Struct SinkCatalog

pub struct SinkCatalog {

    pub id: SinkId,
    pub schema_id: SchemaId,
    pub database_id: DatabaseId,
    pub name: String,
    pub definition: String,
    columns: Vec<ColumnCatalog>,
    pub plan_pk: Vec<ColumnOrder>,
    pub downstream_pk: Vec<usize>,
    pub distribution_key: Vec<usize>,
    pub properties: BTreeMap<String, String>,
    pub owner: UserId,
    pub sink_type: SinkType,
    pub format_desc: Option<SinkFormatDesc>,
    pub connection_id: Option<ConnectionId>,
    pub created_at_epoch: Option<Epoch>,
    pub initialized_at_epoch: Option<Epoch>,
    pub db_name: String,
    pub sink_from_name: String,
    pub target_table: Option<TableId>,
    pub created_at_cluster_version: Option<String>,
    pub initialized_at_cluster_version: Option<String>,
    pub create_type: CreateType,
    pub secret_refs: BTreeMap<String, PbSecretRef>,
    pub original_target_columns: Vec<ColumnCatalog>,
}

the catalog of the sink. There are two kind of schema here. The full schema is all columns stored in the column which is the sink executor/fragment’s output schema. The visible schema contains the columns whose is_hidden is false, which is the columns sink out to the external system. The distribution key and all other keys are indexed in the full schema.

Fields

id: SinkId

Id of the sink.

schema_id: SchemaId

Schema of the sink.

database_id: DatabaseId

Database of the sink.

name: String

Name of the sink.

definition: String

The full CREATE SINK definition of the sink.

columns: Vec<ColumnCatalog>

All columns of the sink. Note that this is NOT sorted by columnId in the vector.

plan_pk: Vec<ColumnOrder>

Primary keys of the sink. Derived by the frontend.

downstream_pk: Vec<usize>

User-defined primary key indices for upsert sink.

distribution_key: Vec<usize>

Distribution key indices of the sink. For example, if distribution_key = [1, 2], then the distribution keys will be columns[1] and columns[2].

properties: BTreeMap<String, String>

The properties of the sink.

owner: UserId

Owner of the sink.

sink_type: SinkType

format_desc: Option<SinkFormatDesc>

connection_id: Option<ConnectionId>

Sink may use a privatelink connection to connect to the downstream system.

created_at_epoch: Option<Epoch>

initialized_at_epoch: Option<Epoch>

db_name: String

Name of the database

sink_from_name: String

Name for the table info for Debezium sink

target_table: Option<TableId>

created_at_cluster_version: Option<String>

initialized_at_cluster_version: Option<String>

create_type: CreateType

secret_refs: BTreeMap<String, PbSecretRef>

The secret reference for the sink, mapping from property name to secret id.

original_target_columns: Vec<ColumnCatalog>

Only for the sink whose target is a table. Columns of the target table when the sink is created. At this point all the default columns of the target table are all handled by the project operator in the sink plan.

Implementations

impl SinkCatalog

pub fn to_proto(&self) -> PbSink

pub fn create_sql(&self) -> String

Returns the SQL statement that can be used to create this sink.

pub fn visible_columns(&self) -> impl Iterator<Item = &ColumnCatalog>

pub fn full_columns(&self) -> &[ColumnCatalog]

pub fn full_schema(&self) -> Schema

pub fn visible_schema(&self) -> Schema

pub fn downstream_pk_indices(&self) -> Vec<usize>

pub fn unique_identity(&self) -> String

Trait Implementations

impl Clone for SinkCatalog

fn clone(&self) -> SinkCatalog

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for SinkCatalog

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl From<&Sink> for SinkCatalog

fn from(pb: &PbSink) -> Self

Converts to this type from the input type.

impl From<Sink> for SinkCatalog

fn from(pb: PbSink) -> Self

Converts to this type from the input type.