OpenAPI definition

AI Lake - Pipe Tables

Endpoint:/

(BETA) Run ANALYZE TABLE for tables in a database instance

POST

/api/v1/ailake/database/instances/{instanceId}/analyzeStatistics

(BETA) Collects CBO statistics for tables in a StarRocks database. Works for both internal (native/PIPE) and external (Iceberg) catalogs. If tableNames is empty, all tables are analyzed.

(BETA) Run ANALYZE TABLE for tables in a database instance › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.

(BETA) Run ANALYZE TABLE for tables in a database instance › Headers

operation-idstring

(BETA) Run ANALYZE TABLE for tables in a database instance › Request Body

tableNamesstring[]
Table names to analyze. If empty or null, all tables in the database are analyzed.

(BETA) Run ANALYZE TABLE for tables in a database instance › Responses

Statistics analysis scheduled.

(BETA) List AI Lake pipe tables

GET

/api/v1/ailake/database/instances/{instanceId}/pipeTables

(BETA) Lists active pipe tables in the given AI Lake database instance.

(BETA) List AI Lake pipe tables › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.

(BETA) List AI Lake pipe tables › query Parameters

pagestring · min: 0
Zero-based page number.
Zero-based page number.
Default: 0
sizestring · min: 1 · max: 500
Number of items per page.
Number of items per page.
Default: 50
metaIncludestring[] · unique

(BETA) List AI Lake pipe tables › Responses

AI Lake pipe tables successfully retrieved

dataobject[] · required
Array of resource items
linksobject · required
Links object for a list response

metaobject
Metadata for a paginated list response

(BETA) Create a new AI Lake pipe table

POST

/api/v1/ailake/database/instances/{instanceId}/pipeTables

(BETA) Creates a pipe-backed OLAP table in the given AI Lake database instance. Infers schema from parquet files. Returns an operation-id header the client can use to poll for progress.

(BETA) Create a new AI Lake pipe table › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.

(BETA) Create a new AI Lake pipe table › Headers

operation-idstring

(BETA) Create a new AI Lake pipe table › Request Body

pathPrefixstring · required
Path prefix to the parquet files (e.g. 'my-dataset/year=2024/'). All parquet files must be at a uniform depth under the prefix — either all directly under the prefix, or all under a consistent Hive partition hierarchy (e.g. year=2024/month=01/). Mixed layouts (files at multiple depths) are not supported.
sourceStorageNamestring · required
Name of the pre-configured S3/MinIO ObjectStorage source
tableNamestring · required
Name of the OLAP table to create. Must match ^[a-z][a-z0-9_-]{0,62}$

aggregationOverridesobject
Maps non-key column names to their StarRocks aggregation function (SUM, MIN, MAX, REPLACE, REPLACE_IF_NOT_NULL, HLL_UNION, BITMAP_UNION, PERCENTILE_UNION). Required for every non-key column when keyConfig type is 'aggregate'. Ignored for other key types.
columnExpressionsobject
Per-target-column projection overrides. Each entry emits <function>(<column>) AS <key> in the SELECT list of the generated CREATE PIPE ... AS INSERT; keys absent from the map are projected as-is. Required for AGGREGATE-KEY tables that include native HLL columns (StarRocks rejects raw VARBINARY into HLL columns).
columnOverridesobject
Override inferred column types. Maps column names to SQL type strings (e.g. {"year": "INT", "event_date": "DATE"}). Applied after parquet schema inference.
distributionConfigobject
Distribution configuration for the OLAP table.
keyConfigobject
Key configuration for the table data model.
maxVarcharLengthinteger · int32
Cap VARCHAR(N) to this length when N exceeds it. 0 = no cap.
partitionConfigobject
Partition configuration for the table.
pollingIntervalSecondsinteger · int32
How often (in seconds) the pipe polls for new files. 0 or null = use server default.
tablePropertiesobject
CREATE TABLE PROPERTIES key-value pairs. Defaults to {"replication_num": "1"}.

