Lucene indexes creation

From BioUML platform
Revision as of 11:40, 10 July 2013 by Tagir Valeev (Talk | contribs)

Jump to: navigation, search

Lucene indexes are necessary for Lucene search to work. Indexes are created for each database separately in databases collection and stored inside luceneIndexes subfolder of database folder on the disk.


Contents

Indexes in BioUML workbench

In BioUML workbench indexes are created automatically when you try to search the corresponding database:

Lucene indexes creation workbench.png

An indexes creation window will appear prompting you on which indexes you want to create. Upon pressing Ok button indexes recreation procedure will start. It may time a lot of time depending on the database:

Lucene indexes creation workbench process.png

After it finishes you may use the search.

Indexes in BioUML server

To create indexes in BioUML server you have to use separate helper application which can be launched using rebuild_indexes.sh (on Unix) or rebuild_indexes.bat (on Windows) script. Place this script into $SERVER_PATH folder (see BioUML server installation).

This application reads the configuration from rebuild_indexes.properties file, so you have to create one. It contains several rows in key=value manner. The following keys are defined:

user
user name (e-mail of the user to log in to BioUML server during indexes creation). This user must be a server administrator for the current server.
pass
password for the specified user
database
semicolon-separated list of databases to reindex
folder
name of indexes folder. Must be luceneIndex

An example of rebuild_indexes.properties file is shown below.

user=administrator@myserver.example.com
pass=secretAdministratorPassword
database=Ensembl;GO;Reactome
folder=luceneIndex

After this file is created, launch rebuild_indexes script and it will try to create the missing indexes. It's desired to shutdown the BioUML server for the indexes creation procedure.

Recreating indexes after the database update

When you update the database indexes will become obsolete. The easiest way to update them is to terminate BioUML, remove luceneIndexes subfolder and recreate the indexes using the procedure above.

Personal tools
Namespaces

Variants
Actions
BioUML platform
Community
Modelling
Analysis & Workflows
Collaborative research
Development
Virtual biology
Wiki
Toolbox