;
} else {
$comment_id = $commententry->import();
$this->comments_done++;
}
} else {
$this->comments_skipped++;
}
}
}
/*
* Search for either a linked image or a non linked image within the supplied html
* or
* Ref: http://www.the-art-of-web.com/php/parse-links/
* "]*href=(\"??)([^\" >]*?)\\1[^>]*>(.*)<\/a>"
* http://wordpress.org/extend/plugins/blogger-image-import/
* "]+href\=([\"'`])(.*)\\1[^<]*?]*src\=([\"'`])(.*)\\3[^>]*>"
*/
function get_images($content) {
$highrez = array();
$lowrez = array();
//First images with links
//Might be nice to expand this top try and get Alt and/or Title attributes for use as description
$regexp = "]*href\=([\"'`])([^> ]*?)\\1[^<]*?]*src\=([\"'`])([^\> ]*?)\\3[^>]*>";
if (preg_match_all("/$regexp/siU", $content, $matches1, PREG_SET_ORDER)) {
//http://www.techrepublic.com/article/17-useful-functions-for-manipulating-arrays-in-php/5792851
foreach ($matches1 as $match) {
if ($this->isurlimage($match[2])) {
$highrez[$match[4]] = $match[2];
} else {
$lowrez[$match[4]] = '';
}
}
}
//Now any image (n.b. this overlaps the previous set)
$regexp = "]*src\=([\"'`])([^\> ]*?)\\1[^>]*>";
if (preg_match_all("/$regexp/siU", $content, $matches2, PREG_SET_ORDER)) {
foreach ($matches2 as $match) {
$lowrez[$match[2]] = '';
}
}
//Remove any rows from this second set that are already in the first set and merge two sets of results
$images = array_merge($lowrez, $highrez);
return $images;
}
/**
* Update all of the images in the posts that have already been imported
*/
function process_images() {
$postsprocessed = $this->images_progress;
if ($postsprocessed == 0) {
$imagesprocessed = 0;
$imagesskipped = 0;
} else {
$imagesprocessed = $this->images_done;
$imagesskipped = $this->images_skipped;
}
$batchsize = 20;
$loadedposts = get_posts( array(
'meta_key' => 'blogger_blog',
'meta_value' => $this->host,
'posts_per_page' => $batchsize,
'offset' => $postsprocessed,
'post_status' => array('draft', 'publish', 'future')
));
//Stop if nothing left
if (count($loadedposts) == 0) {
return true;
}
foreach($loadedposts as $importedpost) {
$importedcontent = $importedpost->post_content;
$author = get_post_meta($importedpost->ID, 'blogger_author', true);
$img_count = 0; //Count of images for this post
foreach($this->get_images($importedcontent) as $lowrez => $highrez) {
if (!$this->image_filter($lowrez)) {
//Pass null for description so that the default (filename) is used, might be good to use Alt tag instead?
$newcontent = $this->import_image($importedpost->ID, $lowrez, $highrez, null, $img_count, $importedcontent, $this->host, $author);
if (!is_wp_error($newcontent)) {
$importedcontent = $newcontent;
$img_count++;
} else {
$imagesskipped++;
}
}
}
$imagesprocessed += $img_count;
$importedpost->post_content = $importedcontent;
wp_update_post($importedpost);
$postsprocessed++;
$this->images_done = $imagesprocessed;
$this->images_progress = $postsprocessed;
$this->images_skipped = $imagesskipped;
}
unset($loadedposts);
return;
}
function image_urlremap($url, $large) {
/* Fix problem urls
e.g. change https://lh4.googleusercontent.com/-nt66qhxzDyY/TZOD-RhTYMI/AAAAAAAACd4/Elzm1smRFb4/s800-h/Ski%2520Trip.jpg to
to https://lh4.googleusercontent.com/-nt66qhxzDyY/TZOD-RhTYMI/AAAAAAAACd4/Elzm1smRFb4/s800/Ski%2520Trip.jpg
Could use a apply_filter here to allow users to add their own tweeks
*/
$pattern = '/(\/)(s\d*)-h(\/)/i';
$replacement = '$1$2$3';
$img = preg_replace($pattern, $replacement, $url);
/* Strip out ? and # on the end of files */
$pattern = '/(.*)[#\?].*/i';
$replacement = '$1';
$img = preg_replace($pattern, $replacement, $img);
if ($large) {
// For images on blogger we can swap /sXXX/ with for example /s1600/ to get a larger file.
