def saveToCassandra(rdd, keyspace=None, table=None, columns=None, row_format=None, keyed=None,
write_conf=None, **write_conf_kwargs):
'''
Saves an RDD to Cassandra. The RDD is expected to contain dicts with keys mapping to CQL
columns.
Arguments:
@param rdd(RDD):
The RDD to save. Equals to self when invoking saveToCassandra on a monkey patched RDD.
@param keyspace(string):in
The keyspace to save the RDD in. If not given and the rdd is a CassandraRDD the same
keyspace is used.
@param table(string):
The CQL table to save the RDD in. If not given and the rdd is a CassandraRDD the same
table is used.
Keyword arguments:
@param columns(iterable):
The columns to save, i.e. which keys to take from the dicts in the RDD.
If None given all columns are be stored.
@param row_format(RowFormat):
Make explicit how to map the RDD elements into Cassandra rows.
If None given the mapping is auto-detected as far as possible.
@param keyed(bool):
Make explicit that the RDD consists of key, value tuples (and not arrays of length
two).
@param write_conf(WriteConf):
A WriteConf object to use when saving to Cassandra
@param **write_conf_kwargs:
WriteConf parameters to use when saving to Cassandra
'''
keyspace = keyspace or getattr(rdd, 'keyspace', None)
if not keyspace:
raise ValueError("keyspace not set")
table = table or getattr(rdd, 'table', None)
if not table:
raise ValueError("table not set")
# create write config as map
write_conf = WriteConf.build(write_conf, **write_conf_kwargs)
write_conf = as_java_object(rdd.ctx._gateway, write_conf.settings())
# convert the columns to a string array
columns = as_java_array(rdd.ctx._gateway, "String", columns) if columns else None
helper(rdd.ctx) \
.saveToCassandra(
rdd._jrdd,
keyspace,
table,
columns,
row_format,
keyed,
write_conf,
)
def deleteFromCassandra(dstream, keyspace=None, table=None, deleteColumns=None,
keyColumns=None,
row_format=None, keyed=None, write_conf=None,
**write_conf_kwargs):
"""Delete data from Cassandra table, using data from the RDD as primary
keys. Uses the specified column names.
Arguments:
@param dstream(DStream)
The DStream to join. Equals to self when invoking
joinWithCassandraTable on a monkey patched RDD.
@param keyspace(string):in
The keyspace to save the RDD in. If not given and the rdd is a
CassandraRDD the same keyspace is used.
@param table(string):
The CQL table to save the RDD in. If not given and the rdd is a
CassandraRDD the same table is used.
Keyword arguments:
@param deleteColumns(iterable):
The list of column names to delete, empty ColumnSelector means full
row.
@param keyColumns(iterable):
The list of column names to delete, empty ColumnSelector means full
row.
@param row_format(RowFormat):
Primary key columns selector, Optional. All RDD primary columns
columns will be checked by default
@param keyed(bool):
Make explicit that the RDD consists of key, value tuples (and not
arrays of length two).
