Here’s a simple example to demonstrate Yahoo! Pipes:

  • Assuming that links on Hacker News are interesting…
  • …GitHub repositories mentioned on Hacker News must be awesome ;-)

If we can agree on that, here’s a pipe that does the job + an extra touch:

Here’s how it works:

  1. Take the RSS feed of Hacker News (
  2. Apply a filter: keep only items whose link starts with
  3. Extract the repository name component from the link and append it to the title to make it easier to see

The first 2 steps are admittedly nobrainers. But the last step is a bit tricky, and it takes two operations:

Extract repository name component and append to title

First, the Rename operator is to make a copy of in item.repo. Notice that in the new name field (after Copy As) you must NOT write the item. prefix, but simply the name “repo”.

Next, the Regex operator has two steps:

  1. In item.repo, replace the pattern https://([^/]+/[^/]+).* with $1, where $1 is a back reference to whatever was matched in (...) in the pattern. So for example if the original value is, it will be replaced with

  2. In item.title, replace $ with ` (${repo}), where ${repo} is the value of item.repo, which the previous step created. And replacing $ effectively means appending to the input string, in this case item.title`.

blog comments powered by Disqus