// Use a standardised large size so we can control quality vs filesize.
$pattern = '/(\/)(s\d*)(\/)/i';
$replacement = '$1s'.Blogger_Importer::LARGE_IMAGE_SIZE.
'$3';
$img = preg_replace($pattern, $replacement, $img);
}
return $img;
}
function image_filter($url) {
// Do we exclude this particular image?
// Don't include images that are already loaded onto this site
// Could use a apply_filter here to allow users to add their own tweeks
return (substr($url, 0, strlen(site_url())) == site_url());
}
function import_image($post_id, $lowrez, $highrez, $description, $imgcount, $postcontent, $blogname, $author) {
/* Import a new image unless we specifically filter it out or if it has already been downloaded on another page.
Based on http://wordpress.stackexchange.com/questions//media-sideload-image-file-name and the tumblr-importer
Simple filename cleaning as characters such as +, % cause issues ref: http://wordpress.org/extend/plugins/uploadplus/
It's processing links of a form similar to these as provided by the "get_images" function
or
If the high resolution (linked) file is not an image then the low resolution version is downloaded.
*/
$lowrez_old = $lowrez;
$highrez_old = $highrez;
$highrezispage = false;
$lowrez = $this->image_urlremap($lowrez, false);
if ($lowrez == '')
return new WP_Error('Not an image', $message = __('Lowrez not an image', 'blogger-importer'), $data = array($lowrez_old, $highrez_old));
if ($highrez != '') {
$highrez = $this->image_urlremap($highrez, true);
} else {
$highrez = $this->image_urlremap($lowrez, true);
}
if (!$att_id = $this->image_exists($lowrez)) {
//Option to add a timeout to download_url, but don't use the wp_remote_get timeout as that's much shorter than the default here of 300s
$tmp = @download_url($highrez);
if (is_wp_error($tmp)) {
@unlink($tmp); // clean up, copied this from other examples but how is this supposed to work if $tmp is an error??
//Don't exit as can still try the small image
}
// If the highrez was not an image then try the lowrex
if (!$this->is_image($tmp, $highrez)) {
$highrezispage = true; //That image was not valid
$tmp = @download_url($lowrez); // Option to add a timeout here
if (is_wp_error($tmp)) {
@unlink($tmp); // clean up
return $tmp; // output wp_error
}
if (!$this->is_image($tmp, $lowrez)) {
@unlink($tmp); // clean up None of items are actually images, for example might be a single pixel, deliberately filtered out or a 404 error?
return new WP_Error('No Images', __('None of the images are valid', 'blogger-importer'), $data = array($lowrez_old, $highrez_old));
}
}
$new_name = preg_replace('/[^A-Za-z0-9._ ]/i', '-', basename($lowrez));
$file_array = array('name' => $new_name, 'tmp_name' => $tmp);
if ( empty( $description ) ) {
$description = $new_name;
}
$att_id = media_handle_sideload($file_array, $post_id, $description);
if (is_wp_error($att_id)) {
@unlink($file_array['tmp_name']);
return $att_id;
}
// Link attachment upto old url, store the author so we can replace it later
add_post_meta($att_id, 'blogger_permalink', $lowrez);
add_post_meta($att_id, 'blogger_blog', $blogname, true);
add_post_meta($att_id, 'blogger_author', $author, true);
if ($highrezispage) //Not an image so store so we can link later
add_post_meta($att_id, 'blogger_largeimgispage', true);
} else {
//Image already exists, check if the high rez one was valid last time
$tmp = get_post_meta($att_id, 'blogger_largeimgispage', true);
if ($tmp == true)
$highrezispage = true;
}
//Always treat picassa webs as image so they get replaced with the new High rez link
if (substr($highrez, 0, 27) == 'http://picasaweb.google.com')
$highrezispage = false;
//Replace the image strings
if (!$highrezispage && $highrez_old != '') {
$imagesrc = wp_get_attachment_image_src($att_id, 'full');
$highrez_new = reset($imagesrc);
$postcontent = str_replace($highrez_old, $highrez_new, $postcontent);
}
$imagesrc = wp_get_attachment_image_src($att_id, 'medium');
$lowrez_new = reset($imagesrc);
$postcontent = str_replace($lowrez_old, $lowrez_new, $postcontent);
//Set the first image to be the post thumbnail (zero index)
if ($imgcount == 0) {
set_post_thumbnail($post_id, $att_id);
}
//media handle sideload moves the file so there should be no temp file left but cleanup just incase.
