Database Reference
In-Depth Information
all the rows added since the start time but possibly additional rows that were
inserted at an earlier time.
When we introduced these time slicing references, we mentioned that they
could be used in most places where a table name is expected. A notable
exception is in mutation operations. For example, you cannot use a time
slice as the destination for a load or query job. Deleting a time slice would
be useful but is not currently permitted. It is conceivable that some of these
restrictions may be lifted in the future, but for now you are restricted to
using time slices as sources of data.
Selecting Tables
The list operation on the REST table resource allows clients to paginate
through the list of tables in a dataset. If there are only a handful of tables
in a dataset, this works great. However, finding one or a few tables in a
large dataset can be cumbersome using this paginated approach. To simplify
locating tables, the service provides a metatable that contains a record for
each table in the dataset:
<
dataset
>.__TABLES__
Listing 11.4
offers a quick tour of its features.
Listing 11.4
Using the dataset metatable
$
bq show ch11.__TABLES
__
Table 317752944021:ch11.__TABLES__
Last modified Schema
Total Rows Total B>
----------------- --------------------------------
------------ ------->
13 Nov 22:44:52 |- project_id: string 4
|- dataset_id: string
|- table_id: string
|- creation_time: integer
|- last_modified_time: integer
|- row_count: integer
|- size_bytes: integer
|- type: integer
