manager

class pyhs.manager.Manager(read_servers=None, write_servers=None, debug=False)

High-level client for HandlerSocket.

This should be used in most cases except ones that you need fine-grained control over index management, low-level operations, etc. For such cases ReadSocket and WriteSocket can be used.

Constructor initializes both read and write sockets.

Parameters:

• read_servers (list of tuples or None) – list of tuples that define HandlerSocket read instances. See format in HandlerSocket constructor.
• write_servers (list of tuples or None) – list of tuples that define HandlerSocket write instances. Format is the same as in read_servers.
• debug (bool) – enable debug mode by passing True.

find(db, table, operation, fields, values, index_name=None, limit=0, offset=0)

Finds rows that meet values with comparison operation in given db and table.

Returns a list of lists of pairs. First item in pair is field name, second is its value. For example, if two rows with two columns each are returned:

[[('field', 'first_row_value'), ('otherfield', 'first_row_othervalue')],
 [('field', 'second_row_value'), ('otherfield', 'second_row_othervalue')]]

Parameters:

• db (string) – database name
• table (string) – table name
• operation (string) – logical comparison operation to use over columns. Currently allowed operations are defined in FIND_OPERATIONS. Only one operation is allowed per call.
• fields (list) – list of table’s fields to get, ordered by inclusion into the index.
• values (list) – values to compare to, ordered the same way as items in fields.
• index_name (string or None) – name of the index to open, default is PRIMARY.
• limit (integer) – optional limit of results. Default is one row. In case multiple rows are expected to be returned, limit must be set explicitly, HS wont get all found rows by default.
• offset (integer) – optional offset of rows to search for.

Return type:

list of lists of tuples

insert(db, table, fields, index_name=None)

Inserts a single row into given table.

Parameters:

• db (string) – database name.
• table (string) – table name.
• fields (list of lists) – list of (column, value) pairs to insert into the table.
• index_name (string or None) – name of the index to open, default is PRIMARY.

Return type:

bool

update(db, table, operation, fields, values, update_values, index_name=None, limit=0, offset=0, return_original=False)

Update row(s) that meet conditions defined by operation, fields values in a given table.

Parameters:

• db (string) – database name
• table (string) – table name
• operation (string) – logical comparison operation to use over columns. Currently allowed operations are defined in FIND_OPERATIONS. Only one operation is allowed per call.
• fields (list) – list of table’s fields to use, ordered by inclusion into the index.
• values (list) – values to compare to, ordered the same way as items in fields.
• update_values (list) – values to update, ordered the same way as items in fields.
• index_name (string or None) – name of the index to open, default is PRIMARY.
• limit (integer) – optional limit of rows. Default is one row. In case multiple rows are expected to be updated, limit must be set explicitly, HS wont update all found rows by default.
• offset (integer) – optional offset of rows to search for.
• return_original (bool) – if set to True, method will return a list of original values in affected rows. Otherwise - number of affected rows (this is default behaviour).

Return type:

int or list

incr(db, table, operation, fields, values, step=['1'], index_name=None, limit=0, offset=0, return_original=False)

Increments row(s) that meet conditions defined by operation, fields values in a given table.

Parameters:

• db (string) – database name
• table (string) – table name
• operation (string) – logical comparison operation to use over columns. Currently allowed operations are defined in FIND_OPERATIONS. Only one operation is allowed per call.
• fields (list) – list of table’s fields to use, ordered by inclusion into the index.
• values (list) – values to compare to, ordered the same way as items in fields.
• step (list) – list of increment steps, ordered the same way as items in fields.
• index_name (string or None) – name of the index to open, default is PRIMARY.
• limit (integer) – optional limit of rows. Default is one row. In case multiple rows are expected to be updated, limit must be set explicitly, HS wont update all found rows by default.
• offset (integer) – optional offset of rows to search for.
• return_original (bool) – if set to True, method will return a list of original values in affected rows. Otherwise - number of affected rows (this is default behaviour).

Return type:

int or list

decr(db, table, operation, fields, values, step=['1'], index_name=None, limit=0, offset=0, return_original=False)

Decrements row(s) that meet conditions defined by operation, fields values in a given table.

Parameters:

• db (string) – database name
• table (string) – table name
• operation (string) – logical comparison operation to use over columns. Currently allowed operations are defined in FIND_OPERATIONS. Only one operation is allowed per call.
• fields (list) – list of table’s fields to use, ordered by inclusion into the index.
• values (list) – values to compare to, ordered the same way as items in fields.
• step (list) – list of decrement steps, ordered the same way as items in fields.
• index_name (string or None) – name of the index to open, default is PRIMARY.
• limit (integer) – optional limit of rows. Default is one row. In case multiple rows are expected to be updated, limit must be set explicitly, HS wont update all found rows by default.
• offset (integer) – optional offset of rows to search for.
• return_original (bool) – if set to True, method will return a list of original values in affected rows. Otherwise - number of affected rows (this is default behaviour).

Return type:

int or list

delete(db, table, operation, fields, values, index_name=None, limit=0, offset=0, return_original=False)

Delete row(s) that meet conditions defined by operation, fields values in a given table.

Parameters:

• db (string) – database name
• table (string) – table name
• operation (string) – logical comparison operation to use over columns. Currently allowed operations are defined in FIND_OPERATIONS. Only one operation is allowed per call.
• fields (list) – list of table’s fields to use, ordered by inclusion into the index.
• values (list) – values to compare to, ordered the same way as items in fields.
• index_name (string or None) – name of the index to open, default is PRIMARY.
• limit (integer) – optional limit of rows. Default is one row. In case multiple rows are expected to be deleted, limit must be set explicitly, HS wont delete all found rows by default.
• offset (integer) – optional offset of rows to search for.
• return_original (bool) – if set to True, method will return a list of original values in affected rows. Otherwise - number of affected rows (this is default behaviour).

Return type:

int or list

get(db, table, fields, value)

A wrapper over find() that gets a single row with a single field look up.

Returns a list of pairs. First item in pair is field name, second is its value.

If multiple result rows, different comparison operation or composite indexes are needed please use find() instead.

Parameters:

• db (string) – database name.
• table (string) – table name.
• fields (list) – list of table’s fields to get, ordered by inclusion into the index. First item must always be the look up field.
• value (string) – a look up value.

Return type:

list of tuples

purge()

Purges all read and write connections. All requests after that operation will open new connections, index caches will be cleaned too.