package google.cloud.bigquery.storage.v1beta1

Get desktop application:
View/edit binary Protocol Buffers messages

BigQuery storage API. The BigQuery storage API can be used to read data stored in BigQuery.

• rpc CreateReadSession (CreateReadSessionRequest, ReadSession)

  storage.proto:56

  Creates a new read session. A read session divides the contents of a BigQuery table into one or more streams, which can then be used to read data from the table. The read session also specifies properties of the data to be read, such as a list of columns or a push-down filter describing the rows to be returned. A particular row can be read by at most one stream. When the caller has reached the end of each stream in the session, then all the data in the table has been read. Read sessions automatically expire 24 hours after they are created and do not require manual clean-up by the caller.

  message CreateReadSessionRequest

  storage.proto:200

  Creates a new read session, which may include additional options such as requested parallelism, projection filters and constraints.

  • optional TableReference table_reference = 1

    Required. Reference to the table to read.

  • string parent = 6

    Required. String of the form `projects/{project_id}` indicating the project this ReadSession is associated with. This is the project that will be billed for usage.

  • optional TableModifiers table_modifiers = 2

    Any modifiers to the Table (e.g. snapshot timestamp).

  • int32 requested_streams = 3

    Initial number of streams. If unset or 0, we will provide a value of streams so as to produce reasonable throughput. Must be non-negative. The number of streams may be lower than the requested number, depending on the amount parallelism that is reasonable for the table and the maximum amount of parallelism allowed by the system. Streams must be read starting from offset 0.

  • optional TableReadOptions read_options = 4

    Read options for this session (e.g. column selection, filters).

  • DataFormat format = 5

    Data output format. Currently default to Avro.

  • ShardingStrategy sharding_strategy = 7

    The strategy to use for distributing data among multiple streams. Currently defaults to liquid sharding.

• rpc ReadRows (ReadRowsRequest, stream ReadRowsResponse)

  storage.proto:77

  Reads rows from the table in the format prescribed by the read session. Each response contains one or more table rows, up to a maximum of 10 MiB per response; read requests which attempt to read individual rows larger than this will fail. Each request also returns a set of stream statistics reflecting the estimated total number of rows in the read stream. This number is computed based on the total table size and the number of active streams in the read session, and may change as other streams continue to read data.

  message ReadRowsRequest

  storage.proto:265

  Requesting row data via `ReadRows` must provide Stream position information.

  • optional StreamPosition read_position = 1

    Required. Identifier of the position in the stream to start reading from. The offset requested must be less than the last row read from ReadRows. Requesting a larger offset is undefined.

  message ReadRowsResponse

  storage.proto:330

  Response from calling `ReadRows` may include row data, progress and throttling information.

  • oneof rows

    Row data is returned in format specified during session creation.

    • AvroRows avro_rows = 3

      Serialized row data in AVRO format.

    • ArrowRecordBatch arrow_record_batch = 4

      Serialized row data in Arrow RecordBatch format.

  • int64 row_count = 6

    Number of serialized rows in the rows block. This value is recorded here, in addition to the row_count values in the output-specific messages in `rows`, so that code which needs to record progress through the stream can do so in an output format-independent way.

  • optional StreamStatus status = 2

    Estimated stream statistics.

  • optional ThrottleStatus throttle_status = 5

    Throttling status. If unset, the latest response still describes the current throttling status.

• rpc BatchCreateReadSessionStreams (BatchCreateReadSessionStreamsRequest, BatchCreateReadSessionStreamsResponse)

  storage.proto:87

  Creates additional streams for a ReadSession. This API can be used to dynamically adjust the parallelism of a batch processing task upwards by adding additional workers.

  message BatchCreateReadSessionStreamsRequest

  storage.proto:356

  Information needed to request additional streams for an established read session.

  • optional ReadSession session = 1

    Required. Must be a non-expired session obtained from a call to CreateReadSession. Only the name field needs to be set.

  • int32 requested_streams = 2

    Required. Number of new streams requested. Must be positive. Number of added streams may be less than this, see CreateReadSessionRequest for more information.

  message BatchCreateReadSessionStreamsResponse

  storage.proto:369

  The response from `BatchCreateReadSessionStreams` returns the stream identifiers for the newly created streams.

  • repeated Stream streams = 1

    Newly added streams.

• rpc FinalizeStream (FinalizeStreamRequest, protobuf.Empty)

  storage.proto:109

  Triggers the graceful termination of a single stream in a ReadSession. This API can be used to dynamically adjust the parallelism of a batch processing task downwards without losing data. This API does not delete the stream -- it remains visible in the ReadSession, and any data processed by the stream is not released to other streams. However, no additional data will be assigned to the stream once this call completes. Callers must continue reading data on the stream until the end of the stream is reached so that data which has already been assigned to the stream will be processed. This method will return an error if there are no other live streams in the Session, or if SplitReadStream() has been called on the given Stream.

  message FinalizeStreamRequest

  storage.proto:375

  Request information for invoking `FinalizeStream`.

  • optional Stream stream = 2

    Stream to finalize.

• rpc SplitReadStream (SplitReadStreamRequest, SplitReadStreamResponse)

  storage.proto:130

  Splits a given read stream into two Streams. These streams are referred to as the primary and the residual of the split. The original stream can still be read from in the same manner as before. Both of the returned streams can also be read from, and the total rows return by both child streams will be the same as the rows read from the original stream. Moreover, the two child streams will be allocated back to back in the original Stream. Concretely, it is guaranteed that for streams Original, Primary, and Residual, that Original[0-j] = Primary[0-j] and Original[j-n] = Residual[0-m] once the streams have been read to completion. This method is guaranteed to be idempotent.

  message SplitReadStreamRequest

  storage.proto:381

  Request information for `SplitReadStream`.

  • optional Stream original_stream = 1

    Stream to split.

  • float fraction = 2

    A value in the range (0.0, 1.0) that specifies the fractional point at which the original stream should be split. The actual split point is evaluated on pre-filtered rows, so if a filter is provided, then there is no guarantee that the division of the rows between the new child streams will be proportional to this fractional value. Additionally, because the server-side unit for assigning data is collections of rows, this fraction will always map to to a data storage boundary on the server side.

  message SplitReadStreamResponse

  storage.proto:396

  Response from `SplitReadStream`.

  • optional Stream primary_stream = 1

    Primary stream, which contains the beginning portion of |original_stream|. An empty value indicates that the original stream can no longer be split.

  • optional Stream remainder_stream = 2

    Remainder stream, which contains the tail of |original_stream|. An empty value indicates that the original stream can no longer be split.

Arrow RecordBatch.

Used in: ReadRowsResponse

• bytes serialized_record_batch = 1

  IPC serialized Arrow RecordBatch.

• int64 row_count = 2

  The count of rows in the returning block.

Arrow schema.

Used in: ReadSession

• bytes serialized_schema = 1

  IPC serialized Arrow schema.

Avro rows.

Used in: ReadRowsResponse

• bytes serialized_binary_rows = 1

  Binary serialized rows in a block.

• int64 row_count = 2

  The count of rows in the returning block.

Avro schema.

Used in: ReadSession

• string schema = 1

  Json serialized schema, as described at https://avro.apache.org/docs/1.8.1/spec.html

Data format for input or output data.

Used in: CreateReadSessionRequest

• DATA_FORMAT_UNSPECIFIED = 0

  Data format is unspecified.

• AVRO = 1

  Avro is a standard open source row based file format. See https://avro.apache.org/ for more details.

• ARROW = 3

Used in: StreamStatus

• float at_response_start = 1

  The fraction of rows assigned to the stream that have been processed by the server so far, not including the rows in the current response message. This value, along with `at_response_end`, can be used to interpolate the progress made as the rows in the message are being processed using the following formula: `at_response_start + (at_response_end - at_response_start) * rows_processed_from_response / rows_in_response`. Note that if a filter is provided, the `at_response_end` value of the previous response may not necessarily be equal to the `at_response_start` value of the current response.

• float at_response_end = 2

  Similar to `at_response_start`, except that this value includes the rows in the current response.

Information returned from a `CreateReadSession` request.

Used as response type in: BigQueryStorage.CreateReadSession

Used as field type in: BatchCreateReadSessionStreamsRequest

• string name = 1

  Unique identifier for the session, in the form `projects/{project_id}/locations/{location}/sessions/{session_id}`.

• optional protobuf.Timestamp expire_time = 2

  Time at which the session becomes invalid. After this time, subsequent requests to read this Session will return errors.

• oneof schema

  The schema for the read. If read_options.selected_fields is set, the schema may be different from the table schema as it will only contain the selected fields.

  • AvroSchema avro_schema = 5

    Avro schema.

  • ArrowSchema arrow_schema = 6

    Arrow schema.

• repeated Stream streams = 4

  Streams associated with this session.

• optional TableReference table_reference = 7

  Table that this ReadSession is reading from.

• optional TableModifiers table_modifiers = 8

  Any modifiers which are applied when reading from the specified table.

• ShardingStrategy sharding_strategy = 9

  The strategy to use for distributing data among the streams.

Strategy for distributing data among multiple streams in a read session.

Used in: CreateReadSessionRequest, ReadSession

• SHARDING_STRATEGY_UNSPECIFIED = 0

  Same as LIQUID.

• LIQUID = 1

  Assigns data to each stream based on the client's read rate. The faster the client reads from a stream, the more data is assigned to the stream. In this strategy, it's possible to read all data from a single stream even if there are other streams present.

• BALANCED = 2

  Assigns data to each stream such that roughly the same number of rows can be read from each stream. Because the server-side unit for assigning data is collections of rows, the API does not guarantee that each stream will return the same number or rows. Additionally, the limits are enforced based on the number of pre-filtering rows, so some filters can lead to lopsided assignments.

Information about a single data stream within a read session.

Used in: BatchCreateReadSessionStreamsResponse, FinalizeStreamRequest, ReadSession, SplitReadStreamRequest, SplitReadStreamResponse, StreamPosition

• string name = 1

  Name of the stream, in the form `projects/{project_id}/locations/{location}/streams/{stream_id}`.

Expresses a point within a given stream using an offset position.

Used in: ReadRowsRequest

• optional Stream stream = 1

  Identifier for a given Stream.

• int64 offset = 2

  Position in the stream.

Progress information for a given Stream.

Used in: ReadRowsResponse

• int64 estimated_row_count = 1

  Number of estimated rows in the current stream. May change over time as different readers in the stream progress at rates which are relatively fast or slow.

• float fraction_consumed = 2

  A value in the range [0.0, 1.0] that represents the fraction of rows assigned to this stream that have been processed by the server. In the presence of read filters, the server may process more rows than it returns, so this value reflects progress through the pre-filtering rows. This value is only populated for sessions created through the BALANCED sharding strategy.

• optional Progress progress = 4

  Represents the progress of the current stream. Note: This value is under development and should not be used. Use `fraction_consumed` instead.

• bool is_splittable = 3

  Whether this stream can be split. For sessions that use the LIQUID sharding strategy, this value is always false. For BALANCED sessions, this value is false when enough data have been read such that no more splits are possible at that point or beyond. For small tables or streams that are the result of a chain of splits, this value may never be true.

All fields in this message optional.

Used in: CreateReadSessionRequest, ReadSession

• optional protobuf.Timestamp snapshot_time = 1

  The snapshot time of the table. If not set, interpreted as now.

Options dictating how we read a table.

Used in: CreateReadSessionRequest

• repeated string selected_fields = 1

  Optional. Names of the fields in the table that should be read. If empty, all fields will be read. If the specified field is a nested field, all the sub-fields in the field will be selected. The output field order is unrelated to the order of fields in selected_fields.

• string row_restriction = 2

  Optional. SQL text filtering statement, similar to a WHERE clause in a query. Currently, only a single predicate that is a comparison between a column and a constant value is supported. Aggregates are not supported. Examples: "int_field > 5" "date_field = CAST('2014-9-27' as DATE)" "nullable_field is not NULL" "st_equals(geo_field, st_geofromtext("POINT(2, 2)"))" "numeric_field BETWEEN 1.0 AND 5.0"

Table reference that includes just the 3 strings needed to identify a table.

Used in: CreateReadSessionRequest, ReadSession

• string project_id = 1

  The assigned project ID of the project.

• string dataset_id = 2

  The ID of the dataset in the above project.

• string table_id = 3

  The ID of the table in the above dataset.

Information on if the current connection is being throttled.

Used in: ReadRowsResponse

• int32 throttle_percent = 1

  How much this connection is being throttled. 0 is no throttling, 100 is completely throttled.