(BETA) Create a new AI Lake pipe table › Responses

Accepted

(BETA) Get an AI Lake pipe table

GET

/api/v1/ailake/database/instances/{instanceId}/pipeTables/{tableName}

(BETA) Returns full details of the specified pipe table.

(BETA) Get an AI Lake pipe table › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.
tableNamestring · required
Pipe table name.
Pipe table name.

(BETA) Get an AI Lake pipe table › Responses

AI Lake pipe table successfully retrieved

dataobject · required
A single JSON:API resource item
linksobject · required
Links object for a list response

(BETA) Delete an AI Lake pipe table

DELETE

/api/v1/ailake/database/instances/{instanceId}/pipeTables/{tableName}

(BETA) Drops the pipe and OLAP table and removes the record. Returns an operation-id header the client can use to poll for progress.

(BETA) Delete an AI Lake pipe table › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.
tableNamestring · required
Pipe table name.
Pipe table name.

(BETA) Delete an AI Lake pipe table › Headers

operation-idstring

(BETA) Delete an AI Lake pipe table › Responses

Accepted

(BETA) Refresh a pipe table partition

POST

/api/v1/ailake/database/instances/{instanceId}/pipeTables/{tableName}/refresh

(BETA) Deletes all rows for the specified Hive partition and re-loads them from S3. Use after overwriting a partition file in object storage with corrected data. Returns an operation-id header the client can use to poll for progress.

(BETA) Refresh a pipe table partition › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.
tableNamestring · required
Pipe table name.
Pipe table name.

(BETA) Refresh a pipe table partition › Headers

operation-idstring

(BETA) Refresh a pipe table partition › Request Body

partitionSpecobject · required
Partition column values identifying the partition to refresh. Keys must match the table's partition_columns exactly. Example: {"date": "2026-01-01"}

(BETA) Refresh a pipe table partition › Responses

Accepted

AI Lake - Databases AI Lake - Services & Operations

(BETA) Run ANALYZE TABLE for tables in a database instance

POST

/api/v1/ailake/database/instances/{instanceId}/analyzeStatistics

(BETA) Collects CBO statistics for tables in a StarRocks database. Works for both internal (native/PIPE) and external (Iceberg) catalogs. If tableNames is empty, all tables are analyzed.

(BETA) Run ANALYZE TABLE for tables in a database instance › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.

(BETA) Run ANALYZE TABLE for tables in a database instance › Headers

operation-idstring

(BETA) Run ANALYZE TABLE for tables in a database instance › Request Body

tableNamesstring[]
Table names to analyze. If empty or null, all tables in the database are analyzed.

(BETA) Run ANALYZE TABLE for tables in a database instance › Responses

Statistics analysis scheduled.

(BETA) List AI Lake pipe tables

GET

/api/v1/ailake/database/instances/{instanceId}/pipeTables

(BETA) Lists active pipe tables in the given AI Lake database instance.

(BETA) List AI Lake pipe tables › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.

(BETA) List AI Lake pipe tables › query Parameters

pagestring · min: 0
Zero-based page number.
Zero-based page number.
Default: 0
sizestring · min: 1 · max: 500
Number of items per page.
Number of items per page.
Default: 50
metaIncludestring[] · unique

(BETA) List AI Lake pipe tables › Responses

AI Lake pipe tables successfully retrieved

dataobject[] · required
Array of resource items
linksobject · required
Links object for a list response

metaobject
Metadata for a paginated list response

(BETA) Create a new AI Lake pipe table

POST

/api/v1/ailake/database/instances/{instanceId}/pipeTables

(BETA) Creates a pipe-backed OLAP table in the given AI Lake database instance. Infers schema from parquet files. Returns an operation-id header the client can use to poll for progress.

(BETA) Create a new AI Lake pipe table › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.

