A new bank is made available as SQL dumps in a remote server. You want to retrieve the related files to populate you own database. Here is a way of doing it.

This example is based on the Gene_Ontology database that requires a MySQL database.

The basic steps are :

1. Retrieve the files from the server (usually sql dumps)

protocol=ftp
server=ftp.geneontology.org

remote.dir=/godatabase/archive/full/latest
remote.files=go_.*\\.gz$
remote.excluded.files=.*\\.tar\\.gz$



2. Write the post-process that will populate the database

POST1=GOdatabase
GOdatabase.name=GOdatabase
GOdatabase.exe=GOdatabase.sh
GOdatabase.args=hostname username password
GOdatabase.desc=Populates the gene ontology database
GOdatabase.type=

And the script itself (that has to be located in the process directory) : 

cd $datadir/$dirversion/future_release/flat
DBNAME="Gene_Ontology"
SQLFILE=`find go_[[:digit:]]*-assocdb-data`
echo "CREATE DATABASE IF NOT EXISTS $DBNAME" | mysql --user=$2 --password=$3 --host=$1
cat $datadir/$dirversion/future_release/flat/$SQLFILE | mysql --user=$2 --password=$3 --host=$1 $DBNAME

3.  Launch BioMAJ.
   

Attached files :

• .properties file
• script file