def load_schema(self):
"""
Loads the schema definition for this column family from
Cassandra and updates comparator and validation classes if
neccessary.
"""
ksdef = self.pool.execute("get_keyspace_description", use_dict_for_col_metadata=True)
try:
self._cfdef = ksdef[self.column_family]
except KeyError:
nfe = NotFoundException()
nfe.why = "Column family %s not found." % self.column_family
raise nfe
self.super = self._cfdef.column_type == "Super"
self._load_comparator_classes()
self._load_validation_classes()
self._load_key_class()
def load_schema(self):
"""
Loads the schema definition for this column family from
Cassandra and updates comparator and validation classes if
neccessary.
"""
ksdef = self.pool.execute('get_keyspace_description',
use_dict_for_col_metadata=True)
try:
self._cfdef = ksdef[self.column_family]
except KeyError:
nfe = NotFoundException()
nfe.why = 'Column family %s not found.' % self.column_family
raise nfe
self.super = self._cfdef.column_type == 'Super'
self._load_comparator_classes()
self._load_validation_classes()
self._load_key_class()
def load_schema(self):
"""
Loads the schema definition for this column family from
Cassandra and updates comparator and validation classes if
neccessary.
"""
try:
try:
self._obtain_connection()
ksdef = self._tlocal.client.get_keyspace_description(use_dict_for_col_metadata=True)
self._cfdef = ksdef[self.column_family]
except KeyError:
nfe = NotFoundException()
nfe.why = 'Column family %s not found.' % self.column_family
raise nfe
finally:
self._release_connection()
self.super = self._cfdef.column_type == 'Super'
self._load_comparator_classes()
self._load_validation_classes()
self._load_key_class()
def load_schema(self):
"""
Loads the schema definition for this column family from
Cassandra and updates comparator and validation classes if
neccessary.
"""
try:
try:
self._obtain_connection()
ksdef = self._tlocal.client.get_keyspace_description(use_dict_for_col_metadata=True)
self._cfdef = ksdef[self.column_family]
except KeyError:
nfe = NotFoundException()
nfe.why = 'Column family %s not found.' % self.column_family
raise nfe
finally:
self._release_connection()
self.super = self._cfdef.column_type == 'Super'
self._load_comparator_classes()
self._load_validation_classes()
self._load_key_class()
def get(self, key, columns=None, column_start="", column_finish="",
column_reversed=False, column_count=100, include_timestamp=False,
super_column=None, read_consistency_level=None, include_ttl=False):
"""
Fetches all or part of the row with key `key`.
The columns fetched may be limited to a specified list of column names
using `columns`.
Alternatively, you may fetch a slice of columns or super columns from a row
using `column_start`, `column_finish`, and `column_count`.
Setting these will cause columns or super columns to be fetched starting with
`column_start`, continuing until `column_count` columns or super columns have
been fetched or `column_finish` is reached. If `column_start` is left as the
empty string, the slice will begin with the start of the row; leaving
`column_finish` blank will cause the slice to extend to the end of the row.
Note that `column_count` defaults to 100, so rows over this size will not be
completely fetched by default.
If `column_reversed` is ``True``, columns are fetched in reverse sorted order,
beginning with `column_start`. In this case, if `column_start` is the empty
string, the slice will begin with the end of the row.
You may fetch all or part of only a single super column by setting `super_column`.
If this is set, `column_start`, `column_finish`, `column_count`, and `column_reversed`
will apply to the subcolumns of `super_column`.
To include every column's timestamp in the result set, set `include_timestamp` to
``True``. Results will include a ``(value, timestamp)`` tuple for each column.
To include every column's ttl in the result set, set `include_ttl` to
``True``. Results will include a ``(value, ttl)`` tuple for each column.
If this is a standard column family, the return type is of the form
``{column_name: column_value}``. If this is a super column family and `super_column`
is not specified, the results are of the form
``{super_column_name: {column_name, column_value}}``. If `super_column` is set,
the super column name will be excluded and the results are of the form
``{column_name: column_value}``.
