Monthly Archives: February 2009

YQL environment files

Filed under feature, news

We’ve introduced a new env query parameter to both the YQL console and the web service. The env parameter expects a URL value that points to a plain text file anywhere on the web (including subversion repositories, github etc). The file should contain a list of use statements that specify all the open data tables that you’d like to be available to your YQL queries for that request. For example, if I want all of the New York Times open data tables hosted at javarants I would create a file like this:

use '';
use '';
use '';
use '';
use '';
use '';
use '';

…and serve it from somewhere on the web. If you want to try this new feature on some useful open data tables in the console, you can use a few of the ones currently in the github open data tables project like this:

Try it in the console!. If you look to the “table” list on the right of the console, you’ll see an extra 9 APIs and 30 tables to explore!

Changelog for build 911

Filed under changelog
  • Console and web service now support an “env” query parameter. This is a url that points to an environment/shell file that YQL will load. Currently this enables multiple tables to be used at once.
  • “Use” table no longer requires the “as” clause. When “as” is absent it will default to the final path segment filename minus the extension. e.g. use 'http://somewhere/service.api.xml; select * from service.api;
  • Show tables now shows where the open data definition was defined
  • Desc table format tweaked
  • Open data table paging fixes, including better support for none paging but flexible result sized services and fixed page models. Moved “matrix” to the paging parent element.
  • Offset bug fixed in table(offset,number) processing

Changelog for build 822

Filed under changelog
  • NOT LIKE added
  • MATCHES regex matching added
  • New USE verb for open data table importing
  • Support for open data table definitions

Open data tables added to YQL

Filed under news

We’re pleased to announce that our last update now enables developers can create, use, and share table definitions over for any public API on the web, in addition to the Yahoo! APIs we support today.

For example, one of the requests we had when using our weather table was how to get an ID that would work for an international location. We pointed people to the API, but that needed a URL to be formatted up in a specific way. Now you can make that a table and use it:

use '' as ws;
select * from weather.forecast where location in (select id from ws(1) where query='bonn, germany')

Try it.

While we believe this is a great start, it’s work in progress. We’re already adding more capabilities and adjusting the open data table XML syntax as we get feedback and go forward. You can read more about how to go about creating open data tables in our documentation.