gdata-books-clojure

Clojure friendly wrapper for Google's GData Books API. (javadocs)

Usage

If you are using [leiningen][leiningen], add gdata-books-clojure to your dependencies:

[gdata-books-clojure "1.0.0-SNAPSHOT"]

First, tell the library your name, your application's name, and version number, in "yourname-appname-1.0" format:

   (set-application-name "joeking-kewlapp-1.0")

Google uses this information to identify applications that are abusing the API.

Then you can search for books by title, or title and author:

 user> (search "Stranger in a Strange Land")
 #<VolumeFeed {VolumeFeed com.google.gdata.data.books.VolumeFeed@1d95643}>

 user> (search {:title "Stranger in a Strange Land", :author "Robert Heinlein"})
 #<VolumeFeed {VolumeFeed com.google.gdata.data.books.VolumeFeed@578b06}>

 user> (search {:title "Stranger in a Strange Land", :author "Heather L. Katz"})
 #<VolumeFeed {VolumeFeed com.google.gdata.data.books.VolumeFeed@13f54ae}>

search returns a VolumeFeed which represents an Atom feed for search results for the book. You can use the helper functions in gdata-books-clojure to tease out the pieces of information you want, without having to know the structure of the atom feed or the the java API:

user> (map creators (entries (search "Stranger in a Strange Land")))
(("Robert A. Heinlein") ("Gary Younge") ("Robert Anson Heinlein") ("Heather L. Katz") ...)

user> (map titles (entries (search {:title "Stranger in a Strange Land", :author "Heather L. Katz"})))
(("Stranger in a strange land" "Muslim political identity in the context of liberal democracy"))

user> (map titles (entries (search {:author "H Beam Piper"})))
(("Little Fuzzy") ("Space Viking") ("Uller Uprising") ("Time Crime") ("The Cosmic Computer") ...)

Using these building blocks, you could build really complicated queries:

  ;; Search for books with "Monkeys" then search for all books by the
  ;; authors of the monkey books
  (->> (search "Monkeys")
       entries
       (map creators)
       flatten
       distinct
       (map #(search {:author %})))

TODO

• Only the first 10 results are returned in the feed from search. I was thinking it might be fun to use some kind of lazy computation to hide the fact that the feeds are paged.
• I haven't really used much of the API for what I am trying to do. Some of this might not actually work the way I expect.
• I am still a little unsure of the right way to test stuff like this in clojure: in static languages, I would try to do to stub out the Google, but the tests I have seem to work OK when you have access to the internet, so...

License

Copyright (C) 2010 Chris Bilson

Distributed under the Eclipse Public License, the same as Clojure.