def main(argv):
usage = "Usage: dbschema [options] [{resource}]"
args, kwargs = parse(argv)
host = kwargs.get('host', settings.DEFAULT_CASSANDRA_HOST)
port = kwargs.get('port', settings.DEFAULT_CASSANDRA_PORT)
try:
system = SystemManager("%s:%s" % (host, port))
# FOR TESTING ONLY
# print str(argv)
if len(argv) == 0:
list_keyspaces(system)
return
if len(argv) == 1:
ksname = argv[0]
list_keyspace(system, ksname)
if len(argv) == 2:
ksname = argv[0]
cfname = argv[1]
if cfname != 'ANY':
list_column_family(system, ksname, cfname)
else:
list_keyspace(system, ksname)
except:
error(output, excinfo(), 2)
def main(argv):
usage = "Usage: dbschema [options] [{resource}]"
args, kwargs = parse(argv)
host = kwargs.get('host', settings.DEFAULT_CASSANDRA_HOST)
port = kwargs.get('port', settings.DEFAULT_CASSANDRA_PORT)
try:
system = SystemManager("%s:%s" % (host, port))
# FOR TESTING ONLY
# print str(argv)
if len(argv) == 0:
list_keyspaces(system)
return
if len(argv) == 1:
ksname = argv[0]
list_keyspace(system, ksname)
if len(argv) == 2:
ksname = argv[0]
cfname = argv[1]
if cfname != 'ANY':
list_column_family(system, ksname, cfname)
else:
list_keyspace(system, ksname)
except:
error(output, excinfo(), 2)
def main(argv):
usage = "Usage: dbdiscover.py"
args, kwargs = parse(argv)
host = kwargs.get('host', settings.DEFAULT_CASSANDRA_HOST)
port = kwargs.get('port', settings.DEFAULT_CASSANDRA_PORT)
try:
system = SystemManager("%s:%s" % (host, port))
ksprops = system.get_keyspace_properties("system").keys()
# header = ["keyspace"] + sorted(ksprops)
header = ["keyspace", "column_family"]
writer = csv.writer(output)
writer.writerow(header)
for keyspace in system.list_keyspaces():
ksinfo = system.get_keyspace_column_families(keyspace, use_dict_for_col_metadata=True)
attrs = [
"id",
"column_type",
"comment",
"comparator_type",
"default_validation_class",
"gc_grace_seconds",
"key_alias",
"key_cache_save_period_in_seconds",
"key_cache_size",
"key_validation_class",
"max_compaction_threshold",
"memtable_operations_in_millions",
"memtable_throughput_in_mb",
"memtable_flush_after_mins",
"merge_shards_chance",
"min_compaction_threshold",
"read_repair_chance",
"replicate_on_write",
"row_cache_provider",
"row_cache_save_period_in_seconds",
"row_cache_size",
"subcomparator_type",
]
for cfname, cfdef in ksinfo.iteritems():
ks = ("keyspace="+ keyspace)
cf = ("column_family="+ cfname)
row =(ks, cf)
writer.writerow(row)
return
except:
error(output, excinfo(), 2)
def main(argv):
usage = "Usage: dbdiscover.py"
args, kwargs = parse(argv)
host = kwargs.get('host', settings.DEFAULT_CASSANDRA_HOST)
port = kwargs.get('port', settings.DEFAULT_CASSANDRA_PORT)
try:
system = SystemManager("%s:%s" % (host, port))
ksprops = system.get_keyspace_properties("system").keys()
# header = ["keyspace"] + sorted(ksprops)
header = ["keyspace", "column_family"]
writer = csv.writer(output)
writer.writerow(header)
for keyspace in system.list_keyspaces():
ksinfo = system.get_keyspace_column_families(
keyspace, use_dict_for_col_metadata=True)
attrs = [
"id",
"column_type",
"comment",
"comparator_type",
"default_validation_class",
"gc_grace_seconds",
"key_alias",
"key_cache_save_period_in_seconds",
"key_cache_size",
"key_validation_class",
"max_compaction_threshold",
"memtable_operations_in_millions",
"memtable_throughput_in_mb",
"memtable_flush_after_mins",
"merge_shards_chance",
"min_compaction_threshold",
"read_repair_chance",
"replicate_on_write",
"row_cache_provider",
"row_cache_save_period_in_seconds",
"row_cache_size",
"subcomparator_type",
]
for cfname, cfdef in ksinfo.iteritems():
ks = ("keyspace=" + keyspace)
cf = ("column_family=" + cfname)
row = (ks, cf)
writer.writerow(row)
return
except:
error(output, excinfo(), 2)
def main(argv):
usage = "Usage: dbcql {query}"
args, kwargs = parse(argv)
if len(args) > 1:
error(output, "Unexpected argument: '%s'" % args[1], 2)
if len(args) < 1:
error(output, "No query argument", 2)
keyspace = kwargs.get('keyspace', None)
host = kwargs.get('host', settings.DEFAULT_CASSANDRA_HOST)
port = kwargs.get('port', settings.DEFAULT_CASSANDRA_PORT)
# UNDONE: credentials ..
query = args[0]
if query is None:
error(output, "Command requires a single query argument", 2)