"""
packed_key = self._pack_key(key)
single_column = columns is not None and len(columns) == 1
if (not self.super and single_column) or \
(self.super and super_column is not None and single_column):
column = None
if self.super and super_column is None:
super_column = columns[0]
else:
column = columns[0]
cp = self._column_path(super_column, column)
col_or_super = self.pool.execute('get', packed_key, cp,
read_consistency_level or self.read_consistency_level)
return self._cosc_to_dict([col_or_super], include_timestamp, include_ttl)
else:
cp = self._column_parent(super_column)
sp = self._slice_predicate(columns, column_start, column_finish,
column_reversed, column_count, super_column)
list_col_or_super = self.pool.execute('get_slice', packed_key, cp, sp,
read_consistency_level or self.read_consistency_level)
if len(list_col_or_super) == 0:
raise NotFoundException()
return self._cosc_to_dict(list_col_or_super, include_timestamp, include_ttl)
def __init__(self,
client,
column_family,
buffer_size=1024,
read_consistency_level=ConsistencyLevel.ONE,
write_consistency_level=ConsistencyLevel.ONE,
timestamp=gm_timestamp,
super=False,
dict_class=dict,
autopack_names=True,
autopack_values=True):
"""
Constructs an abstraction of a Cassandra column family or super column family.
Operations on this, such as `get` or `insert` will get data from or
insert data into the corresponding Cassandra column family.
:Parameters:
`client`: :class:`cassandra.Cassandra.Client`
Cassandra client with thrift API
`column_family`: string
The name of this ColumnFamily
`buffer_size`: integer
When calling `get_range`, the intermediate results need to be
buffered if we are fetching many rows, otherwise the Cassandra
server will overallocate memory and fail. This is the size of
that buffer.
`read_consistency_level`: :class:`pycassa.cassandra.ttypes.ConsistencyLevel`
Affects the guaranteed replication factor before returning from
any read operation
`write_consistency_level`: :class:`pycassa.cassandra.ttypes.ConsistencyLevel`
Affects the guaranteed replication factor before returning from
any write operation
`timestamp`: function
The default timestamp function returns:
int(time.mktime(time.gmtime()))
Or the number of seconds since Unix epoch in GMT.
Set timestamp to replace the default timestamp function with your
own.
`super`: bool
Whether this ColumnFamily has SuperColumns
`dict_class` : class (must act like the dict type)
The default dict_class is :class:`dict`.
If the order of columns matter to you, pass your own dictionary
class, or python 2.7's new :class:`collections.OrderedDict`. All returned
rows and subcolumns are instances of this.
`autopack_names`: bool
Whether column and supercolumn names should be packed automatically
based on the comparator and subcomparator for the column
family. This does not typically work when used with
:class:`~pycassa.columnfamilymap.ColumnFamilyMap`.
`autopack_values`: bool
Whether column values should be packed automatically based on
the validator_class for a given column. This should probably
be set to ``False`` when used with a
:class:`~pycassa.columnfamilymap.ColumnFamilyMap`.
