Skip to content

A command line tool for provisioning and configuring the Retrieve and Rank Service and the Document Conversion Service.

License

Notifications You must be signed in to change notification settings

IBM-Watson/kale

Repository files navigation

Kale

kale is a command line tool for provisioning and configuring the IBM Watson Retrieve and Rank Service and the IBM Watson Document Conversion Service.

We expect kale to run on any modern Java, Java 6 or newer. We test kale on Ubuntu Linux, Red Hat Linux, Windows and MacOS.

Use: download and run

Download

Download the kale tool from our releases.

Setup and run

Recommendation: create a short alias to setup a command named kale.

On Linux, OSX and other Unix-style systems, the alias command can be used like this:

alias kale="java -jar /full/path/to/kale-1.8.1-standalone.jar"

And on Windows, the doskey command can be used like this:

doskey kale=java -jar C:\full\path\to\kale-1.8.1-standalone.jar $*

Now the kale command should be available. Try:

kale help

Example run

Here is an unedited run through of provisioning commands to prepare a Solr collection as a target for running the Data Crawler.

After downloading kale from releases I opened a terminal window. The $ is my command prompt. I typed each command shown on the lines starting with $. All other text is output from running kale.

$ alias 'kale=java -jar /Users/ba/Downloads/kale-1.8.1-standalone.jar'
$ kale login
Endpoint (default: https://api.ng.bluemix.net)?
Using endpoint 'https://api.ng.bluemix.net'
Username? [email protected]
Password?
Logging in...
Using org '[email protected]'
Using space 'dev'
Loading services...
Log in successful!

Current environment:
   user:                         [email protected]
   endpoint:                     https://api.ng.bluemix.net
   org:                          [email protected]
   space:                        dev

$ kale create space example

Space 'example' has been created and selected for future actions.

$ kale create document_conversion example-dc
Creating document_conversion service 'example-dc' using the 'standard' plan.
Creating key for service 'example-dc'.

Service 'example-dc' has been created and selected for future actions.

$ kale create retrieve_and_rank example-rnr
Creating retrieve_and_rank service 'example-rnr' using the 'standard' plan.
Creating key for service 'example-rnr'.

Service 'example-rnr' has been created and selected for future actions.

$ kale create cluster example-cluster
Creating cluster 'example-cluster' in 'example-rnr'.

Cluster 'example-cluster' has been created and selected for future actions.
It will take a few minutes to become available.

$ kale list services
Available services in the 'example' space:
   [standard] document_conversion service named: example-dc

   [standard] retrieve_and_rank service named: example-rnr
      Cluster name: example-cluster, size: free, status: NOT_AVAILABLE

Currently using the following selections:
   document_conversion service:  example-dc
   retrieve_and_rank service:    example-rnr
   cluster:                      example-cluster

$ kale list services
Available services in the 'example' space:
   [standard] document_conversion service named: example-dc

   [standard] retrieve_and_rank service named: example-rnr
      Cluster name: example-cluster, size: free, status: READY
         configs:
         collections:

Currently using the following selections:
   document_conversion service:  example-dc
   retrieve_and_rank service:    example-rnr
   cluster:                      example-cluster

$ kale create solr-config english
Creating configuration 'english' in 'example-rnr/example-cluster'.

Solr configuration named 'english' has been created and selected for future actions.

$ kale create collection example-collection
Creating collection 'example-collection' in 'example-rnr/example-cluster' using config 'english'.

Collection 'example-collection' has been created and selected for future actions.

$ kale create crawler-config

Created two files for setting up the Data Crawler:
    'orchestration_service.conf' contains document_conversion service connection information.
    'orchestration_service_config.json' contains configurations sent to the 'index_document' API call.

The two files created at the end of this run, orchestration_service.conf and orchestration_service_config.json can be dropped, unmodified, into a Data Crawler configuration to point the crawler run at the services we just created.

Development

Automated testing

We automatically run the unit tests on Travis-CI (on Ubuntu Linux) and on AppVeyor (on Windows). Our developers typically run tests on MacOS.

Preparation for build, test and run

Get the Leiningen tool, lein, from http://leiningen.org/, and put it on your PATH. On Mac OSX, brew install leiningen works well. Most Linux package managers do not do a good job of packaging Leiningen, so please do the direct installation from http://leiningen.org/.

Run tests

$ lein test

Or, to run the tests and produce a code coverage report:

$ lein cloverage

Development run of the tool itself

$ lein run <command>

Building a jar

$ lein uberjar
Compiling ...
...
Created .../kale/target/uberjar/kale-1.9.0-SNAPSHOT.jar
Created .../kale/target/uberjar/kale-1.9.0-SNAPSHOT-standalone.jar

Now you can run the tool with a simple java command line:

$ java -jar target/uberjar/kale-1.9.0-SNAPSHOT-standalone.jar

To make running kale easier, it's recommended that you create an alias (see instructions under "Setup and run").

Contributing

We are thrilled you are considering contributing to kale! Please read our guidelines for contributing.

Licensing

©️ Copyright IBM Corp. 2016

All code contained within this project repository or any subdirectories is licensed according to the terms of the MIT license, which can be viewed in the file LICENSE.

Open Source @ IBM

Find more open source projects on the IBM Github Page

About

A command line tool for provisioning and configuring the Retrieve and Rank Service and the Document Conversion Service.

Resources

License

Stars

Watchers

Forks

Packages

No packages published