@unlink($tmp);
// incase something goes wrong
if ($postcontent == '') {
return new WP_Error('Empty Content', __("Attempting to write back empty content", 'blogger-importer'), $data = array($lowrez_old, $highrez_old));
}
return $postcontent;
}
function is_image($file, $filename) {
//Is the downloaded file really an image
//e.g. it looked like an image from the URL but when downloaded it was something else perhaps a html page
//Also filter out tracking images of 1 pixel square
//Found that wp_check_filetype_and_ext and wp_match_mime_types was giving false positives
$imgstats = @getimagesize($file);
if (!$imgstats) {
return false;
}
return (($imgstats[0] > 1) && ($imgstats[1] > 1));
}
function image_exists($lowrez) {
global $wpdb;
return $wpdb->get_var($wpdb->prepare("SELECT ID FROM $wpdb->posts p INNER JOIN $wpdb->postmeta m ON p.ID = m.post_id AND meta_key = 'blogger_permalink' WHERE post_type = 'attachment' AND meta_value = %s LIMIT 0 , 1",
$lowrez));
}
function process_links() {
//Update all of the links in the blog
global $wpdb;
$postsprocessed = $this->links_progress;
if ($postsprocessed == 0) {
$linksprocessed = 0;
} else {
$linksprocessed = $this->links_done;
}
$batchsize = 20;
$oldurlsearch = $this->host;
if (substr($oldurlsearch, 0, 3) == 'www.') {
$oldurlsearch = substr($oldurlsearch, 3, strlen($oldurlsearch - 3));
}
$oldurlsearch = str_replace('.', '\.', $oldurlsearch);
$blogspot = stripos($oldurlsearch, '\.blogspot\.');
if ($blogspot) { //Blogspot addresses can be international e.g. myblog.blogspot.com, myblog.blogspot.com.au or myblog.blogspot.co.uk or myblog.blogspot.de both resolve to the same blog.
//See http://www.searchenginejournal.com/google-blogger-url-censorship/39724/
$oldurlsearch = substr($oldurlsearch, 0, $blogspot + 12).
'[\w\.]{2,6}';
}
$loadedposts = get_posts(array('meta_key' => 'blogger_blog', 'meta_value' => $this->host, 'posts_per_page' => $batchsize, 'offset' => $postsprocessed, 'post_status' => array('draft', 'publish', 'future')));
//Stop if nothing left
if (count($loadedposts) == 0) {
return true;
}
foreach($loadedposts as $importedpost) {
$importedcontent = $importedpost->post_content;
$regexp = ']*href=([\"\'`])(https?:\/\/(?:www\.)*'.$oldurlsearch.'\/)([^\" >]*?)\1[^>]*>(.*)<\/a>';
if (preg_match_all("/$regexp/siU", $importedcontent, $matches, PREG_SET_ORDER)) {
foreach($matches as $match) {
$HostURL = substr($match[2], 0, strlen($match[2]) - 1); //e.g. http://minitemp.blogspot.co.uk
$PageURL = '/'.$match[3]; //e.g. '/2011/04/what-happens-if-blog-title-is-really.html'
$sql = $wpdb->prepare("SELECT post_id FROM $wpdb->postmeta m
inner join $wpdb->posts p on p.id = m.post_id and post_type = 'post'
where meta_key = '%s'
and meta_value = '%s'
", 'blogger_permalink', $PageURL);
$linkpostid = $wpdb->get_var($sql);
if ($linkpostid != 0) {
$NewURL = get_permalink($linkpostid);
} else { // Page not found, update content with just the new domain
$NewURL = site_url($PageURL);
}
$importedcontent = str_replace($HostURL.$PageURL, $NewURL, $importedcontent); $linksprocessed++;
}
if ($importedcontent == '') {
return new WP_Error('Empty Content', __("Attempting to write back empty content"));
}
$importedpost->post_content = $importedcontent;
wp_update_post($importedpost);
}
$postsprocessed++;
//For some reason the intermediate values are not getting written, is it that the options are cached hence not read back?