"""
self.client = client
self.column_family = column_family
self.buffer_size = buffer_size
self.read_consistency_level = read_consistency_level
self.write_consistency_level = write_consistency_level
self.timestamp = timestamp
self.super = super
self.dict_class = dict_class
self.autopack_names = autopack_names
self.autopack_values = autopack_values
# Determine the ColumnFamily type to allow for auto conversion
# so that packing/unpacking doesn't need to be done manually
self.cf_data_type = None
self.col_name_data_type = None
self.supercol_name_data_type = None
self.col_type_dict = dict()
col_fam = None
try:
col_fam = client.get_keyspace_description()[self.column_family]
except KeyError:
raise NotFoundException('Column family %s not found.' %
self.column_family)
if col_fam is not None:
if self.autopack_names:
if not self.super:
self.col_name_data_type = col_fam.comparator_type
else:
self.col_name_data_type = col_fam.subcomparator_type
self.supercol_name_data_type = self._extract_type_name(
col_fam.comparator_type)
index = self.col_name_data_type = self._extract_type_name(
self.col_name_data_type)
if self.autopack_values:
self.cf_data_type = self._extract_type_name(
col_fam.default_validation_class)
for name, cdef in col_fam.column_metadata.items():
self.col_type_dict[name] = self._extract_type_name(
cdef.validation_class)
def get_indexed_slices(self,
index_clause,
columns=None,
column_start="",
column_finish="",
column_reversed=False,
column_count=100,
include_timestamp=False,
super_column=None,
read_consistency_level=None):
"""
Fetches a list of KeySlices from a Cassandra server based on an index clause
:Parameters:
`index_clause`: :class:`~pycassa.cassandra.ttypes.IndexClause`
Limits the keys that are returned based on expressions that compare
the value of a column to a given value. At least one of the
expressions in the IndexClause must be on an indexed column.
.. seealso:: meth::pycassa.index.create_index_clause() and
meth::pycassa.index.create_index_expression().
`columns`: [str]
Limit the columns or super_columns fetched to the specified list
`column_start`: str
Only fetch when a column or super_column is >= column_start
`column_finish`: str
Only fetch when a column or super_column is <= column_finish
`column_reversed`: bool
Fetch the columns or super_columns in reverse order. This will do
nothing unless you passed a dict_class to the constructor.
`column_count`: int
Limit the number of columns or super_columns fetched per key
`include_timestamp`: bool
If true, return a (value, timestamp) tuple for each column
`super_column`: str
Return columns only in this super_column
`read_consistency_level`: :class:`pycassa.cassandra.ttypes.ConsistencyLevel`
Affects the guaranteed replication factor before returning from
any read operation
:Returns:
if include_timestamp == True: {key : {column : (value, timestamp)}}
else: {key : {column : value}}
"""
(super_column, column_start,
column_finish) = self._pack_slice_cols(super_column, column_start,
column_finish)
packed_cols = None
if columns is not None:
packed_cols = []
for col in columns:
packed_cols.append(
self._pack_name(col, is_supercol_name=self.super))
cp = ColumnParent(column_family=self.column_family,
super_column=super_column)
sp = create_SlicePredicate(packed_cols, column_start, column_finish,
column_reversed, column_count)
# Pack the values in the index clause expressions
new_exprs = []
for expr in index_clause.expressions:
new_exprs.append(IndexExpression(self._pack_name(expr.column_name), expr.op, \
self._pack_value(expr.value, expr.column_name)))
index_clause.expressions = new_exprs
keyslice_list = self.client.get_indexed_slices(
cp, index_clause, sp, self._rcl(read_consistency_level))
if len(keyslice_list) == 0:
raise NotFoundException()
return self._convert_KeySlice_list_to_dict_class(
keyslice_list, include_timestamp)
def get(self,
key,
columns=None,
column_start="",
column_finish="",
column_reversed=False,
column_count=100,
include_timestamp=False,
super_column=None,
read_consistency_level=None):
"""
Fetch a key from a Cassandra server
:Parameters:
`key`: str
The key to fetch
`columns`: [str]
Limit the columns or super_columns fetched to the specified list
`column_start`: str
Only fetch when a column or super_column is >= column_start
`column_finish`: str
Only fetch when a column or super_column is <= column_finish
`column_reversed`: bool
Fetch the columns or super_columns in reverse order. This will do
nothing unless you passed a ``dict_class`` to the constructor.
