<?php

/**
*
* a simple RSS aggregator for Wordpress
* 
* more info: http://laughingmeme.org/
* author: kellan@protest.net
* license: GPL
* version: 0.02
*
* INSTALL:
*   1. drop in <yourblog>/wp-content
*   2. drop a copy of MapgieRSS into <yourblog>/wp-content
*   3. update wp-agg.ini
*   4. install a cron job to run periodically,
*   5. alternately browse to http://<yourblog>/wp-content/wp-rss-aggregate.php
*/

define('MAGPIE_DIR', 'magpierss/');

require_once MAGPIE_DIR.'rss_fetch.inc';
require_once MAGPIE_DIR.'rss_utils.inc';
require_once '../wp-config.php';
require_once '../wp-includes/wp-db.php';

$feeds = parse_config();

$total_new = 0;
foreach ($feeds as $f) {
	$new_count = feed2wp($wpdb, $f);
	$total_new = $total_new + $new_count;
}
echo "Added $total_new new posts";

function feed2wp($wpdb, $f) {
	$feed = fetch_rss($f['url']);
	$new_count = 0;
	foreach ($feed->items as $item) {
		$post = item_to_post($wpdb, $item, $f);
		// echo "<pre>"; var_dump($post); echo "</pre>";
		$new = add_post($wpdb, $post);
		if ( $new ) { $new_count++; }
	}
	return $new_count;
}

function item_to_post($wpdb, $item, $f) {
	$post = array();
	$post['post_title'] 	= $wpdb->escape($item['title']);
	$post['post_author'] = ($item['dc']['creator']) ? $item['dc']['creator'] : $item['dc']['contributor'];
	$post['post_author'] = author_to_id($wpdb, $post['post_author']);
	
	if ( isset($item['content']['encoded']) and $item['content']['encoded'] ) {
		$post['post_content']   = $wpdb->escape($item['content']['encoded']);
	} else {
		$post['post_content'] 	= $wpdb->escape($item['description']);
	}
	
	$post['post_name'] 	 = sanitize_title($post_title);
	$epoch 	= parse_w3cdtf($item['dc']['date']);
	$date 	= date('Y-m-d H:i:s', $epoch);
	$gmt 	= gmdate('Y-m-d H:i:s', $epoch);
	$post['post_date'] = $date;
	$post['post_modified'] = $date;
	$post['post_date_gmt'] = $gmt;
	$post['post_modified_gmt'] = $gmt;
	$post['wp_agg_rss_id'] = $item['link'];
	
	$post['cats'] = $f['cats'];
	
	if ( $item['dc']['subject'] ) {
		if ( strpos($f['url'], 'del.icio.us') !== false ) {
			$post['cats'] = array_merge($post['cats'], explode(' ', $item['dc']['subject']));
		}
		else {
			$post['cats'][] = $item['dc']['subject'];
		}
	}
	
	return $post;
}

function add_post($wpdb, $post) {
	$result = $wpdb->get_var(
		"SELECT meta_id FROM $wpdb->postmeta 
		WHERE meta_key='wp_agg_rss_id' AND meta_value='$post[wp_agg_rss_id]'");
	
	// if we've already added this item, then skip	
	if ($result) {
		return false;
	}
	
	$postquery = sprintf(
		"INSERT INTO $wpdb->posts (
			ID, post_author, post_date, post_date_gmt, post_content, 
			post_title, post_name, post_modified, post_modified_gmt)
		VALUES ('%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s')",
			'0', $post['post_author'], $post['post_date'], 
			$post['post_date_gmt'], $post['post_content'], $post['post_title'], 
			$post['post_name'], $post['post_modified'], $post['post_modified_gmt']
	);
	
	error_log("post:$postquery");
	
	$result = $wpdb->query($postquery);
	
	$postId = mysql_insert_id ($wpdb->dbh);
	$cat_ids = lookup_categories($wpdb, $post['cats']);
	add_to_category($wpdb, $postId, $cat_ids);
	add_rss_id($wpdb, $postId, $post['wp_agg_rss_id']);
	return true;
}

# 

function add_to_category($wpdb, $postId, $post_categories) {
	if (!$post_categories) $post_categories[] = 1;
	foreach ($post_categories as $post_category) {
		// Double check it's not there already
		$exists = $wpdb->get_row(
			"SELECT * FROM $wpdb->post2cat 
			WHERE post_id = $postId AND category_id = $post_category");

		 if (!$exists) { 
			$wpdb->query("
			INSERT INTO $wpdb->post2cat
			(post_id, category_id)
			VALUES
			($postId, $post_category)
			");
		}
	}
}

function add_rss_id($wpdb, $postId, $id) {
	$result = $wpdb->query("
		INSERT INTO $wpdb->postmeta 
		(post_id,meta_key,meta_value) 
		VALUES ('$postId','wp_agg_rss_id','$id')");
}

// look up (and create) an author
function author_to_id($wpdb, $author) {
	$id = $wpdb->get_var(
		"SELECT ID from $wpdb->users WHERE 
			user_login = '$author' OR
			user_firstname = '$author' OR
			user_nickname = '$author' OR
			user_description = '$author' OR
			user_nicename = '$author'");
	if ($id) {
		return $id;
	}
	$wpdb->query(
		"INSERT INTO $wpdb->users 
			(ID,user_login,user_firstname,user_nickname) 
		VALUES ('0', '$author','$author','$author')");
	$authorId = mysql_insert_id ($wpdb->dbh);

	return $authorId;	
}


// look up (and create) category ids from a list of categories
function lookup_categories ($wpdb, $cats) {
	if ( !count($cats) ) {
		return array();
	}
	
	# i'd kill for a decent map function in PHP
	# but that would require functiosn to be first class object, or at least
	# coderef support
	$cat_strs = array();
	foreach ( $cats as $c ) {
		$c = $wpdb->escape($c); $c = "'$c'"; 
		$cat_strs[] = $c;	
	}

	$cat_sql = join(',', $cat_strs);
	$sql = "SELECT cat_ID,cat_name from $wpdb->categories WHERE cat_name IN ($cat_sql)";
	$results = $wpdb->get_results($sql);
	
	$cat_ids 	= array();
	$cat_found	= array();
	foreach ( $results as $row ) {
		
		$cat_ids[] = $row->cat_ID;
		$cat_found[] = $row->cat_name;
	}
	
	$cat_unknown = array_diff($cats, $cat_found);
	if ( count($cat_unknown) ) {
		$sql =  "INSERT INTO $wpdb->categories (cat_name, category_nicename) 
			VALUES ('%s', '%s')";
			
		foreach ( $cat_unknown as $new_cat ) {
			$nice_cat = sanitize_title($new_cat);
			$wpdb->query(sprintf($sql, $new_cat, $nice_cat));
			$cat_ids[] = mysql_insert_id ($wpdb->dbh);
		}
	}
	return $cat_ids;
}

# use ini files because i'm lazy
#
# example config.  source name is ignored
# cats are split on ':'
#
# [source name]
# url = http://laughingmeme.org/index.rdf
# cats = LM
#
# [source name2]
# url = http://del.icio.us/tag/coffee
# cats = addiction:caffeine:seattle


function parse_config($file='wp-agg.ini') {
	$parsed = parse_ini_file($file, true);
	$feeds = array();
	foreach ($parsed as $name => $sec) {
		if (!$sec['url']) { continue; }
		$sec['cats'] = explode(':', $sec['cats']);
		$feeds[] = $sec;
	}
	return $feeds;
}


?>