# A query may consist of multiple expressions. We execute each of
# the expressions in order and output the results from the final
# expression. The primary scenario is:
#
# "USE {keyspace}; SELECT * FROM .."
#
# However, arbitrary query expressions may be combined in this way.
expressions = query.split(';')
# Scan the expressions looking for anything we need to disable
for expression in expressions:
# UNDONE: The following disables usage of the CLQ DROP command
# in order to prevent users from inadvertently (or maliciously)
# dropping of critical keyspace data until we can properly enable
# the capability by integrating with Splunk's role based access
# control (where this would be an admin-only capability).
if re.match("\s*drop ", expression, re.I):
error(output, "The CQL DROP command has been disabled.", 2)
connection = None
cursor = None
try:
connection = cql.connect(host, port, keyspace)
cursor = connection.cursor()
# Everything looks ok .. so run them.
for expression in expressions:
cursor.execute(expression)
# A 'SELECT *' expression requires special handling because the
# results may be "ragged", meaning that we don't know what fields
# will appear for any given row. In order to handle this case we
# scan an initial `batchsize` set of rows, collecting all fields
# that we see. We then use those fields for the results header and
# collect any additional fields that we subsequently see into an
# "_extra" MV field. Note that a row in Cassandra can contain up
# to 2B columns, so "SELECT *" may have other issues aside from the
# incrimental processing impact on this search script.
if re.search("select\s+\*", expression, re.I):
ragged(cursor)
else:
normal(cursor)
except:
error(output, excinfo(), 2)
finally:
if cursor is not None:
cursor.close()
if connection is not None:
connection.close()
def main(argv):
usage = "Usage: dbcql {query}"
args, kwargs = parse(argv)
if len(args) > 1:
error(output, "Unexpected argument: '%s'" % args[1], 2)
if len(args) < 1:
error(output, "No query argument", 2)
keyspace = kwargs.get('keyspace', None)
host = kwargs.get('host', settings.DEFAULT_CASSANDRA_HOST)
port = kwargs.get('port', settings.DEFAULT_CASSANDRA_PORT)
# UNDONE: credentials ..
query = args[0]
if query is None:
error(output, "Command requires a single query argument", 2)
# A query may consist of multiple expressions. We execute each of
# the expressions in order and output the results from the final
# expression. The primary scenario is:
#
# "USE {keyspace}; SELECT * FROM .."
#
# However, arbitrary query expressions may be combined in this way.
expressions = query.split(';')
# Scan the expressions looking for anything we need to disable
for expression in expressions:
# UNDONE: The following disables usage of the CLQ DROP command
# in order to prevent users from inadvertently (or maliciously)
# dropping of critical keyspace data until we can properly enable
# the capability by integrating with Splunk's role based access
# control (where this would be an admin-only capability).
if re.match("\s*drop ", expression, re.I):
error(output, "The CQL DROP command has been disabled.", 2);
connection = None
cursor = None
try:
connection = cql.connect(host, port, keyspace)
cursor = connection.cursor()
# Everything looks ok .. so run them.
for expression in expressions:
cursor.execute(expression)
# A 'SELECT *' expression requires special handling because the
# results may be "ragged", meaning that we don't know what fields
# will appear for any given row. In order to handle this case we
# scan an initial `batchsize` set of rows, collecting all fields
# that we see. We then use those fields for the results header and
# collect any additional fields that we subsequently see into an
# "_extra" MV field. Note that a row in Cassandra can contain up
# to 2B columns, so "SELECT *" may have other issues aside from the
# incrimental processing impact on this search script.
if re.search("select\s+\*", expression, re.I):
ragged(cursor)
else:
normal(cursor)
except:
error(output, excinfo(), 2)
finally:
if cursor is not None:
cursor.close()
if connection is not None:
connection.close()
fields = fields.split(',')
cfpath = cfpath.split('.')
if len(cfpath) != 2:
error(output, "Invalid column family path", 2)
ksname, cfname = cfpath
try:
server = "%s:%d" % (host, port)
pool = pycassa.connect(ksname, [server])
cfam = pycassa.ColumnFamily(pool, cfname)
except pycassa.cassandra.c08.ttypes.InvalidRequestException, e:
error(output, e.why, 2)
except pycassa.cassandra.c08.ttypes.NotFoundException, e:
error(output, e.why, 2)
except:
error(output, excinfo(), 2)
for line in sys.stdin:
if line == '\n': break
reader = csv.DictReader(sys.stdin)
header = reader.fieldnames
if keycol not in header:
error(output, "Key field '%s' not found" % keycol, 2)
header.extend(fields)
writer = csv.DictWriter(output, header)
writer.writer.writerow(header)
while True:
