table

package
v0.2.0-rc1 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Mar 20, 2025 License: Apache-2.0 Imports: 42 Imported by: 19

Documentation

Index

Constants

View Source 
const (
	ArrowFieldDocKey = "doc"
	// Arrow schemas that are generated from the Parquet library will utilize
	// this key to identify the field id of the source Parquet field.
	// We use this when converting to Iceberg to provide field IDs
	ArrowParquetFieldIDKey = "PARQUET:field_id"
)

constants to look for as Keys in Arrow field metadata

View Source 
const (
	WriteDataPathKey                        = "write.data.path"
	WriteMetadataPathKey                    = "write.metadata.path"
	WriteObjectStorePartitionedPathsKey     = "write.object-storage.partitioned-paths"
	WriteObjectStorePartitionedPathsDefault = true
	ObjectStoreEnabledKey                   = "write.object-storage.enabled"
	ObjectStoreEnabledDefault               = false

	DefaultNameMappingKey = "schema.name-mapping.default"

	MetricsModeColumnConfPrefix    = "write.metadata.metrics.column"
	DefaultWriteMetricsModeKey     = "write.metadata.metrics.default"
	DefaultWriteMetricsModeDefault = "truncate(16)"

	ManifestMergeEnabledKey     = "commit.manifest-merge.enabled"
	ManifestMergeEnabledDefault = false

	ManifestTargetSizeBytesKey     = "commit.manifest.target-size-bytes"
	ManifestTargetSizeBytesDefault = 8 * 1024 * 1024 // 8 MB

	ManifestMinMergeCountKey     = "commit.manifest.min-count-to-merge"
	ManifestMinMergeCountDefault = 100

	WritePartitionSummaryLimitKey     = "write.summary.partition-limit"
	WritePartitionSummaryLimitDefault = 0

	MetadataDeleteAfterCommitEnabledKey     = "write.metadata.delete-after-commit.enabled"
	MetadataDeleteAfterCommitEnabledDefault = false

	MetadataPreviousVersionsMaxKey     = "write.metadata.previous-versions-max"
	MetadataPreviousVersionsMaxDefault = 100
)
View Source 
const (
	InitialSortOrderID  = 1
	UnsortedSortOrderID = 0
)
View Source 
const DefaultFormatVersion = 2
View Source 
const MainBranch = "main"
View Source 
const ScanNoLimit = -1
View Source 
const (
	ScanOptionArrowUseLargeTypes = "arrow.use_large_types"
)

Variables

View Source 
var (
	ErrInvalidMetadataFormatVersion = errors.New("invalid or missing format-version in table metadata")
	ErrInvalidMetadata              = errors.New("invalid metadata")
)
View Source 
var (
	ErrInvalidOperation = errors.New("invalid operation value")
	ErrMissingOperation = errors.New("missing operation key")
)
View Source 
var (
	ErrInvalidSortDirection = errors.New("invalid sort direction, must be 'asc' or 'desc'")
	ErrInvalidNullOrder     = errors.New("invalid null order, must be 'nulls-first' or 'nulls-last'")
)
View Source 
var ErrInvalidRefType = errors.New("invalid snapshot ref type, should be 'branch' or 'tag'")
View Source 
var UnsortedSortOrder = SortOrder{OrderID: UnsortedSortOrderID, Fields: []SortField{}}

A default Sort Order indicating no sort order at all

Functions

func ArrowSchemaToIceberg 

func ArrowSchemaToIceberg(sc *arrow.Schema, downcastNsTimestamp bool, nameMapping iceberg.NameMapping) (*iceberg.Schema, error)

func ArrowTypeToIceberg 

func ArrowTypeToIceberg(dt arrow.DataType, downcastNsTimestamp bool) (iceberg.Type, error)

func NewRemoveSnapshotRefUpdate added in v0.2.0 

func NewRemoveSnapshotRefUpdate(ref string) *removeSnapshotRefUpdate

NewRemoveSnapshotRefUpdate creates a new update that removes a snapshot reference from the table metadata.

func NewSetPropertiesUpdate added in v0.2.0 

func NewSetPropertiesUpdate(updates iceberg.Properties) *setPropertiesUpdate

NewSetPropertiesUpdate creates a new update that sets the given properties in the table metadata.

func SchemaToArrowSchema 

func SchemaToArrowSchema(sc *iceberg.Schema, metadata map[string]string, includeFieldIDs, useLargeTypes bool) (*arrow.Schema, error)

SchemaToArrowSchema converts an Iceberg schema to an Arrow schema. If the metadata parameter is non-nil, it will be included as the top-level metadata in the schema. If includeFieldIDs is true, then each field of the schema will contain a metadata key PARQUET:field_id set to the field id from the iceberg schema.

func ToRequestedSchema 

func ToRequestedSchema(ctx context.Context, requested, fileSchema *iceberg.Schema, batch arrow.Record, downcastTimestamp, includeFieldIDs, useLargeTypes bool) (arrow.Record, error)

ToRequestedSchema will construct a new record batch matching the requested iceberg schema casting columns if necessary as appropriate.

