Jython script to connect to VIVO via SDB and export a model.

export_vivo.py

"""
Jython script to connect to VIVO via SDB and export a model. 
Requires Jython 2.5 or later.  Download from: http://www.jython.org/downloads.html

Assumes the VIVO harvester is on your path.

- change the database connection information to point to your VIVO database. 
- change the model name to specify the VIVO model you would like to export. 
- change the output_file name to the location where you want to save the RDF.
- for larger models you will want to increase the memory available to Java. 
    e.g. jython -J-Xmx2048m export_vivo.py

"""

from org.vivoweb.harvester.util.repo import SDBJenaConnect

jc = SDBJenaConnect(
    'jdbc:mysql://localhost/vivo',
    'vivo',
    'pass',
    'MySQL',
    #class
    'com.mysql.jdbc.Driver',
    #layout
    'layout2',
    #modelname
    'http://vivo.school.edu/modelname'
    )

#Echo the model name. 
print jc.getModelName()

output_file = "export.rdf"

#Change this to exportRdfToFile(output_file, 'N3') to get a different serialization.
jc.exportRdfToFile(output_file)

jc.close()

export_vivo_models.py

"""
Jython script to connect to VIVO via SDB and export named graphs.
Requires Jython 2.5 or later.  Download from: http://www.jython.org/downloads.html

Assumes the VIVO harvester is on your path.

- change the database connection information to point to your VIVO database. 
- change the model names to specify the VIVO models you would like to export. 
- change the DIR name to the location where you want to save the RDF.
- for larger models you will want to increase the memory available to Java. 
    e.g. jython -J-Xmx2048m export_vivo_models.py
    e.g. jython -J-Xmx4096m export_vivo_models.py
"""

import os
from org.vivoweb.harvester.util.repo import SDBJenaConnect

DB_HOST = 'localhost'
DB = 'vivo'
DB_USER = 'user'
DB_PASS = 'pass'
#Directory for saving data
DIR = '/tmp/vivo'

def scrub_uri(name):
    #Clean up a HTTP URI so it can be used as a file name.
    return name.lstrip('http://').replace('/', '-')


#Add your list of models here.
models = [
        "http://vitro.mannlib.cornell.edu/default/vitro-kb-2",
        "http://vitro.mannlib.cornell.edu/default/vitro-kb-inf",
        "http://localhost/data/mynamedgraph",
]


print '-----Dumping---'
for model_name in models:
    #Echo the model name.
    jc = SDBJenaConnect(
        'jdbc:mysql://{0}/{1}?useUnicode=yes&characterEncoding=utf8'.format(DB_HOST, DB),
        DB_USER,
        DB_PASS,
        'MySQL',
        #class
        'com.mysql.jdbc.Driver',
        #layout
        'layout2',
        #modelname
        model_name
    )
    #Print model name for debugging
    print jc.getModelName()
    #File name for export
    fname = os.path.join(DIR, '%s.n3' % scrub_uri(model_name))
    jc.exportRdfToFile(fname, 'N3')
    jc.close()