Pipes Translate Module

Earlier this year http://babelfish.yahoo.com was shut down. The Pipes translate module relied on this service and has also been deprecated.

Pipe developers have been looking for a solution to this gap. One of the available free solutions we’ve been using is the Microsoft Translate API with Pipes.

First you need to sign up with Microsoft to get your translate key and username.

When filling out their form you can enter your domain or pipes.yahoo.com in the “Redirect URI” field. (E.g. http://pipes.yahoo.com)

The only 2 things you need to remember is the Client ID (which you assign) and the Client secret (which they generate).

You can do that here: https://datamarket.azure.com/developer/applications/

After you set your clientid and get your client secret, clone this Pipe: http://pipes.yahoo.com/pipes/pipe.info?_id=c33dfcb6228fcf820afd1bac96833d4c and add it to your favorites by “staring” it.

When building your Pipe you can access it via the Favorites folder in the Pipes Editor.

You can see from this Pipes example how to use the Translate Pipe to fine tune what you want your Pipe to translate: http://pipes.yahoo.com/pipes/pipe.edit?_id=ac325bbbc6c5603c8c8e33dd26fe3fc3

Here i’m putting the Bing Translate Pipe into a Loop and translating the RSS feed titles from Spanish to English.

You can clone and then modify the Pipe to your liking.

Remember the free version of Microsoft Translator will give you 2 million characters of translation per month.

For developers familiar with YQL there is also a YQL Microsoft Translate table available here. As with Pipes, you’ll need to sign up for the client ID and client secret from Microsoft.

Posted in news | Comments Off

New Module: XPATH Fetch Page

We are introducing a new module, the XPATH Fetch Page.

We are also going to deprecate the Fetch Page module at the end of June. So please convert your existing Pipes that use the Fetch Page module to the XPATH Fetch Page module.

To use the XPATH Fetch Page module, first enter the URL of the site you want. By default, the module will output the DOM elements as items in the preview pane. You can optionally use the “Emit items as string” checkbox if you need the html as a string.

You can use the “Extract using XPATH” field to fine tune what you need from the HTML Page. For example, if I want all the links in the page I can simply use “//a” to grab all links. If I want all the images in the html I can do “//img”. Read more on XPATH. You can also find XPATH statements using firebug/developer tools to target data that you want in a HTML page.

Previously with the older Fetch Page module you would have to wrangle with regex, splits and other complicated methods to get the data you wanted. The new XPATH Fetch Page module is more powerful, easier and more inline with todays standards.

Currently this module will extract the page and fix malformed tags using Tidy. You have the option to run the parser using support for HTML4 (by default) or checking the “Use HTML5 parser” checkbox to use the HTML5 parser. We recommend using the HTML5 parser when using this module for most cases.

Click here for an example Pipe using the new XPATH Fetch Page module.

You can also use Pipes special variable substitution method (e.g ${<dom node path here>}) to construct new content from the dom nodes. For example:

${td.0.span.0.a.content} will pull content from that dom path, viewed in the preview pane.

and a longer example:

Company name / Ticker: ${td.0.a.content} ${td.0.p}<br>Underwriter: ${td.1.p.content.0} ${td.1.p.content.1}<br>Price Range: ${td.2.p.content}<br>Shares: ${td.3.p}<br>Pricing Date: ${td.4.p}

You can use the Regex module to help you build this new content. View an example that uses this method here.

Note on usage: The module will only fetch HTML pages under 1.5MB and the page must also be indexable (e.g. allowed by the site’s robots.txt file.) If you do not want your page made available to this module, please add it to your robots.txt file, or add the following tag into the page’s element:

<meta name=”robots” content=”noindex”>

Posted in news | Comments Off

Using Yahoo! Pipes and the YQL Module

In this talk from YUIConf 2011, YQL engineer Paul Donnelly demonstrates the features of the Yahoo! Pipes editor and explains how you can use Pipes and YQL to power your web apps, create mashups, and more.

