Christian Heilmann

Getting a list of Flickr photos by location and/or search term with a YQL open table

Displaying photos from Flickr can be daunting. The API needs authentication and the RSS, JSON or LOLcode output is very limited. The way around is using YQL and its Flickr tables. That way it is pretty easy to search Flickr:

select * from flickr.photos.search where text=”panda”

Try out the Panda search in the YQL console.

The output format has a lot of information in there, but sadly enough, not all. For example the real name of the owner or the description is missing. Therefore you need to go through yet another Flickr API to get the whole data set:

select * from flickr.photos.info where photo_id in(
select id from flickr.photos.search where text=”panda”
)

Guess what? You can also try this more detailed query in the console.

I’ve shown before how easy it is to display Flickr Photos retrieved that way:


The issue with that though is that it uses JavaScript and JavaScript may be turned off (think Blackberries). Of course you can do the same thing in PHP but I’d wager to say more people to JavaScript than PHP these days.

The main issue is that Flickr returns the photos in a pretty weird format and that you need a script like the one above to turn it into a simple HTML list.

The good news is that YQL with the execute command allows you to embed JavaScript in your open tables. That way you can write a table that does all the necessary transformations and returns the data as a simple list for immediate use:




select * from {table} where location=”london,UK”
Christian Heilmann
http://www.wait-till-i.com/2009/11/01/getting-a-list-of-flickr-photos-by-location-andor-search-term-with-a-yql-open-table
Searches Flickr by location and/or search term and returns an HTML list that can be immediately used in a mashup.





You’ll notice that while the E4X support is very powerful, it can be a bit confusing to look at on first sight. Once you got your head around though it becomes much cleaner that way.

You can use this table like any other open table via the use command in YQL:

use “http://github.com/codepo8/yql-tables/raw/master/flickr/flickr.photolist.xml” as flickr;
select * from flickr where text=”me” and location=”uk” and amount=20

try it in the console.

I’ve wrapped one more API in there – the Yahoo Geo API to determine a place from a name should you want to search by location. All in all you have three parameters in this open table – all of which are optional:

If you look at the table source, you can also see that I hard-wired the license of the photos to 4 which is CC-BY. So if you link the photos back to Flickr you both satisfied Flickr’s terms and the original photographer’s.

Now, the easiest way to use this output is by using YQL’s JSON-P-X output format. This is XML with a callback which returns a JSON object with the HTML as a string instead of a convoluted JSON object. See the JSON-P-X output here.

That way you can easily use it in JavaScript:


And also in PHP:

$url = ‘http://query.yahooapis.com/v1/public/yql?q=use%20%22http://github.com/codepo8/yql-tables/raw/master/flickr/flickr.photolist.xml%22%20as%20flickr;%20select%20*%20from%20flickr%20where%20text=%22me%22%20and%20location=%22uk%22%20and%20amount=20&format=xml&diagnostics=false’;
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
$output = curl_exec($ch);
curl_close($ch);
$output = preg_replace(‘/.*
‘,$output);
$output = preg_replace(‘/.*?>/’,’‘,$output);
$output = preg_replace(‘//’,’‘,$output);
echo $output;
?>

You can see both in action on the demo page.

So by using YQL open tables you can not only access complex APIs with YQL, but you could also write complete mashups in JavaScript and have them executed in a safe environment on Yahoo’s servers. Your end of the mashup is simply the display which could be a form that works with Ajax when JavaScript is available and renders a static page in PHP (or whatever other server-side language) when JavaScript is turned off. You only need to do one HTTP request – the rest is executed and cached on the YQL server farm – everybody wins.