Group items tagged

 * Implements background tasks for * <a href="http://code.google.com/appengine/docs/java/overview.html">Google App * Engine for Java</a>, based on the * <a href="http://code.google.com/appengine/articles/deferred.html">Python 'deferred' * library</a>; simplifies use of the <a href="http://code.google.com/appengine/docs/java/taskqueue/overview.html"> * Task Queue Java API</a> by automatically handling the serialization and * deserializtion of complex task arguments.

to write large batch processing jobs without having to think about the plumbing details.

it is a very easy way to perform some operation on every single Entity of a given Kind in your datastore in parallel

What would you have to build for yourself if Mapper weren’t available? Begin querying over every Entity in chained Task Queues Store beginning and end cursors (introduced in 1.3.5) Create tasks to work with chunks of your datastore Write the code to manipulate your data Build an interface to control your batch jobs Build a callback system for your multitudes of parallelized workers to call when the entire task has completed It’s certainly not a trivial amount of work

You create a java project only with the shared files (NOT a GAE project). You DON'T copy any libraries into the lib folder. You should add the required third-party components into your main project. Then you rightclick on the shared project's *source* folder, go to "Build Path -> Remove from build path" item in the context menu. You may, by the way, rename the source folder, so that it has a name different from the default source folder name. Next you rightclick on the main project's folder, go to "Build Path -> Link source".

This article studies the proxy technique first and then turns to the Google AJAX Feed API method, giving you a chance to intermix Java™and JavaScript coding.

You need a service that, when given a feed URL, connects to that site, downloads its contents, and sends them back to the caller. (For a shell-line parallel, consider the wget or curl command.) You can do this many ways, and Listing 2 shows a simple way to accomplish the task. Because I decided to call my remote proxy RemoteProxy, the server-side class had to be called RemoteProxyImpl; Impl stands for "Implementation."

The SOP won't let your code get data from another site, but there's an exception: You can download and execute JavaScript code using the <script ... /> tag. If the code you download happens to include data and calls a function of yours that puts the data to good use, then you have managed to bypass the SOP. Here's the idea behind the Google AJAX Feed API: It uses the <script ... /> tag to call a Google site that works as a proxy. The remote site gets the feed data and returns it in the form of JavaScript code. The downloaded JavaScript code calls your function so you can process the incoming XML

I believe that Twig is the only library that can store objects with entire collections embedded as components. So the Columns could actually be stored in the same entity as the Table. This means that querying or reading Tables is _much_ faster. If the tables are read more than written this would be ideal. Docs are a bit light but basically you just define an embedded collection like this: class Table { @Key String name; @Component Collection<Column> columns; } and thats it! The columns are then stored as a multi-valued property so you can even query properties them like "show all tables with a column named 'age'".