User friendly container for Google Cloud Bigtable Table.

class, instance)[source]#

Bases: object

Representation of a Google Cloud Bigtable Table.


We don’t define any properties on a table other than the name. The only other fields are column_families and granularity, The column_families are not stored locally and granularity is an enum with only one value.

We can use a Table to:

  • table_id (str) – The ID of the table.
  • instance (Instance) – The instance that owns the table.
column_family(column_family_id, gc_rule=None)[source]#

Factory to create a column family associated with this table.

  • column_family_id (str) – The ID of the column family. Must be of the form [_a-zA-Z0-9][-_.a-zA-Z0-9]*.
  • gc_rule (GarbageCollectionRule) – (Optional) The garbage collection settings for this column family.
Return type:



A column family owned by this table.

create(initial_split_keys=None, column_families=())[source]#

Creates this table.


A create request returns a _generated.table_pb2.Table but we don’t use this response.

  • initial_split_keys (list) – (Optional) List of row keys that will be used to initially split the table into several tablets (Tablets are similar to HBase regions). Given two split keys, "s1" and "s2", three tablets will be created, spanning the key ranges: [, s1), [s1, s2), [s2, ).
  • column_families (list) – (Optional) List or other iterable of ColumnFamily instances.

Delete this table.


List the column families owned by this table.

Return type:dict
Returns:Dictionary of column families attached to this table. Keys are strings (column family names) and values are ColumnFamily instances.
Raises:ValueError if the column family name from the response does not agree with the computed name from the column family ID.

Table name used in requests.


This property will not change if table_id does not, but the return value is not cached.

The table name is of the form

Return type:str
Returns:The table name.
read_row(row_key, filter_=None)[source]#

Read a single row from this table.

  • row_key (bytes) – The key of the row to read from.
  • filter (RowFilter) – (Optional) The filter to apply to the contents of the row. If unset, returns the entire row.
Return type:

PartialRowData, NoneType


The contents of the row if any chunks were returned in the response, otherwise None.


ValueError if a commit row chunk is never encountered.

read_rows(start_key=None, end_key=None, limit=None, filter_=None)[source]#

Read rows from this table.

  • start_key (bytes) – (Optional) The beginning of a range of row keys to read from. The range will include start_key. If left empty, will be interpreted as the empty string.
  • end_key (bytes) – (Optional) The end of a range of row keys to read from. The range will not include end_key. If left empty, will be interpreted as an infinite string.
  • limit (int) – (Optional) The read will terminate after committing to N rows’ worth of results. The default (zero) is to return all results.
  • filter (RowFilter) – (Optional) The filter to apply to the contents of the specified row(s). If unset, reads every column in each row.
Return type:



A PartialRowsData convenience wrapper for consuming the streamed results.

row(row_key, filter_=None, append=False)[source]#

Factory to create a row associated with this table.


At most one of filter_ and append can be used in a Row.

  • row_key (bytes) – The key for the row being created.
  • filter (RowFilter) – (Optional) Filter to be used for conditional mutations. See DirectRow for more details.
  • append (bool) – (Optional) Flag to determine if the row should be used for append mutations.
Return type:



A row owned by this table.


ValueError if both filter_ and append are used.


Read a sample of row keys in the table.

The returned row keys will delimit contiguous sections of the table of approximately equal size, which can be used to break up the data for distributed tasks like mapreduces.

The elements in the iterator are a SampleRowKeys response and they have the properties offset_bytes and row_key. They occur in sorted order. The table might have contents before the first row key in the list and after the last one, but a key containing the empty string indicates “end of table” and will be the last response given, if present.


Row keys in this list may not have ever been written to or read from, and users should therefore not make any assumptions about the row key structure that are specific to their use case.

The offset_bytes field on a response indicates the approximate total storage space used by all rows in the table which precede row_key. Buffering the contents of all rows between two subsequent samples would require space roughly equal to the difference in their offset_bytes fields.

Return type:GrpcRendezvous
Returns:A cancel-able iterator. Can be consumed by calling next() or by casting to a list and can be cancelled by calling cancel().