func TypeToArrowType 

func TypeToArrowType(t iceberg.Type, includeFieldIDs bool, useLargeTypes bool) (arrow.DataType, error)

TypeToArrowType converts a given iceberg type, into the equivalent Arrow data type. For dealing with nested fields (List, Struct, Map) if includeFieldIDs is true, then the child fields will contain a metadata key PARQUET:field_id set to the field id.

func VisitArrowSchema 

func VisitArrowSchema[T any](sc *arrow.Schema, visitor ArrowSchemaVisitor[T]) (res T, err error)

func WithMaxRefAgeMs added in v0.2.0 

func WithMaxRefAgeMs(maxRefAgeMs int64) setSnapshotRefOption

func WithMaxSnapshotAgeMs added in v0.2.0 

func WithMaxSnapshotAgeMs(maxSnapshotAgeMs int64) setSnapshotRefOption

func WithMinSnapshotsToKeep added in v0.2.0 

func WithMinSnapshotsToKeep(minSnapshotsToKeep int) setSnapshotRefOption

Types

type ArrowSchemaVisitor 

type ArrowSchemaVisitor[T any] interface {
	Schema(*arrow.Schema, T) T
	Struct(*arrow.StructType, []T) T
	Field(arrow.Field, T) T
	List(arrow.ListLikeType, T) T
	Map(mt *arrow.MapType, keyResult T, valueResult T) T
	Primitive(arrow.DataType) T
}

ArrowSchemaVisitor is an interface that can be implemented and used to call VisitArrowSchema for iterating

type CatalogIO added in v0.2.0 

type CatalogIO interface {
	LoadTable(context.Context, Identifier, iceberg.Properties) (*Table, error)
	CommitTable(context.Context, *Table, []Requirement, []Update) (Metadata, string, error)
}

type FileScanTask 

type FileScanTask struct {
	File          iceberg.DataFile
	DeleteFiles   []iceberg.DataFile
	Start, Length int64
}

type Identifier 

type Identifier = []string

type LocationProvider added in v0.2.0 

type LocationProvider interface {
	NewTableMetadataFileLocation(newVersion int) (string, error)
	NewMetadataLocation(metadataFileName string) string
}

func LoadLocationProvider added in v0.2.0 

func LoadLocationProvider(tableLocation string, tableProps iceberg.Properties) (LocationProvider, error)

type Metadata 

type Metadata interface {
	// Version indicates the version of this metadata, 1 for V1, 2 for V2, etc.
	Version() int
	// TableUUID returns a UUID that identifies the table, generated when the
	// table is created. Implementations must throw an exception if a table's
	// UUID does not match the expected UUID after refreshing metadata.
	TableUUID() uuid.UUID
	// Location is the table's base location. This is used by writers to determine
	// where to store data files, manifest files, and table metadata files.
	Location() string
	// LastUpdatedMillis is the timestamp in milliseconds from the unix epoch when
	// the table was last updated. Each table metadata file should update this
	// field just before writing.
	LastUpdatedMillis() int64
	// LastColumnID returns the highest assigned column ID for the table.
	// This is used to ensure fields are always assigned an unused ID when
	// evolving schemas.
	LastColumnID() int
	// Schemas returns the list of schemas, stored as objects with their
	// schema-id.
	Schemas() []*iceberg.Schema
	// CurrentSchema returns the table's current schema.
	CurrentSchema() *iceberg.Schema
	// PartitionSpecs returns the list of all partition specs in the table.
	PartitionSpecs() []iceberg.PartitionSpec
	// PartitionSpec returns the current partition spec that the table is using.
	PartitionSpec() iceberg.PartitionSpec
	// DefaultPartitionSpec is the ID of the current spec that writers should
	// use by default.
	DefaultPartitionSpec() int
	// LastPartitionSpecID is the highest assigned partition field ID across
	// all partition specs for the table. This is used to ensure partition
	// fields are always assigned an unused ID when evolving specs.
	LastPartitionSpecID() *int
	// Snapshots returns the list of valid snapshots. Valid snapshots are
	// snapshots for which all data files exist in the file system. A data
	// file must not be deleted from the file system until the last snapshot
	// in which it was listed is garbage collected.
	Snapshots() []Snapshot
	// SnapshotByID find and return a specific snapshot by its ID. Returns
	// nil if the ID is not found in the list of snapshots.
	SnapshotByID(int64) *Snapshot
	// SnapshotByName searches the list of snapshots for a snapshot with a given
	// ref name. Returns nil if there's no ref with this name for a snapshot.
	SnapshotByName(name string) *Snapshot
	// CurrentSnapshot returns the table's current snapshot.
	CurrentSnapshot() *Snapshot
	// Ref returns the snapshot ref for the main branch.
	Ref() SnapshotRef
	// Refs returns a list of snapshot name/reference pairs.
	Refs() iter.Seq2[string, SnapshotRef]
	// SnapshotLogs returns the list of snapshot logs for the table.
	SnapshotLogs() iter.Seq[SnapshotLogEntry]
	// SortOrder returns the table's current sort order, ie: the one with the
	// ID that matches the default-sort-order-id.
	SortOrder() SortOrder
	// SortOrders returns the list of sort orders in the table.
	SortOrders() []SortOrder
	// DefaultSortOrder returns the ID of the current sort order that writers
	// should use by default.
	DefaultSortOrder() int
	// Properties is a string to string map of table properties. This is used
	// to control settings that affect reading and writing and is not intended
	// to be used for arbitrary metadata. For example, commit.retry.num-retries
	// is used to control the number of commit retries.
	Properties() iceberg.Properties
	// PreviousFiles returns the list of metadata log entries for the table.
	PreviousFiles() iter.Seq[MetadataLogEntry]
	Equals(Metadata) bool