`column_count`: int
Limit the number of columns or super_columns fetched per key
`include_timestamp` : bool
If true, return a (value, timestamp) tuple for each column
`super_column`: str
Return columns only in this super_column
`read_consistency_level`: :class:`pycassa.cassandra.ttypes.ConsistencyLevel`
Affects the guaranteed replication factor before returning from
any read operation
:Returns:
if include_timestamp == True: {'column': ('value', timestamp)}
else: {'column': 'value'}
"""
super_column, column_start, column_finish = self._pack_slice_cols(
super_column, column_start, column_finish)
packed_cols = None
if columns is not None:
packed_cols = []
for col in columns:
packed_cols.append(
self._pack_name(col, is_supercol_name=self.super))
cp = ColumnParent(column_family=self.column_family,
super_column=super_column)
sp = create_SlicePredicate(packed_cols, column_start, column_finish,
column_reversed, column_count)
list_col_or_super = self.client.get_slice(
key, cp, sp, self._rcl(read_consistency_level))
if len(list_col_or_super) == 0:
raise NotFoundException()
return self._convert_ColumnOrSuperColumns_to_dict_class(
list_col_or_super, include_timestamp)
def __init__(self, pool, column_family, buffer_size=1024,
read_consistency_level=ConsistencyLevel.ONE,
write_consistency_level=ConsistencyLevel.ONE,
timestamp=gm_timestamp, super=False,
dict_class=util.OrderedDict, autopack_names=True,
autopack_values=True):
"""
An abstraction of a Cassandra column family or super column family.
Operations on this, such as :meth:`get` or :meth:`insert` will get data from or
insert data into the corresponding Cassandra column family with
name `column_family`.
`pool` is a :class:`~pycassa.pool.ConnectionPool` that the column
family will use for all operations. A connection is drawn from the
pool before each operations and is returned afterwards. Note that
the keyspace to be used is determined by the pool.
When calling :meth:`get_range()` or :meth:`get_indexed_slices()`,
the intermediate results need to be buffered if we are fetching many
rows, otherwise the Cassandra server will overallocate memory and fail.
`buffer_size` is the size of that buffer in number of rows. The default
is 1024.
`read_consistency_level` and `write_consistency_level` set the default
consistency levels for every operation; these may be overridden
per-operation. These should be instances of
:class:`~pycassa.cassandra.ttypes.ConsistencyLevel`. These default
to level ``ONE``.
Each :meth:`insert()` or :meth:`remove` sends a timestamp with every
column. The `timestamp` parameter is a function that is used to get
this timestamp when needed. The default function is :meth:`gm_timestamp()`.
Results are returned as dictionaries. :class:`~pycassa.util.OrderedDict` is
used by default so that order is maintained. A different class, such as
:class:`dict` may be used instead by passing `dict_class`.
By default, column family definitions will be examined to determine
what data type Cassandra expects for column names and values. When
columns are retrieved or inserted, their names and values will be
packed or unpacked if necessary to convert them to or from their
binary representation. Automatic packing of names and values can
be individually enabled or disabled with `autopack_names` and
`autopack_values`. When using :class:`~pycassa.columnfamilymap.ColumnFamilyMap`,
these should both be set to ``False``.
"""
self.pool = pool
self._tlocal = threading.local()
self._tlocal.client = None
self.column_family = column_family
self.buffer_size = buffer_size
self.read_consistency_level = read_consistency_level
self.write_consistency_level = write_consistency_level
self.timestamp = timestamp
self.dict_class = dict_class
self.autopack_names = autopack_names
self.autopack_values = autopack_values
# Determine the ColumnFamily type to allow for auto conversion
# so that packing/unpacking doesn't need to be done manually
self.cf_data_type = None
self.col_name_data_type = None
self.supercol_name_data_type = None
self.col_type_dict = dict()
col_fam = None
try:
try:
self._obtain_connection()
col_fam = self._tlocal.client.get_keyspace_description(use_dict_for_col_metadata=True)[self.column_family]
except KeyError:
nfe = NotFoundException()
nfe.why = 'Column family %s not found.' % self.column_family
raise nfe
finally:
self._release_connection()
if col_fam is not None:
self.super = col_fam.column_type == 'Super'
if self.autopack_names:
if not self.super:
self.col_name_data_type = col_fam.comparator_type
else:
self.col_name_data_type = col_fam.subcomparator_type
self.supercol_name_data_type = util.extract_type_name(col_fam.comparator_type)
index = self.col_name_data_type = util.extract_type_name(self.col_name_data_type)
if self.autopack_values:
self.cf_data_type = util.extract_type_name(col_fam.default_validation_class)
for name, cdef in col_fam.column_metadata.items():
self.col_type_dict[name] = util.extract_type_name(cdef.validation_class)
def __init__(self, pool, column_family, buffer_size=1024,
read_consistency_level=ConsistencyLevel.ONE,
write_consistency_level=ConsistencyLevel.ONE,
timestamp=gm_timestamp, super=False,
dict_class=OrderedDict, autopack_names=True,
autopack_values=True):
"""
Constructs an abstraction of a Cassandra column family or super column family.
