Web API to download, resize, and save images from flickr public images feed.

In the project's root folder run the command clj -m rflick.core to start Jetty Web Server on port 3000.

There is a single endpoint, http://localhost:3000, that is responsible for handling the requests.

• num-images

  Optional, integer between 1 and 20.

  Number of images to be downloaded, resized, and saved. Flickr's feed page returns 20 items, that is why the API limit is 20. If not provided, all images returned by the feed will be processed.

• width

  Required, integer.

  Width of the resized images.

• height

  Required, integer.

  Height of the resized images.

Empty body, status 202 Accepted.

The processing was implemented as a pipeline of steps connected by core.async channels. For each incoming request the pipeline is built, execution happens and the channels are closed.

The server is a simple ring handler augumented with a couple of middlewares, executed on Jetty via ring-jetty-adapter.

• Tests: tests where developed on the REPL but a test suite would be nice.
• Error handling: there is basic error handling that is used during network calls: the retrying high-order function. Places that could be improved with error handling are:
  • Writing resized image files to disk.
  • Parsing request query parameters.
• Functions visibility: some functions could be private but I decided to keep them public to make testing easier/faster.

Probably a lot more. :-)