On a WordPress site I’ve been working on I’m using SlideDeck as a feature article slider on the homepage.  The only problem was I have additional articles populating underneath it and the articles from the SlideDeck were showing up down there as well.

We didn’t want that behavior so I researched how to filter out SlideDeck posts from the post loop.  At this time, it’s apparently easy to filter for content that has a custom meta-tag, but you can’t easily filter for content that doesn’t have a custom meta-tag applied to it.  It’s because when a post doesn’t have a custom met-tag attached to it that custom meta-tag doesn’t come up as “null” or “0”  it simply doesn’t come up at all.  So there’s no way to use query_posts to filter out SlideDeck posts, you have to make custom database query and use $wpdb->get_results.

This post from SlideDeck’s customer support website did most of the heavy lifting for me in putting together the exact query I needed to filter out only SlideDeck posts.  The only problem was that this gave me no insight as to how I could paginate the posts.  Thankfully google came to rescue once again and I found this post and this post which helped me piece together the script.  I posted my findings on SlideDeck’s customer service website, but I’m posted it here as well so I can easily find it again if I need it.

So for the code, inside loop.php before have_posts() I put…

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

if(is_home()) {
    //query_posts(array( 'meta_key' => 'slidedeck_post_featured') );
    global $wpdb;
    $ppp = intval(get_query_var('posts_per_page'));
 
    $paged = (get_query_var('paged')) ? get_query_var('paged') : 1;
    $offset = ($paged-1) * $ppp;
    $sql = "SELECT p.*
        FROM {$wpdb->posts} AS p
        LEFT JOIN {$wpdb->postmeta} AS pm
        ON pm.post_id = p.ID AND pm.meta_key = %s
        WHERE p.post_status = %s AND p.post_type = %s AND pm.meta_value IS NULL
	ORDER BY p.post_date desc
        LIMIT $offset, $ppp";
    $posts = $wpdb->get_results($wpdb->prepare( $sql, '_slidedeck_post_featured', 'publish', 'post' ));
}

if(is_home())  will make it so it will only filter out SlideDeck posts on the homepage, and apparently the have_posts() works exactly as it normally does even when I change the loop using $wpdb->get_results, which means I don’t have to do much to the rest of the page in order for this to work.  WordPress, you rock <3

This will paginate the posts according to the number of items I have set in the WordPress settings, but for some reason when I did this on the last page it printed out empty posts to fill out that page.

For instance, say I only have 7 posts left on the last page, but I have it set to show a maximum of 10 posts per page – it would print out 3 empty posts on that page.

To combat this I put the following code inside have_posts() so it wouldn’t print anything if the post ID was null.

1
2
3

if(get_the_ID() > 0) {
    //html for posts
}

Pretty straight forward and everything has been working great so far 🙂

References

http://support.slidedeck.com/slidedeck/topics/wp_query_a_list_of_posts_not_featured_in_the_slide_deck?from_gsfn=true

http://wordpress.org/support/topic/custom-loop-pagination
http://www.scriptygoddess.com/archives/2011/02/23/wordpress-pagination-woes-solved-i-hope/

For a project we wanted to bring in outside links and display them in a box on the homepage.  Instead of building out a whole custom system to handle the link, I suggested posing them through Facebook.

So in order to be able to style the feed however we wanted I put together a simple Curl code in PHP to pull in the Facebook Page RSS feed (every page has a feed it’ll be formated like this => http://www.facebook.com/feeds/page.php?id=245091913103&format=rss20).

This will check if there’s a link inside the body of the status update and use that, but then default to the link to the status message in Facebook if that doesn’t exist.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

<?php
        $curl = curl_init();
        curl_setopt($curl, CURLOPT_URL, "http://www.facebook.com/feeds/page.php?id=245091913103&format=rss20");
        curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
        curl_setopt($curl, CURLOPT_CONNECTTIMEOUT, 0);
        curl_setopt($curl, CURLOPT_FOLLOWLOCATION, 1);
        curl_setopt($curl, CURLOPT_USERAGENT, "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.3) Gecko/20070309 Firefox/2.0.0.3");
        $xmlTwitter = curl_exec($curl);
        curl_close($curl);
 
        $xmlObjTwitter = simplexml_load_string($xmlTwitter);
        $tempCounter = 0;
 
        foreach ($xmlObjTwitter->channel->item as $item) {                    
            if ( $tempCounter < 9 ){
                preg_match('/(http|https|ftp)(.*?) (id|target)/i', $item->description, $matches);
                $matches[0] = str_replace('" id','',$matches[0]);
                echo '<li><a href="'.($matches[0] ? $matches[0] : $item->link).'" target="_blank">'.$item->title.'</a>
                <div>'.date('F jS, Y',strtotime($item->pubDate)).'</div></li>';
            }
 
            $tempCounter += 1;
        }
 
        echo '</ul>';
?>

So, I’ve just spent the last half hour tracking down this annoying bug in WordPress 😛

I’m really glad I use an outdated theme (or at least one without custom header support) so I could track this puppy down.