Operations on this, such as :meth:`get` or :meth:`insert` will get data from or
insert data into the corresponding Cassandra column family.
:param pool: A connection pool to a Cassandra cluster
:type client: :class:`~pycassa.pool.AbstractPool`
:param column_family: The name of the column family
:type column_family: string
:param buffer_size: When calling :meth:`get_range()` or
:meth:`get_indexed_slices()`, the intermediate results need
to be buffered if we are fetching many rows, otherwise the
Cassandra server will overallocate memory and fail. This
is the size of that buffer in number of rows.
:type buffer_size: int
:param read_consistency_level: Affects the guaranteed replication factor
before returning from any read operation
:type read_consistency_level: :class:`~pycassa.cassandra.ttypes.ConsistencyLevel`
:param write_consistency_level: Affects the guaranteed replication
factor before returning from any write operation
:type write_consistency_level: :class:`~pycassa.cassandra.ttypes.ConsistencyLevel`
:param timestamp:
The default timestamp function returns
``int(time.mktime(time.gmtime()))``,
the number of seconds since Unix epoch in GMT.
Set this to replace the default timestamp function with your own.
:type timestamp: function
:param dict_class: The default dict_class is :class:`~pycassa.util.OrderedDict`.
All returned rows and subcolumns are instances of this.
:type dict_class: class
:param autopack_names: Whether column and supercolumn names should
be packed automatically based on the comparator and subcomparator
for the column family. This does not typically work when used with
:class:`~pycassa.columnfamilymap.ColumnFamilyMap`.
:type autopack_names: bool
:param autopack_values: Whether column values should be packed
automatically based on the validator_class for a given column.
This should probably be set to ``False`` when used with a
:class:`~pycassa.columnfamilymap.ColumnFamilyMap`.
:type autopack_values: bool
:param super: Whether this column family has super columns. This
is detected automatically since 0.5.1.
.. deprecated:: 0.5.1
:type super: bool
"""
self.pool = pool
self.client = None
self.column_family = column_family
self.buffer_size = buffer_size
self.read_consistency_level = read_consistency_level
self.write_consistency_level = write_consistency_level
self.timestamp = timestamp
self.dict_class = dict_class
self.autopack_names = autopack_names
self.autopack_values = autopack_values
# Determine the ColumnFamily type to allow for auto conversion
# so that packing/unpacking doesn't need to be done manually
self.cf_data_type = None
self.col_name_data_type = None
self.supercol_name_data_type = None
self.col_type_dict = dict()
col_fam = None
try:
self.client = self.pool.get()
col_fam = self.client.get_keyspace_description(use_dict_for_col_metadata=True)[self.column_family]
except KeyError:
nfe = NotFoundException()
nfe.why = 'Column family %s not found.' % self.column_family
raise nfe
finally:
self.client.return_to_pool()
if col_fam is not None:
self.super = col_fam.column_type == 'Super'
if self.autopack_names:
if not self.super:
self.col_name_data_type = col_fam.comparator_type
else:
self.col_name_data_type = col_fam.subcomparator_type
self.supercol_name_data_type = self._extract_type_name(col_fam.comparator_type)
index = self.col_name_data_type = self._extract_type_name(self.col_name_data_type)
if self.autopack_values:
self.cf_data_type = self._extract_type_name(col_fam.default_validation_class)
for name, cdef in col_fam.column_metadata.items():
self.col_type_dict[name] = self._extract_type_name(cdef.validation_class)
def __init__(self,
pool,
column_family,
buffer_size=1024,
read_consistency_level=ConsistencyLevel.ONE,
write_consistency_level=ConsistencyLevel.ONE,
timestamp=gm_timestamp,
super=False,
dict_class=util.OrderedDict,
autopack_names=True,
autopack_values=True):
"""
An abstraction of a Cassandra column family or super column family.
