Struct Table

pub struct Table { /* private fields */ }

Iceberg table

Implementations

impl Table

Public interface of the table.

pub fn builder() -> CreateTableBuilder

Creates a new table builder with default configuration

Returns a CreateTableBuilder initialized with default properties:

• WRITE_PARQUET_COMPRESSION_CODEC: “zstd”
• WRITE_PARQUET_COMPRESSION_LEVEL: “1”
• WRITE_OBJECT_STORAGE_ENABLED: “false”

Returns

• CreateTableBuilder - A builder for configuring and creating a new table

Example

use iceberg_rust::table::Table;

let builder = Table::builder()
    .with_name("my_table")
    .with_schema(schema);

pub async fn new( identifier: Identifier, catalog: Arc<dyn Catalog>, metadata: TableMetadata, ) -> Result<Self, Error>

Creates a new table instance with the given identifier, catalog and metadata

Arguments

• identifier - The unique identifier for this table in the catalog
• catalog - The catalog that this table belongs to
• metadata - The table’s metadata containing schema, partitioning, etc.

Returns

• Result<Table, Error> - The newly created table instance or an error

This is typically called by catalog implementations rather than directly by users. For creating new tables, use Table::builder() instead.

pub fn identifier(&self) -> &Identifier

Returns the unique identifier for this table in the catalog

The identifier contains both the namespace and name that uniquely identify this table within its catalog.

Returns

• &Identifier - A reference to this table’s identifier

pub fn catalog(&self) -> Arc<dyn Catalog>

Returns a reference to the catalog containing this table

The returned catalog reference is wrapped in an Arc to allow shared ownership and thread-safe access to the catalog implementation.

Returns

• Arc<dyn Catalog> - A thread-safe reference to the table’s catalog

pub fn object_store(&self) -> Arc<dyn ObjectStore>

Returns the object store for this table’s location

The object store is determined by the table’s location and is used for reading and writing table data files. The returned store is wrapped in an Arc to allow shared ownership and thread-safe access.

Returns

• Arc<dyn ObjectStore> - A thread-safe reference to the table’s object store

pub fn current_schema(&self, branch: Option<&str>) -> Result<&Schema, Error>

Returns the current schema for this table, optionally for a specific branch

Arguments

• branch - Optional branch name to get the schema for. If None, returns the main branch schema

Returns

• Result<&Schema, Error> - The current schema if found, or an error if the schema cannot be found

Errors

Returns an error if the schema ID cannot be found in the table metadata

pub fn metadata(&self) -> &TableMetadata

Returns a reference to this table’s metadata

The metadata contains all table information including:

• Schema definitions
• Partition specifications
• Snapshots
• Sort orders
• Table properties

Returns

• &TableMetadata - A reference to the table’s metadata

pub fn into_metadata(self) -> TableMetadata

Consumes the table and returns its metadata

This method takes ownership of the table instance and returns just the underlying TableMetadata. This is useful when you no longer need the table instance but want to retain its metadata.

Returns

• TableMetadata - The owned metadata from this table

pub async fn manifests( &self, start: Option<i64>, end: Option<i64>, ) -> Result<Vec<ManifestListEntry>, Error>

Returns manifest list entries for snapshots within the given sequence range

Arguments

• start - Optional starting snapshot ID (exclusive). If None, includes from the beginning
• end - Optional ending snapshot ID (inclusive). If None, uses the current snapshot

Returns

• Result<Vec<ManifestListEntry>, Error> - Vector of manifest entries in the range, or an empty vector if no current snapshot exists

Errors

Returns an error if:

• The end snapshot ID is invalid
• Reading the manifest list fails

pub async fn datafiles<'a>( &self, manifests: &'a [ManifestListEntry], filter: Option<Vec<bool>>, sequence_number_range: (Option<i64>, Option<i64>), ) -> Result<impl Stream<Item = Result<ManifestEntry, Error>> + 'a, Error>

Returns a stream of manifest entries for the given manifest list entries

Arguments

• manifests - List of manifest entries to read data files from
• filter - Optional vector of boolean predicates to filter manifest entries
• sequence_number_range - Tuple of (start, end) sequence numbers to filter entries by

Returns

• Result<impl Stream<Item = Result<ManifestEntry, Error>>, Error> - Stream of manifest entries that match the given filters

Type Parameters

• 'a - Lifetime of the manifest list entries reference

Errors

Returns an error if reading any manifest file fails

pub async fn datafiles_contains_delete( &self, start: Option<i64>, end: Option<i64>, ) -> Result<bool, Error>

Check if datafiles contain deletes

pub fn new_transaction(&mut self, branch: Option<&str>) -> TableTransaction<'_>

Creates a new transaction for atomic modifications to this table

Arguments

• branch - Optional branch name to create the transaction for. If None, uses the main branch

Returns

• TableTransaction - A new transaction that can be used to atomically modify this table

The transaction must be committed for any changes to take effect. Multiple operations can be chained within a single transaction.

Trait Implementations

impl Clone for Table

fn clone(&self) -> Table

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Table

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

Formats the value using the given formatter.