clj-icat-direct

A Clojure library for directly connecting to the iRODS ICAT database and running queries against it. Uses Korma to connect, which might be overkill right now, but provides connection-pooling and a way forward from simply running strings containing SQL against the ICAT.

See the LICENSE file for licensing information.

Testing from the REPL

This library is configured to add the path ./repl to the source file search path in REPL mode, so that developers can easily test using the Clojure REPL. To use this feature, first create two files: $HOME/.irods/.qa-db.json for QA, and $HOME/.irods/.prod-db.json for production. Note that these files will contain sensitive connection information for the ICAT database. Be sure to set the file permissions accordingly. Each file should look like this:

{
    "host": "icat.example.org",
    "port": 5432,
    "user": "somedbuser",
    "password": "somedbpassword"
}

Once you have the files in place, you can easily configure the REPL to access the ICAT database by calling init-prod or init-qa from the clj-icat-direct.repl-utils namespace:

user=> (require '[clj-icat-direct.repl-utils :as ru])
nil

user=> (ru/init-prod)
;; output omitted

After configuring the REPL, you can other functions in clj-icat-direct as usual:

user=> (require '[clj-icat-direct.icat :as icat])
nil

user=> (icat/number-of-items-in-folder "a" "example" "/example/home/a" :any)
25

The repl directory is also included in .gitignore so that you can add additional namespaces to it without having to worry very much about accidentally including them in a commit. This can be very helpful if you have an editor that can interactively run Clojure code.