(BETA) Create a new AI Lake pipe table › Headers

operation-idstring

(BETA) Create a new AI Lake pipe table › Request Body

pathPrefixstring · required
Path prefix to the parquet files (e.g. 'my-dataset/year=2024/'). All parquet files must be at a uniform depth under the prefix — either all directly under the prefix, or all under a consistent Hive partition hierarchy (e.g. year=2024/month=01/). Mixed layouts (files at multiple depths) are not supported.
sourceStorageNamestring · required
Name of the pre-configured S3/MinIO ObjectStorage source
tableNamestring · required
Name of the OLAP table to create. Must match ^[a-z][a-z0-9_-]{0,62}$

aggregationOverridesobject
Maps non-key column names to their StarRocks aggregation function (SUM, MIN, MAX, REPLACE, REPLACE_IF_NOT_NULL, HLL_UNION, BITMAP_UNION, PERCENTILE_UNION). Required for every non-key column when keyConfig type is 'aggregate'. Ignored for other key types.
columnExpressionsobject
Per-target-column projection overrides. Each entry emits <function>(<column>) AS <key> in the SELECT list of the generated CREATE PIPE ... AS INSERT; keys absent from the map are projected as-is. Required for AGGREGATE-KEY tables that include native HLL columns (StarRocks rejects raw VARBINARY into HLL columns).
columnOverridesobject
Override inferred column types. Maps column names to SQL type strings (e.g. {"year": "INT", "event_date": "DATE"}). Applied after parquet schema inference.
distributionConfigobject
Distribution configuration for the OLAP table.
keyConfigobject
Key configuration for the table data model.
maxVarcharLengthinteger · int32
Cap VARCHAR(N) to this length when N exceeds it. 0 = no cap.
partitionConfigobject
Partition configuration for the table.
pollingIntervalSecondsinteger · int32
How often (in seconds) the pipe polls for new files. 0 or null = use server default.
tablePropertiesobject
CREATE TABLE PROPERTIES key-value pairs. Defaults to {"replication_num": "1"}.

(BETA) Create a new AI Lake pipe table › Responses

Accepted

(BETA) Get an AI Lake pipe table

GET

/api/v1/ailake/database/instances/{instanceId}/pipeTables/{tableName}

(BETA) Returns full details of the specified pipe table.

(BETA) Get an AI Lake pipe table › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.
tableNamestring · required
Pipe table name.
Pipe table name.

(BETA) Get an AI Lake pipe table › Responses

AI Lake pipe table successfully retrieved

dataobject · required
A single JSON:API resource item
linksobject · required
Links object for a list response

(BETA) Delete an AI Lake pipe table

DELETE

/api/v1/ailake/database/instances/{instanceId}/pipeTables/{tableName}

(BETA) Drops the pipe and OLAP table and removes the record. Returns an operation-id header the client can use to poll for progress.

(BETA) Delete an AI Lake pipe table › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.
tableNamestring · required
Pipe table name.
Pipe table name.

(BETA) Delete an AI Lake pipe table › Headers

operation-idstring

(BETA) Delete an AI Lake pipe table › Responses

Accepted

(BETA) Refresh a pipe table partition

POST

/api/v1/ailake/database/instances/{instanceId}/pipeTables/{tableName}/refresh

(BETA) Refresh a pipe table partition › path Parameters

instanceIdstring · required
Database instance identifier. Accepts the database name (preferred) or UUID.
Database instance identifier. Accepts the database name (preferred) or UUID.
tableNamestring · required
Pipe table name.
Pipe table name.

(BETA) Refresh a pipe table partition › Headers

operation-idstring

(BETA) Refresh a pipe table partition › Request Body

partitionSpecobject · required
Partition column values identifying the partition to refresh. Keys must match the table's partition_columns exactly. Example: {"date": "2026-01-01"}

(BETA) Refresh a pipe table partition › Responses

Accepted