EDI-NG_client

EDI-NG (Next Generation) is a configurable HTML form page generator. Its purpose is to provide advanced forms with validation features in order to create geo-metadata.

Form pages are defined using XML files called templates. Once the form is filled in, it gets posted to a back end (namely EDI-NG_server, https://github.com/SP7-Ritmare/EDI-NG_server), which will create a metadata XML file based on the rules defined in the XML template file.

You can find plenty of template samples in the dist/templates directory. These template files are meant both as an example and as part of projects we are working at.

Please note that the metadataEndpoints sparqlEndpoints and datasource URIs point to our own endpoints, so you will need to customise them based on your own overall architecture

Creating a form based on a template file saved locally as "TEST_v1.00.xml" is as easy as using a javascript snippet like this inside a pure HTML page:

  $(window).load(function() {
      edi.loadLocalTemplate("TEST", "1.00", onTemplateLoaded);
  });

Warning

Due to CORS policies, some browsers might not allow loading from local files (i.e. local templates) In this case load templates from http(s) as in:

 $(window).load(function() {
     edi.loadTemplate("TEST", "1.00", onTemplateLoaded);
 });

How to Cite

Please, when you use EDI-NG_client cite as:

Pavesi, F., A. Basoni, C. Fugazza, S. Menegon, A. Oggioni, M. Pepe, P. Tagliolato, and P. Carrara. “EDI – A Template-Driven Metadata Editor for Research Data.” Journal of Open Research Software - JORS 4 (October 25, 2016). doi: 10.5334/jors.106.

Quick Start

The easiest way for a hands-on trial of EDI-NG is to download the demo standalone JAR here. It includes both a demo client and server, and it requires JAVA 7+.

If you just want to run EDI-NG client, all you need to do is in the dist directory. You can then deploy directory contents to a Web Server of your choice, or access it via the filesystem.

The "templates" directory contains a few sample templates: you should customise them to point to your own EDI-NG_server installation. To do that, replace content of the

<metadataEndpoint>...</metadataEndpoint>

tag.

E.g., replace

<metadataEndpoint>http://edi.get-it.it/</metadataEndpoint>

with

<metadataEndpoint>http://localhost:8080/</metadataEndpoint>

if you are running your own EDI-NG_Server on your local machine.

All included templates have mainly two purposes:

• serving for our own projects
• serving as an example to make your own templates

In particular, you can prepare your own SPARQL endpoints to feed your templates' combo boxes or autocompletion fields. That is why you can assign a specific SPARQL endpoint to datasources in your templates and even have a default SPARQL endpoint for them. E.g., you can replace our SPARQL endpoint

<sparqlEndpoint>http://sp7.irea.cnr.it:8891/sparql</sparqlEndpoint>

with

<sparqlEndpoint>http://url.to.your.sparql.endpoint/</sparqlEndpoint>

The sample pages, meant to illustrate sample templates, have sample buttons on the upper left part of the page:

• "Save locally" ("save EDIML" in older versions) - saves current values of fields to your browser for later use
• "Load last version" ("load EDIML" in older versions) - fills in the fields from a set of values that had been previously saved by the "Save locally" button
• "Send metadata" - sends all values you have filled in, plus all default or calculated values the template specifies, to the EDI-NG_server that is designated as the , so that the corresponding XML metadata are generated; if the "ignore warnings" checkbox is checked, the values can be sent even in the presence of warnings.

Once the XML metadata are generated, a new button appears, "download generated XML", allowing you to save the XML to your local filesystem.

Advanced installation topics

This project is managed by means of the bower/grunt pair. With "bower" we keep dependencies while we use "grunt" to create and populate the working directories and the dist output folder.

Please refer to the bower.json for info about this project's dependencies. In order to use grunt, you'll need to run the following commands at the root directory of this project, to install grunt and needed grunt packages:

npm install grunt
npm install load-grunt-tasks
npm install grunt-contrib-copy
npm install grunt-contrib-concat
npm install grunt-contrib-uglify
npm install grunt-contrib-watch
npm install grunt-bower-concat

If you need to customise this software more deeply, on the other hand, you can do so by editing the bower.json file and then running

bower update

and

grunt

in sequence.

External dependencies

All external dependencies can be found in the bower.json, except for Google Code Prettify, which is used only for debug purposes, so unused most of the time.

Copyright information

Copyright (C) 2013:

Anna Basoni - IREA CNR, Mauro Bastianini - ISMAR CNR, Cristiano Fugazza - IREA CNR, Simone Lanucara - IREA CNR, Stefano Menegon - ISMAR CNR, Tiziano Minuzzo - ISMAR CNR, Alessandro Oggioni - IREA CNR, Fabio Pavesi - IREA CNR, Monica Pepe - IREA CNR, Alessandro Sarretta - ISMAR CNR, Paolo Tagliolato - IREA CNR, Andrea Vianello - ISMAR CNR, Paola Carrara - IREA CNR

Support contact

For support or suggestions you can use the GitHub Issue Tracker, or email fabio(at)adamassoft.it