I have been working on building a theme that is a child theme of good ‘ol 2010 (2011 wasn’t out when I started building :()

I uploaded a portrait image and noticed it got cut off in the image editor like so:

The editor’s will really need to be able to adjust that thumbnail image, so I went digging around and found that a theme called “Modfolio” had this exact same problem.  So, I figure, it’s gotta be something with the theme.  Especially since I didn’t get a whole lot of results…if it was a core issue and everyone was seeing this then I would assume there would be a lot more complaints.

I switched around from my child theme, to 2010, and even 2011, but the same error kept popping up :/  Then I figured I’d check it out in my blog here.  I’m currently using a slightly modified version of Uchilla theme – which hasn’t been updated in ages.  And lo-and-behold the image is coming up fine and dandy in the editor.  I could be wrong on this, but I believe the only file in a theme that can mess with the admin side of things is the functions.php file.  So I just kept hacking away at it until I found the bit of code that was causing me grief.

It’s lines 117 & 188 in the functions file for 2010:

define( 'HEADER_IMAGE_WIDTH', apply_filters( 'twentyten_header_image_width', 940 ) );
define( 'HEADER_IMAGE_HEIGHT', apply_filters( 'twentyten_header_image_height', 198 ) );

That was somehow adding a constraint to the editor preview and for me simply commenting out those two lines did the trick 🙂  I did have to re-upload all the affected images though…

I don’t know what effect that might have on the custom headers…I’m not using them myself so it wasn’t an issue for me.

I recently added in a way for admin users to download a backup of the database to a CMS I work on.  I came up with something between this solution and this one.  Add this little snippet after a database connection and change “table_1” and “table_2” to whichever tables you’d like to output, and add as many tables as you need.

What I like about this script is that it doesn’t save a copy to the server so folks can’t nab it if they know the entire URL, but it can be easily integrated to a web based interface.  It also uses “create table if not exists” which was missing from the other two scripts.  I think a backup should create tables in case you loose the table structure, but it shouldn’t delete the existing data.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

<?php
function datadump ($table) {
    $result .= "# Dump of $table \n";
    $result .= "# Dump DATE : " . date("d-M-Y") ."\n\n";
    
    $row2 = mysql_fetch_row(mysql_query('SHOW CREATE TABLE '.$table));
    $result .= "\n\n".str_replace('CREATE TABLE','CREATE TABLE IF NOT EXISTS',$row2[1]).";\n\n";
    
    $query = mysql_query("select * from $table");
    $num_fields = @mysql_num_fields($query);
    $numrow = mysql_num_rows($query);
    
    for ($i =0; $i<$numrow; $i++) {
        while($row = mysql_fetch_row($query)) {
            $result .= "INSERT INTO ".$table." VALUES(";
            for($j=0; $j<$num_fields; $j++) {
                $row[$j] = addslashes($row[$j]);
                $row[$j] = ereg_replace("\n","\\n",$row[$j]);
                
                if (isset($row[$j])) $result .= "\"$row[$j]\"" ; else $result .= "\"\"";
                if ($j<($num_fields-1)) $result .= ",";
            }
            $result .= ");\n";
        }
    }
    return $result . "\n\n\n";
}
 
$content = datadump('table_1').datadump('table_2');
 
$file_name = "database_backup.sql";
Header("Content-type: application/octet-stream");
Header("Content-Disposition: attachment; filename=$file_name");
echo $content;
?>

I recently just finished a project called “Twitter Critter” It’s a simple program that works with Twitter to generate an avatar.  All you have to do is enter in your own username, or anyone’s username, and it will create an avatar based on your Tweets.

It searches through your latest 200 tweets and uses those words to determine what color, critter, and what accessories your Twitter Critter will have.  It also generates a cute bio for that Critter based on those choices.  There are thousands of possible combinations, and if you come back after doing some Tweeting your result will change!  You can also Tweet out a permalink to the specific Twitter Critter your generated, and with one click you can set it as you avatar on Twitter.

Click here to get started!  http://twitter-critter.contropa.com/

Related Blogs:

http://rakugakiman.blog6.fc2.com/blog-entry-2827.html