$this->links_done = $linksprocessed;
$this->links_progress = $postsprocessed;
}
unset($loadedposts);
return;
}
function isurlimage($srcurl) {
//Process picasaweb links and files that are images
if (substr($srcurl, 0, 27) == 'http://picasaweb.google.com')
return true;
return preg_match('/(?i)\.(jpe?g|png|gif|bmp)$/i', $srcurl);
}
/**
* Parse the atom file
*
* @param string $file Path to XML file for parsing
* @return SimplePie object of the feed
*/
function parse( $file ) {
$data = file_get_contents( $file );
// parse the feed
$feed = new SimplePie();
//set_xxxx methods depreciated (and not working?) replaced with get_registry as per docs
$reg = $feed->get_registry();
$reg->register('Sanitize', 'Blogger_Importer_Sanitize');
$feed->sanitize = $reg->create('Sanitize'); //Should not really need to do this but there seems to be an issue with the SimplePie class?
$reg->register('Item', 'WP_SimplePie_Blog_Item');
$feed->set_raw_data( $data );
$feed->init();
return $feed;
}
// Display import page title
function header() {
echo '
';
}
/**
* Decide if the given meta key maps to information we will want to import
*
* @param string $key The meta key to check
* @return string|bool The key if we do want to import, false if not
*/
function is_valid_meta_key( $key ) {
// skip attachment metadata since we'll regenerate it from scratch
// skip _edit_lock as not relevant for import
if ( in_array( $key, array( '_wp_attached_file', '_wp_attachment_metadata', '_edit_lock' ) ) )
return false;
return $key;
}
/**
* Decide whether or not the importer is allowed to create users.
* Default is true, can be filtered via import_allow_create_users
*
* @return bool True if creating users is allowed
*/
function allow_create_users() {
return apply_filters( 'import_allow_create_users', true );
}
/**
* Added to http_request_timeout filter to force timeout at 60 seconds during import
* @return int 60
*/
function bump_request_timeout( $val ) {
return 60;
}
// return the difference in length between two strings
function cmpr_strlen( $a, $b ) {
return strlen($b) - strlen($a);
}
}
} // class_exists( 'Blogger_Importer' )
function blogger_importer_init() {
load_plugin_textdomain( 'blogger-importer', false, dirname( plugin_basename( __FILE__ ) ) . '/languages/' );
/**
* WordPress Importer object for registering the import callback
* @global WP_Import $wp_import
*/
$GLOBALS['wp_import'] = new Blogger_Importer();
register_importer('blogger', __('Blogger', 'blogger-importer'), __('Import posts, comments, and categories from a Blogger blog and migrate authors to WordPress users.', 'blogger-importer'), array( $GLOBALS['wp_import'], 'dispatch' ) );
}
add_action( 'admin_init', 'blogger_importer_init' );
or
If the high resolution (linked) file is not an image then the low resolution version is downloaded.
*/
$lowrez_old = $lowrez;
$highrez_old = $highrez;
$highrezispage = false;
$lowrez = $this->image_urlremap($lowrez, false);
if ($lowrez == '')
return new WP_Error('Not an image', $message = __('Lowrez not an image', 'blogger-importer'), $data = array($lowrez_old, $highrez_old));
if ($highrez != '') {
$highrez = $this->image_urlremap($highrez, true);
} else {
$highrez = $this->image_urlremap($lowrez, true);
}
if (!$att_id = $this->image_exists($lowrez)) {
//Option to add a timeout to download_url, but don't use the wp_remote_get timeout as that's much shorter than the default here of 300s
$tmp = @download_url($highrez);
if (is_wp_error($tmp)) {
@unlink($tmp); // clean up, copied this from other examples but how is this supposed to work if $tmp is an error??
//Don't exit as can still try the small image
}
// If the highrez was not an image then try the lowrex
if (!$this->is_image($tmp, $highrez)) {
$highrezispage = true; //That image was not valid
$tmp = @download_url($lowrez); // Option to add a timeout here
if (is_wp_error($tmp)) {
@unlink($tmp); // clean up
return $tmp; // output wp_error
}
if (!$this->is_image($tmp, $lowrez)) {
@unlink($tmp); // clean up None of items are actually images, for example might be a single pixel, deliberately filtered out or a 404 error?
