Geo exif data in WordPress

Did you know that WordPress imports EXIF data from your images when they’re uploaded?

If your image contains certain information (aperture, shutter speed, ISO, focal length, camera, creation time, credit, copyright, caption, title), it is in the database and you can use it as I do on my photoblog, Matt Mullenweg uses on his photo pages, and as described on Blogging Tips in WordPress Gallery and EXIF and Mixing Gallery and Blog Posts.

I’ve been thinking for a while that it would be cool to geotag images and show their location on a map. I finally got around to doing it (with some help from my husband, thanks honey!). Here’s how…

Before I get started, this post is PHP heavy and I’m not really going to explain the elements it uses (otherwise it would take me an age to write, and you just as long to read). Other than the posts I’ll link to, tizag.com has a good introductory PHP tutorial.

Geotagging

There’s a few different devices that you can connect to your camera to add GPS data to your photos and some cameras can do it themselves. I found an Aperture plugin, Maperture, which allows you to add the location of your images via Google maps.

I found the EXIF specifications and discovered the EXIF classes to get a map location are GPSLatitude and GPSLongitude (which store degrees, minutes and seconds), and GPSLatitudeRef and GPSLongitudeRef (which store N or S and E or W).

Importing EXIF data to the WordPress database

The information in this section is better handled by a function in a more recent post Add EXIF metadata to WordPress. Unlike what’s outlined below, you don’t have to change core WordPress files with the updated method. The rest of this post is up to date, though.

The file that determines which information is imported from images is wp-admin/includes/image.php. At line 231 (in WP 2.7) is a function called wp_read_image_metadata. To get WP to import the GPS data I added some values to the array:

	$meta = array(
		'aperture' => 0,
		'credit' => '',
		'camera' => '',
		'caption' => '',
		'created_timestamp' => 0,
		'copyright' => '',
		'focal_length' => 0,
		'iso' => 0,
		'shutter_speed' => 0,
		'title' => '',
	);

became,

	$meta = array(
		'aperture' => 0,
		'credit' => '',
		'camera' => '',
		'caption' => '',
		'created_timestamp' => 0,
		'copyright' => '',
		'focal_length' => 0,
		'iso' => 0,
		'latitude' => 0,
		'latitude_ref' => '',
		'longitude' => 0,
		'longitude_ref' => '',
		'shutter_speed' => 0,
		'title' => '',
	);

Then I had to tell WordPress which EXIF data is to be added to the database as latitude, longitude and the references.

After the lines,

if (!empty($exif['ISOSpeedRatings']))
	$meta['iso'] = $exif['ISOSpeedRatings'];

I added,

if (!empty($exif['GPSLatitude']))
	$meta['latitude'] = $exif['GPSLatitude'] ;
if (!empty($exif['GPSLatitudeRef']))
	$meta['latitude_ref'] = trim( $exif['GPSLatitudeRef'] );
if (!empty($exif['GPSLongitude']))
	$meta['longitude'] = $exif['GPSLongitude'] ;
if (!empty($exif['GPSLongitudeRef']))
	$meta['longitude_ref'] = trim( $exif['GPSLongitudeRef'] );

It really just copies the method used for the other EXIF fields.

GPS EXIF format

The difference between the latitude and longitude EXIF, compared to the others (e.g., aperture and shutter speed), is that GPSLatitude and GPSLongitude are stored as 3 rationals, or 3 fractions, which represent degrees, minutes and seconds. WordPress has a function to convert fractions to decimals for the database, but because these have an array of three fractions, it was a little more complicated.

I decided to import them as fractions and divide them out later.

This is what the EXIF (and IPTC) info from the MySQL database looks like,

"image_meta";a:14:{s:8:"aperture";s:1:"8";s:6:"credit";s:0:"";s:6:"camera";s:9:"NIKON D50";s:7:"caption";s:0:"";s:17:"created_timestamp";s:10:"1228061663";s:9:"copyright";s:0:"";s:12:"focal_length";s:2:"50";s:3:"iso";s:3:"200";s:8:"latitude";a:3:{i:0;s:4:"33/1";i:1;s:8:"5237/100";i:2;s:3:"0/1";}s:12:"latitude_ref";s:1:"S";s:9:"longitude";a:3:{i:0;s:5:"151/1";i:1;s:7:"906/100";i:2;s:3:"0/1";}s:13:"longitude_ref";s:1:"E";s:13:"shutter_speed";s:5:"0.004";s:5:"title";s:0:"";}

Unless you’re familiar with viewing database info, it may look like a bunch of jibberish at first, but if you look through, you can see that latitude has the values 33/1, 5237/100 and 0/1, and latitude_ref is S. So, the latitude of where that photo was taken is 33° 52.37′ 0″ S. Likewise for the longitude, which is 151° 9.06′ 0″ E.