(original posting here: http://www.yuiblog.com/blog/2012/01/11/video-yuiconf2011-donnelly/)

Posted in news | Comments Off

Pipes Changelog 2.0.10

  • Frontend Changes
    • String builder, Item builder and URL builder modules now have a max of 40 item per module, increased from 10.
    • In the Pipes Editor, in folders “My Pipes” and “Favorites”, the sort order is now based on last modified Pipe.
  • Engine Changes
    • Date Formatter Module can use Java SimpleDateFormat Patterns (http://y.ahoo.it/qSAA+)
    • SubPipe using default text bug fixed. (http://y.ahoo.it/8O7kH)
    • Some atom feeds not getting y:published emitted – fixed (note: if building own data (item builder, fetch data) and you want y:published emitted – run data through the “Create RSS” Module or “RSS Item Builder” module. y:published is needed to emit a valid RSS pubDate element in final Pipes RSS output)
Posted in changelog | Comments Off

Pipes Changelog 2.0.9

  • Renamed Module “Fetch Site Feed” to “Find First Site Feed“. Changed due to confusing name, which resulted in users using “Fetch Site Feed” instead of “Fetch Feed” – http://y.ahoo.it/uaPda
  • Removed Alert icon in all modules that access a url. Confused users thinking the feed was invalid or unreachable. – http://y.ahoo.it/IoPXn
  • Fixed Namespaced elements issues – http://y.ahoo.it/7x6PZ
  • Fixed Subsequent Split module increases Truncate value – http://y.ahoo.it/pDyu0
  • Fixed Pipe name alias redirect issue: http://y.ahoo.it/yoEsT
Posted in changelog | Comments Off

Pipes DB write outage.

Currently our cloud DB service has an outage that effects writes to the service. Reads are operational and working.

This means creating new pipes or editing existing Pipes will not work. But existing Pipes will still run and produce output as normal.

You can access you Pipe output by entering:

http://pipes.yahoo.com/pipes/pipe.run?_id=<yourpipeid>& _render=rss/json/kml…etc.

e.g. http://pipes.yahoo.com/pipes/pipe.run?_id=ZKJobpaj3BGZOew9G8evXg&_render=json

Posted in news | Comments Off

Pipes Badges – Release Note

Last week, we released upgraded Pipes Badges to be compatible with YUI3. However, there was a slight hiccup and the versions of the List Badge, Map Badge, and Image Badge script files that were produced are problematic. If you recently updated or added badges to your personal site and are experiencing issues, please look at your source code and make sure you are not using the following versions of these files:

  • http://l.yimg.com/a/i/us/pps/imagebadge_1.4.js
  • http://l.yimg.com/a/i/us/pps/listbadge_1.5.js
  • http://l.yimg.com/a/i/us/pps/mapbadge_1.4.js

If you are using any of those versions, please update them to be one of these instead:

  • http://l.yimg.com/a/i/us/pps/imagebadge_1.5.js
  • http://l.yimg.com/a/i/us/pps/listbadge_1.6.js
  • http://l.yimg.com/a/i/us/pps/mapbadge_1.5.js

We fixed the pipe.info page, so you should be getting proper badge code snippets moving forward.

Posted in changelog | Comments Off

Pipes Changelog

Posted in changelog | Comments Off

Flickr Module

With the next release(TBD) we will require any Pipe that uses the Flickr Module to use an API key. You can obtain the Flickr API key here: http://www.flickr.com/services/apps/create/apply/. It’s easy to get a key and please get it ready for when this change happens. You can also use the Private String Module to make sure your API key is not viewable to the public as described in the screenshot below.

Flickr Module

Leave a comment

Webplayer integration

We recently integrated webplayer.yahoo.com into the pipe.info pages when viewing the “List” view.

For more info on webplayer and Pipes please read: http://developer.yahoo.com/blogs/ydn/posts/2011/08/yahoo-pipes-webplayer-awesome/

Leave a comment