return new WP_Error('No Images', __('None of the images are valid', 'blogger-importer'), $data = array($lowrez_old, $highrez_old));
}
}
$new_name = preg_replace('/[^A-Za-z0-9._ ]/i', '-', basename($lowrez));
$file_array = array('name' => $new_name, 'tmp_name' => $tmp);
if ( empty( $description ) ) {
$description = $new_name;
}
$att_id = media_handle_sideload($file_array, $post_id, $description);
if (is_wp_error($att_id)) {
@unlink($file_array['tmp_name']);
return $att_id;
}
// Link attachment upto old url, store the author so we can replace it later
add_post_meta($att_id, 'blogger_permalink', $lowrez);
add_post_meta($att_id, 'blogger_blog', $blogname, true);
add_post_meta($att_id, 'blogger_author', $author, true);
if ($highrezispage) //Not an image so store so we can link later
add_post_meta($att_id, 'blogger_largeimgispage', true);
} else {
//Image already exists, check if the high rez one was valid last time
$tmp = get_post_meta($att_id, 'blogger_largeimgispage', true);
if ($tmp == true)
$highrezispage = true;
}
//Always treat picassa webs as image so they get replaced with the new High rez link
if (substr($highrez, 0, 27) == 'http://picasaweb.google.com')
$highrezispage = false;
//Replace the image strings
if (!$highrezispage && $highrez_old != '') {
$imagesrc = wp_get_attachment_image_src($att_id, 'full');
$highrez_new = reset($imagesrc);
$postcontent = str_replace($highrez_old, $highrez_new, $postcontent);
}
$imagesrc = wp_get_attachment_image_src($att_id, 'medium');
$lowrez_new = reset($imagesrc);
$postcontent = str_replace($lowrez_old, $lowrez_new, $postcontent);
//Set the first image to be the post thumbnail (zero index)
if ($imgcount == 0) {
set_post_thumbnail($post_id, $att_id);
}
//media handle sideload moves the file so there should be no temp file left but cleanup just incase.
@unlink($tmp);
// incase something goes wrong
if ($postcontent == '') {
return new WP_Error('Empty Content', __("Attempting to write back empty content", 'blogger-importer'), $data = array($lowrez_old, $highrez_old));
}
return $postcontent;
}
function is_image($file, $filename) {
//Is the downloaded file really an image
//e.g. it looked like an image from the URL but when downloaded it was something else perhaps a html page
//Also filter out tracking images of 1 pixel square
//Found that wp_check_filetype_and_ext and wp_match_mime_types was giving false positives
$imgstats = @getimagesize($file);
if (!$imgstats) {
return false;
}
return (($imgstats[0] > 1) && ($imgstats[1] > 1));
}
function image_exists($lowrez) {
global $wpdb;
return $wpdb->get_var($wpdb->prepare("SELECT ID FROM $wpdb->posts p INNER JOIN $wpdb->postmeta m ON p.ID = m.post_id AND meta_key = 'blogger_permalink' WHERE post_type = 'attachment' AND meta_value = %s LIMIT 0 , 1",
$lowrez));
}
function process_links() {
//Update all of the links in the blog
global $wpdb;
$postsprocessed = $this->links_progress;
if ($postsprocessed == 0) {
$linksprocessed = 0;
} else {
$linksprocessed = $this->links_done;
}
$batchsize = 20;
$oldurlsearch = $this->host;
if (substr($oldurlsearch, 0, 3) == 'www.') {
$oldurlsearch = substr($oldurlsearch, 3, strlen($oldurlsearch - 3));
}
$oldurlsearch = str_replace('.', '\.', $oldurlsearch);
$blogspot = stripos($oldurlsearch, '\.blogspot\.');
if ($blogspot) { //Blogspot addresses can be international e.g. myblog.blogspot.com, myblog.blogspot.com.au or myblog.blogspot.co.uk or myblog.blogspot.de both resolve to the same blog.
//See http://www.searchenginejournal.com/google-blogger-url-censorship/39724/
$oldurlsearch = substr($oldurlsearch, 0, $blogspot + 12).
'[\w\.]{2,6}';
}
$loadedposts = get_posts(array('meta_key' => 'blogger_blog', 'meta_value' => $this->host, 'posts_per_page' => $batchsize, 'offset' => $postsprocessed, 'post_status' => array('draft', 'publish', 'future')));
//Stop if nothing left
if (count($loadedposts) == 0) {
return true;
}
foreach($loadedposts as $importedpost) {
$importedcontent = $importedpost->post_content;
$regexp = '