Orientdb driver for python that uses the binary protocol.
Pyorient works with orientdb versions from 1.7 to 2.x.
Warning Some issues are experimented with record_create/record_upload and OrientDB < 2.0. These command are strongly discouraged with these versions
NOTICE Prior to version 1.4.9 there was a potential SQL injection vulnerability that now is fixed. (see details , details )
pip install pyorient
OrientDB PyOrient Python Driver
ant
and nose
installed properly./ci/start-ci.sh
from project directorynosetests
Proper documentation will be available soon, for now you have to read the tests.
PyOrient is composed of two layers. At its foundation is the python wrapper around OrientDB's binary protocol. Built upon that - and OrientDB's own SQL language - is the Object-Graph Mapper (or OGM). The OGM layer is documented separately.
client = pyorient.OrientDB("localhost", 2424)
session_id = client.connect( "admin", "admin" )
client.db_create( db_name, pyorient.DB_TYPE_GRAPH, pyorient.STORAGE_TYPE_MEMORY )
client.db_exists( db_name, pyorient.STORAGE_TYPE_MEMORY )
client.db_open( db_name, "admin", "admin" )
client.db_close()
client.db_list()
client.db_size()
client.db_count_records()
cluster_id = client.command( "create class my_class extends V" )
client.command(
"insert into my_class ( 'accommodation', 'work', 'holiday' ) values( 'B&B', 'garage', 'mountain' )"
)
Warning Some issues are experimented with record_create/record_upload and OrientDB < 2.0. These command are strongly discouraged with these versions
rec = { '@my_class': { 'accommodation': 'house', 'work': 'office', 'holiday': 'sea' } }
rec_position = client.record_create( cluster_id, rec )
Warning Some issues are experimented with record_create/record_upload and OrientDB < 2.0. These command are strongly discouraged with these versions
rec3 = { '@my_class': { 'accommodation': 'hotel', 'work': 'home', 'holiday': 'hills' } }
update_success = client.record_update( rec_position._rid, rec_position._rid, rec3, rec_position._version )
client.record_load( rec_position._rid )
def _my_callback(for_every_record):
print(for_every_record)
client.record_load( rec_position._rid, "*:-1", _my_callback )
result = client.query("select from my_class", 10, '*:0')
def _my_callback(for_every_record):
print(for_every_record)
result = client.query_async("select from my_class", 10, '*:0', _my_callback)
client.record_delete( cluster_id, rec_position._rid )
client.db_drop( db_name )
new_cluster_id = client.data_cluster_add(
'my_cluster_1234567', pyorient.CLUSTER_TYPE_PHYSICAL
)
client.db_reload()
client.data_cluster_data_range( new_cluster_id )
client.data_cluster_count( [ 1, 2, 3, 4, 11 ] )
client.data_cluster_drop( new_cluster_id )
client.shutdown( "root", "a_super_secret_password" )
### use a cluster
cluster_id = 3
### execute real create to get some info
rec = { 'accommodation': 'mountain hut', 'work': 'not!', 'holiday': 'lake' }
rec_position = client.record_create( cluster_id, rec )
tx = client.tx_commit()
tx.begin()
### create a new record
rec1 = { 'accommodation': 'home', 'work': 'some work', 'holiday': 'surf' }
rec_position1 = client.record_create( -1, rec1 )
### prepare for an update
rec2 = { 'accommodation': 'hotel', 'work': 'office', 'holiday': 'mountain' }
update_record = client.record_update( cluster_id, rec_position._rid, rec2, rec_position._version )
tx.attach( rec_position1 )
tx.attach( rec_position1 )
tx.attach( update_record )
res = tx.commit()
assert res["#3:1"].holiday == 'mountain'
assert res["#3:2"].holiday == 'surf'
assert res["#3:3"].holiday == 'surf'
cmd = ("begin;"
"let a = create vertex set script = true;"
"let b = select from v limit 1;"
"let e = create edge from $a to $b;"
"commit retry 100;")
edge_result = self.client.batch(cmd)
Since version 27 is introduced an extension to allow use a token based session. This functionality must be enabled on the server config.
client.set_session_token
method.When using the token based authentication, the connections can be shared between users of the same server.
client = pyorient.OrientDB("localhost", 2424)
client.set_session_token( True ) # set true to enable the token based
authentication
client.db_open( "GratefulDeadConcerts", "admin", "admin" )
### store this token somewhere
sessionToken = client.get_session_token()
### destroy the old client, equals to another user/socket/ip ecc.
del client
### create a new client
client = pyorient.OrientDB("localhost", 2424)
### set the previous obtained token to re-attach to the old session
client.set_session_token( sessionToken )
### now the dbOpen is not needed to perform database operations
record = client.query( 'select from V where @rid = #9:1' )
### set the flag again to true if you want to renew the token
client.set_session_token( True ) # set true
client.db_open( "GratefulDeadConcerts", "admin", "admin" )
new_sessionToken = client.get_session_token()
assert sessionToken != new_sessionToken
The GRAPH representation of animals and its food
import pyorient
client = pyorient.OrientDB("localhost", 2424) # host, port
### open a connection (username and password)
client.connect("admin", "admin")
### create a database
client.db_create("animals", pyorient.DB_TYPE_GRAPH, pyorient.STORAGE_TYPE_MEMORY)
### select to use that database
client.db_open("animals", "admin", "admin")
### Create the Vertex Animal
client.command("create class Animal extends V")
### Insert a new value
client.command("insert into Animal set name = 'rat', specie = 'rodent'")
### query the values
client.query("select * from Animal")
[<OrientRecord at 0x7f>..., ...]
### Create the vertex and insert the food values
client.command('create class Food extends V')
client.command("insert into Food set name = 'pea', color = 'green'")
### Create the edge for the Eat action
client.command('create class Eat extends E')
### Lets the rat likes to eat pea
eat_edges = client.command(
"create edge Eat from ("
"select from Animal where name = 'rat'"
") to ("
"select from Food where name = 'pea'"
")"
)
### Who eats the peas?
pea_eaters = client.command("select expand( in( Eat )) from Food where name = 'pea'")
for animal in pea_eaters:
print(animal.name, animal.specie)
'rat rodent'
...
### What each animal eats?
animal_foods = client.command("select expand( out( Eat )) from Animal")
for food in animal_foods:
animal = client.query(
"select name from ( select expand( in('Eat') ) from Food where name = 'pea' )"
)[0]
print(food.name, food.color, animal.name)
'pea green rat'
To Jetbrains tools support, coded with:
Copyright (c) 2014 Niko Usai, Domenico Lupinetti. See LICENSE for details.