@param write_conf(WriteConf):
A WriteConf object to use when saving to Cassandra
@param **write_conf_kwargs:
WriteConf parameters to use when saving to Cassandra
"""
ctx = dstream._ssc._sc
gw = ctx._gateway
# create write config as map
write_conf = WriteConf.build(write_conf, **write_conf_kwargs)
write_conf = as_java_object(gw, write_conf.settings())
# convert the columns to a string array
deleteColumns = as_java_array(gw, "String",
deleteColumns) if deleteColumns else None
keyColumns = as_java_array(gw, "String", keyColumns) \
if keyColumns else None
return helper(ctx).deleteFromCassandra(dstream._jdstream, keyspace, table,
deleteColumns, keyColumns,
row_format,
keyed, write_conf)
def saveToCassandra(dstream, keyspace, table, columns=None, row_format=None, keyed=None, write_conf=None,
**write_conf_kwargs):
ctx = dstream._ssc._sc
gw = ctx._gateway
# create write config as map
write_conf = WriteConf.build(write_conf, **write_conf_kwargs)
write_conf = as_java_object(gw, write_conf.settings())
# convert the columns to a string array
columns = as_java_array(gw, "String", columns) if columns else None
return helper(ctx).saveToCassandra(dstream._jdstream, keyspace, table, columns, row_format, keyed, write_conf)
def saveToCassandra(dstream, keyspace, table, columns=None, row_format=None,
keyed=None,
write_conf=None, **write_conf_kwargs):
ctx = dstream._ssc._sc
gw = ctx._gateway
# create write config as map
write_conf = WriteConf.build(write_conf, **write_conf_kwargs)
write_conf = as_java_object(gw, write_conf.settings())
# convert the columns to a string array
columns = as_java_array(gw, "String", columns) if columns else None
return helper(ctx).saveToCassandra(dstream._jdstream, keyspace, table,
columns, row_format,
keyed, write_conf)
def test_write_conf(self):
rdd = self.sc.parallelize(
[{'key': i, 'text': i, 'int': i} for i in range(10)])
save = partial(rdd.saveToCassandra, self.keyspace, self.table)
save(batch_size=100)
save(batch_buffer_size=100)
save(batch_grouping_key='replica_set')
save(batch_grouping_key='partition')
save(consistency_level='ALL')
save(consistency_level=ConsistencyLevel.LOCAL_QUORUM)
save(parallelism_level=10)
save(throughput_mibps=10)
save(ttl=5)
save(ttl=timedelta(minutes=30))
save(timestamp=time.clock() * 1000 * 1000)
save(timestamp=datetime.now())
save(metrics_enabled=True)
save(write_conf=WriteConf(ttl=3, metrics_enabled=True))
def deleteFromCassandra(rdd,
keyspace=None,
table=None,
deleteColumns=None,
keyColumns=None,
row_format=None,
keyed=None,
write_conf=None,
**write_conf_kwargs):
"""
Delete data from Cassandra table, using data from the RDD as primary
keys. Uses the specified column names.
Arguments:
@param rdd(RDD):
The RDD to save. Equals to self when invoking saveToCassandra on a
monkey patched RDD.
@param keyspace(string):in
The keyspace to save the RDD in. If not given and the rdd is a
CassandraRDD the same keyspace is used.
@param table(string):
The CQL table to save the RDD in. If not given and the rdd is a
CassandraRDD the same table is used.
Keyword arguments:
@param deleteColumns(iterable):
The list of column names to delete, empty ColumnSelector means full
row.
@param keyColumns(iterable):
The list of column names to delete, empty ColumnSelector means full
row.
@param row_format(RowFormat):
Primary key columns selector, Optional. All RDD primary columns
columns will be checked by default
@param keyed(bool):
Make explicit that the RDD consists of key, value tuples (and not
arrays of length two).
@param write_conf(WriteConf):
A WriteConf object to use when saving to Cassandra
@param **write_conf_kwargs:
WriteConf parameters to use when saving to Cassandra
"""
keyspace = keyspace or getattr(rdd, 'keyspace', None)
if not keyspace:
raise ValueError("keyspace not set")
table = table or getattr(rdd, 'table', None)
if not table:
raise ValueError("table not set")
# create write config as map
write_conf = WriteConf.build(write_conf, **write_conf_kwargs)
write_conf = as_java_object(rdd.ctx._gateway, write_conf.settings())
# convert the columns to a string array
deleteColumns = as_java_array(rdd.ctx._gateway, "String", deleteColumns) \
if deleteColumns else None
keyColumns = as_java_array(rdd.ctx._gateway, "String", keyColumns) \
if keyColumns else None
helper(rdd.ctx) \
.deleteFromCassandra(
rdd._jrdd,
keyspace,
table,
deleteColumns,
keyColumns,
row_format,
keyed,
write_conf,
)