NDEx Search REST Service

Provides integrated search REST service using NDEx as a backend. This service runs using an embedded Jetty server and is invoked from the command line.

Requirements

• Centos 7+, Ubuntu 12+, and most other linux distributions should work
• Java 11+ (jdk to build)
• Make (to build)
• Maven 3.6 or higher (to build)

Dependencies deployed on NRNB Nexus

• ndex-enrichment-rest
• ndex-java-client

Building NDEx Search REST Service

Commands below build NDEx Search REST Service assuming machine has Git command line tools installed and above Java modules have been installed:

# In lieu of git one can just download repo and unzip it
git clone https://github.com/cytoscape/ndexsearch-rest.git

cd ndexsearch-rest
mvn clean test install

The above command will create a jar file under target/ named
ndexsearch-rest-<VERSION>-jar-with-dependencies.jar that is a command line application

Running NDEx Search REST Service

The following steps cover how to configure and run integrated search REST service. In the steps below ndexsearch.jar refers to the jar created previously named ndexsearch-rest-<VERSION>-jar-with-dependencies.jar

Step 1 Create directories and configuration file

# create directory
mkdir -p searchdb/logs searchdb/tasks
cd searchdb

# Generate template configuration file
java -jar ndexsearch.jar --mode exampleconf > ndexsearch.conf

The ndexsearch.conf file will look like the following:

# Example configuration file for Search service


# Sets Search database directory
search.database.dir = /tmp

# Sets imageURL to use for results that lack imageURL
search.unset.image.url = http://ndexbio.org/images/new_landing_page_logo.06974471.png

# Sets Search task directory where results from queries are stored
search.task.dir = /tmp/tasks

# Sets HOST URL prefix (value is prefixed to Location header when query is invoked. Can be commented out)
# search.host.url = http://ndexbio.org

# Run Service under embedded Jetty command line parameters
runserver.contextpath = /
runserver.log.dir = /tmp/log
runserver.port = 8080
# Valid log levels DEBUG INFO WARN ERROR ALL
runserver.log.level = INFO

# Sets name of json file containing source results.
# This file expected to reside in search.database.dir directory
source.configurations = source.configurations.json
ndex.user = bob
ndex.password = somepassword
ndex.server = public.ndexbio.org
ndex.useragent = NDExSearch/0.8.0

Replace /tmp paths with full path location to searchdb directory created earlier.

Step 2 Create source.configurations.json file

This file (which resides in search.database.dir) contains information about services this integrated search and utilize.

Run the following to create an example source.configurations.json file:

java -jar ndexsearch.jar --mode examplesourceconfig > source.configurations.json

The source.configurations.json file will look like this:

{
  "sources" : [ {
    "name" : "enrichment",
    "description" : "This is a description of enrichment source",
    "endPoint" : "http://localhost:8095/enrichment/v1/",
    "uuid" : null
  }]
}

Each service under sources has various fields needed to access that service.

The uuid should be set to a unique value for each service.

Step 4 Run the service

jav -jar ndexsearch.jar --mode runserver --conf ndexsearch.conf

COPYRIGHT AND LICENSE

TODO

Acknowledgements

TODO