From Gcube Wiki
Jump to: navigation, search

GeoPortal is a feature-complete framework enabling the publication, access and management of GIS projects consisting of multiple documents, images, and datasets. It can be configured with the XML specification of the GIS project data model.

This document outlines the design rationale, key features, and high-level architecture, the options for their deployment and as well some use cases.

Key features

GeoPortal key features are :

Support for publication lifecycle
By supporting complex Data (Meta + Payloads) archives;
By enableing versioning, workflows, access policies;
By supporting several materialisations (GIS, Databases, ...)
By managing indexes (Meta catalogues, Index GIS layers)
Maximise reusability
By exploiting space-time GeoPortal Service
By allowing for configurable behaviour;
By supporting a generic meta-model;
By offering configurable GUIs (Management grid, Insert/Edit Form, Data Viewers);
External Data Integration
By exploiting OGC standards.



Aim of the framework is to maximize adaptability by supporting custom model and behaviour definition.


GeoPortal Architecture

The framework provides the following Dynamic GUIs, which use the Profile metadata definition to enable tailored:

  • Data Collection : Form assisting users to publish GIS projects;
  • GIS Viewer : allowing any user to visualise projects on a map;
  • Project Viewer : assisting users in accessing information, documents, images and datasets associated with the GIS project.

Dynamic GUIs rely on java client libraries to communicate to the GeoPortal Service, managing validation and management of GIS projects. The service relies on the D4Science Workspace to store and access attached documents and on the D4Science SDI to offer OGC Compliant Services (e.g. WMS, WFS, WCS, etc.). It also uses an internal archive of Profiled Document to manage publication lifecycle and for querying purposes.

Extensible Model

Dataset schemata are defined as Profile, and every managed collection of documents refer to a specific Profile. A complex geo-temporal dataset linked to a Profile is called Profiled Document. This approach allows for the definition of tailored Profiles allowing the management of a heterogeneous collection of documents. Moreover, it supports evolution by intrinsically enabling the extensions of any Profile.

Profile and Profiled Document

Profiled Documents are stored as :

  • Meta: core section for publication management, containing:
    • Ownership
    • Identification
    • Versioning
    • Status
      • Phase
      • User-oriented messages
  • JSON Document metadata;
  • FileSets uploaded along with metadata;
  • Materialised Resources obtained by publishing FileSets, e.g.
    • GIS Layer
    • RDBMS Table
    • Indexing information (GIS centroids, Catalogues IDs ..)

Profiles define collections in a VRE, specifying :

  • Document structure
    • Fields (cardinality, type, constraint, defaults, indexing, GUI declaration)
  • Publication lifecycle handlers
    • Default values
    • Java model for specific business logic
    • Validators
    • Publishers
  • Handlers configuration

Use Cases

The gCube GeoPortal suite is designed to better serve publication management of complex data in Data e-infrastructures. Following examples are hypotetic situations in which the suite could be used in.

Well suited use cases

The suite is designed to easily serve situations like :

  • Managing publication of simple and complex documents in multiple platforms
  • Allow communities to define / reuse / adopt :
    • Shared Document workflow
    • Shared Document structure
    • Shared Collections and Indexes
    • Shared Document Materialization praxis

Less well suited use cases

Although the suite could be used in the following situations, they remain outside of the scope of the application for the time being :

  • Intensive processing of datasets
  • Complex workflows

Case Studies