Calling the metadata from the database

As shown in WordPress Gallery and EXIF, you can get the information in the database using
$imgmeta = wp_get_attachment_metadata( $imgid );
$imgmeta['image_meta']['shutter_speed'], for example.

If I just use that for latitude and longitude, it returns the word “array” because the values are stored as an array. So, I converted the individual array values into slightly friendlier decimals.

I’m using the Thesis theme, so the following goes in my custom_functions.php file. If you’re using another theme you could put it in your theme’s functions.php (if your theme doesn’t have one, create one).

I copied the exif_frac2dec function from image.php (I’m not sure why, but trying to use that function from custom_functions.php didn’t work), and then applied the function to each value of the arrays.

function geo_frac2dec($str) {
	@list( $n, $d ) = explode( '/', $str );
	if ( !empty($d) )
		return $n / $d;
	return $str;
}

function geo_pretty_fracs2dec($fracs) {
	return	geo_frac2dec($fracs[0]) . '° ' .
		geo_frac2dec($fracs[1]) . '′ ' .
		geo_frac2dec($fracs[2]) . '″ ';
}

Then, in the loop that calls the exif info from the database, after the bit that says $imgmeta = wp_get_attachment_metadata( $imgid );, I defined where the latitude and longitude info comes from and ran it through the geo_pretty_fracs2dec function.

$imgmeta = wp_get_attachment_metadata( $imgid );

$latitude = $imgmeta['image_meta']['latitude'];
$longitude = $imgmeta['image_meta']['longitude'];
$lat_ref = $imgmeta['image_meta']['latitude_ref'];
$lng_ref = $imgmeta['image_meta']['longitude_ref'];

Using geo_pretty_fracs2dec($latitude). $lat_ref . ' ' . geo_pretty_fracs2dec($longitude) . $lng_ref outputs something that looks like 33° 52.37′ 0″ S 151° 9.06′ 0″ E.

Link location to Google maps

Google uses a full decimal latitude and longitude; you need to divide minutes by 60 and seconds by 3600 and add them all together. Google also uses negative values to determine South and West.

The following function extracts the array values, divides them accordingly, and adds them together.

function geo_single_fracs2dec($fracs) {
	return	geo_frac2dec($fracs[0]) +
		geo_frac2dec($fracs[1]) / 60 +
		geo_frac2dec($fracs[2]) / 3600;
}

After the other variable definitions from before,

$lat = geo_single_fracs2dec($latitude);
$lng = geo_single_fracs2dec($longitude);
if ($lat_ref == 'S') { $neg_lat = '-'; } else { $neg_lat = ''; }
if ($lng_ref == 'W') { $neg_lng = '-'; } else { $neg_lng = ''; }

The whole list item to call the location reference and link to the map looks like,

if ($latitude != 0 && $longitude != 0) echo '<li>Location: <a href="http://maps.google.com/maps?q=' . $neg_lat . number_format($lat, 6) . '+' . $neg_lng . number_format($lng, 6) . '&z=11">' . geo_pretty_fracs2dec($latitude). $lat_ref . ' ' . geo_pretty_fracs2dec($longitude) . $lng_ref . '</a></li>';

Where the if != 0 says to only do this stuff if the location is not equal to zero, because there’s no point sending people to the South Atlantic Ocean. Would be a problem if that’s where you actually were though… The z=11 bit is the zoom on the map, play around with it until it suits you. Everything else is PHP or stuff I mentioned before, but feel free to ask what stuff does if you don’t understand it.

This will work for other maps, such as Yahoo, you just need to find out what the URL structure for the other map is.

That’s it. My photoblog EXIF uses a combination of the above, the code at Blogging Tips and also some from A Moment of EnLitenment to show shutter speed as fractions.

Have a look at my complete EXIF code to see it all combined, and if you made it to the end of this post, congratulations!