geoNetwork

Introduction to technical details

María Arias de Reyna - GeoCat

Index

  • Search Engine
  • Web Map Viewer
  • Import/Export metadata
  • Medata edit based on templates
  • Harvesting
  • Roles and privileges
  • CSW
  • Configuration Basics

Search Engine

As a (geo)data portal, geoNetwork's main feature

is search through metadata

All data is stored on the database,

but it is not used for searching

Search is based on Lucene 4.9.0

Search Engine

Lucene folder path configurable

Configure the fields used on search on index-fields.xsl

Configure the fields exposed on index-fields.xsl

Available through the q API service

Web Map Viewer

Part of the basic User Interface

Allows users to preview data layers

Read-only (view-only)

Geo-related services can be added (like geocoding)

Import and Export metadata

Standard way: CSW-t

Asynchronous way: Harvesting

geoNetwork services:

Harvesting

Periodical jobs that copy data from other catalogs

Periodicity can be defined on simple cron format

Multiple formats on origin

Users will not be able to modify this records, but can duplicate them

Role and Privileges

geoNetwork works with four types of users:

  • Registered User
  • Editor
  • Reviewer
  • Administrator

Roles can be defined by group

Role and Privileges

The group All is a generic group for public data

Role and Privileges

Records can be batch moved to another user

CSW

GeoNetwork is compliant with the 2.0.2 version of the specification

http(s)://___/srv/$language/csw

CSW

GetCapabilities: general description of the service

DescribeRecord: information model supported by the service

CSW

GetRecordById: data of the record by identifier. Authentication optional

GetRecords: query the service using CQL or OGC Filter. Authentication optional

CSW

Harvest: Force a CSW harvesting (sync or async). Authentication required

Transaction: interface for creating, modifying and deleting catalogue records. Authentication required

Configuration Basics

Based on database settings table and xml files (config*.xml)

Latest versions rely more on Spring library

  • Authentication & Security (LDAP, Shibboleth, CAS)
  • Database connections
  • Definition of Privileges

Configuration Basics

Big catalogs (+100.000 records)

  • Use PostGIS index instead of Shapefiles
  • Tweak postgreSQL with more memory (spatial queries)

Large catalogs (+300.000 records)

  • Consider clustering

Huge catalogs (+1.000.000 records)

  • Check inodes (number of files)
  • Consider use SOLR instead of Lucene

The End

@geocat_bv