Operations on this, such as :meth:`get` or :meth:`insert` will get data from or
insert data into the corresponding Cassandra column family with
name `column_family`.
`pool` is a :class:`~pycassa.pool.ConnectionPool` that the column
family will use for all operations. A connection is drawn from the
pool before each operations and is returned afterwards. Note that
the keyspace to be used is determined by the pool.
When calling :meth:`get_range()` or :meth:`get_indexed_slices()`,
the intermediate results need to be buffered if we are fetching many
rows, otherwise the Cassandra server will overallocate memory and fail.
`buffer_size` is the size of that buffer in number of rows. The default
is 1024.
`read_consistency_level` and `write_consistency_level` set the default
consistency levels for every operation; these may be overridden
per-operation. These should be instances of
:class:`~pycassa.cassandra.ttypes.ConsistencyLevel`. These default
to level ``ONE``.
Each :meth:`insert()` or :meth:`remove` sends a timestamp with every
column. The `timestamp` parameter is a function that is used to get
this timestamp when needed. The default function is :meth:`gm_timestamp()`.
Results are returned as dictionaries. :class:`~pycassa.util.OrderedDict` is
used by default so that order is maintained. A different class, such as
:class:`dict` may be used instead by passing `dict_class`.
By default, column family definitions will be examined to determine
what data type Cassandra expects for column names and values. When
columns are retrieved or inserted, their names and values will be
packed or unpacked if necessary to convert them to or from their
binary representation. Automatic packing of names and values can
be individually enabled or disabled with `autopack_names` and
`autopack_values`. When using :class:`~pycassa.columnfamilymap.ColumnFamilyMap`,
these should both be set to ``False``.
"""
self.pool = pool
self._tlocal = threading.local()
self._tlocal.client = None
self.column_family = column_family
self.buffer_size = buffer_size
self.read_consistency_level = read_consistency_level
self.write_consistency_level = write_consistency_level
self.timestamp = timestamp
self.dict_class = dict_class
self.autopack_names = autopack_names
self.autopack_values = autopack_values
# Determine the ColumnFamily type to allow for auto conversion
# so that packing/unpacking doesn't need to be done manually
self.cf_data_type = None
self.col_name_data_type = None
self.supercol_name_data_type = None
self.col_type_dict = dict()
col_fam = None
try:
try:
self._obtain_connection()
col_fam = self._tlocal.client.get_keyspace_description(
use_dict_for_col_metadata=True)[self.column_family]
except KeyError:
nfe = NotFoundException()
nfe.why = 'Column family %s not found.' % self.column_family
raise nfe
finally:
self._release_connection()
if col_fam is not None:
self.super = col_fam.column_type == 'Super'
if self.autopack_names:
if not self.super:
self.col_name_data_type = col_fam.comparator_type
else:
self.col_name_data_type = col_fam.subcomparator_type
self.supercol_name_data_type = util.extract_type_name(
col_fam.comparator_type)
index = self.col_name_data_type = util.extract_type_name(
self.col_name_data_type)
if self.autopack_values:
self.cf_data_type = util.extract_type_name(
col_fam.default_validation_class)
for name, cdef in col_fam.column_metadata.items():
self.col_type_dict[name] = util.extract_type_name(
cdef.validation_class)
