Puma URL Resolver

Overview

The Puma URL Resolver provides a web service to the Puma Corpus Creator to retrieve publication full-text download URLs or directly the publication raw data.

The API provides a single access point where either a starting URL is provided or the DOI of the document is given. If a DOI is given then a starting URL is obtained from dx.doi.org.

The program will follow a URL and determine where it is redirected too: by doing this it is able to obtain a host and hence journal information.

Publication full-text is obtained through relevant APIs for individual journals.

Quick start

Building

# mvn clean
# mvn package 

Running

Set the following PUMA_RESOLVER_VM_ARGS environment variable for your own needs (eg proxy settings if needed)

# export PUMA_RESOLVER_VM_ARGS='-Ddw.server.applicationConnectors[0].bindHost=127.0.0.1 -Ddw.server.applicationConnectors[0].port=8020 -Ddw.server.adminConnectors[0].bindHost=127.0.0.1  -Ddw.server.adminConnectors[0].port=8021 -Dhttp.proxyHost=your.proxy -Dhttp.proxyPort=proxyPort -Dhttps.proxyHost=your.proxy -Dhttps.proxyPort=proxyPort'  

and then run the included script

# ./run.sh 

Usage

The main URL to obtain full-text download URLs on your local machine (using the port paramters above) sending a DOI is

http://localhost:8020/api/v1?doi=10.1021/cg5000205

Alternatively you can specify a known URL

http://localhost:8020/api/v1/?url=http://dx.doi.org/10.1021/cg5000205

If the relevant journal resolver has been implemented this will return you either the download URL or the full-text raw data.

If it has not been implemented you will receive a response indicating the host where the publication can be found and an error message indicating that the resolver is not (currently) supported.

Integration into Puma Corpus Creator

The Puma Corpus Creator (PCC) will automatically use the Puma URL Resolver if the relevant environment variable has been set.

Before running the Puma Corpus Creator set the following environment variable (assuming you are running the URL resolver on port 8020):

 # export PUMA_PCC_RESOLVER_URL='http://localhost:8020/api/v1'  

When importing documents, PCC can be set to automatically download full-text: if this environment variable is set PCC will attempt to obtain a URL automatically for each document and then download the relevant full-text.

If the environment variable is not set or the Puma URL Resolver does not respond then full-text data will not be obtained.