	NameMapping() iceberg.NameMapping

	LastSequenceNumber() int64
}

Metadata for an iceberg table as specified in the Iceberg spec

https://iceberg.apache.org/spec/#iceberg-table-spec

func NewMetadata added in v0.2.0 

func NewMetadata(sc *iceberg.Schema, partitions *iceberg.PartitionSpec, sortOrder SortOrder, location string, props iceberg.Properties) (Metadata, error)

NewMetadata creates a new table metadata object using the provided schema, information, generating a fresh UUID for the new table metadata. By default, this will generate a V2 table metadata, but this can be modified by adding a "format-version" property to the props map. An error will be returned if the "format-version" property exists and is not a valid version number.

func NewMetadataWithUUID added in v0.2.0 

func NewMetadataWithUUID(sc *iceberg.Schema, partitions *iceberg.PartitionSpec, sortOrder SortOrder, location string, props iceberg.Properties, tableUuid uuid.UUID) (Metadata, error)

NewMetadataWithUUID is like NewMetadata, but allows the caller to specify the UUID of the table rather than creating a new one.

func ParseMetadata 

func ParseMetadata(r io.Reader) (Metadata, error)

ParseMetadata parses json metadata provided by the passed in reader, returning an error if one is encountered.

func ParseMetadataBytes 

func ParseMetadataBytes(b []byte) (Metadata, error)

ParseMetadataBytes is like ParseMetadataString but for a byte slice.

func ParseMetadataString 

func ParseMetadataString(s string) (Metadata, error)

ParseMetadataString is like ParseMetadata, but for a string rather than an io.Reader.

type MetadataBuilder added in v0.2.0 

type MetadataBuilder struct {
	// contains filtered or unexported fields
}

func MetadataBuilderFromBase added in v0.2.0 

func MetadataBuilderFromBase(metadata Metadata) (*MetadataBuilder, error)

func NewMetadataBuilder added in v0.2.0 

func NewMetadataBuilder() (*MetadataBuilder, error)

func (*MetadataBuilder) AddPartitionSpec added in v0.2.0 

func (b *MetadataBuilder) AddPartitionSpec(spec *iceberg.PartitionSpec, initial bool) (*MetadataBuilder, error)

func (*MetadataBuilder) AddSchema added in v0.2.0 

func (b *MetadataBuilder) AddSchema(schema *iceberg.Schema, newLastColumnID int, initial bool) (*MetadataBuilder, error)

func (*MetadataBuilder) AddSnapshot added in v0.2.0 

func (b *MetadataBuilder) AddSnapshot(snapshot *Snapshot) (*MetadataBuilder, error)

func (*MetadataBuilder) AddSortOrder added in v0.2.0 

func (b *MetadataBuilder) AddSortOrder(sortOrder *SortOrder, initial bool) (*MetadataBuilder, error)

func (*MetadataBuilder) AppendMetadataLog added in v0.2.0 

func (b *MetadataBuilder) AppendMetadataLog(entry MetadataLogEntry) *MetadataBuilder

func (*MetadataBuilder) Build added in v0.2.0 

func (b *MetadataBuilder) Build() (Metadata, error)

func (*MetadataBuilder) CurrentSchema added in v0.2.0 

func (b *MetadataBuilder) CurrentSchema() *iceberg.Schema

func (*MetadataBuilder) CurrentSpec added in v0.2.0 

func (b *MetadataBuilder) CurrentSpec() iceberg.PartitionSpec

func (*MetadataBuilder) GetSchemaByID added in v0.2.0 

func (b *MetadataBuilder) GetSchemaByID(id int) (*iceberg.Schema, error)

func (*MetadataBuilder) GetSortOrderByID added in v0.2.0 

func (b *MetadataBuilder) GetSortOrderByID(id int) (*SortOrder, error)

func (*MetadataBuilder) GetSpecByID added in v0.2.0 

func (b *MetadataBuilder) GetSpecByID(id int) (*iceberg.PartitionSpec, error)

func (*MetadataBuilder) HasChanges added in v0.2.0 

func (b *MetadataBuilder) HasChanges() bool

func (*MetadataBuilder) LastUpdatedMS added in v0.2.0 

func (b *MetadataBuilder) LastUpdatedMS() int64

func (*MetadataBuilder) NameMapping added in v0.2.0 

func (b *MetadataBuilder) NameMapping() iceberg.NameMapping

func (*MetadataBuilder) RemoveProperties added in v0.2.0 

func (b *MetadataBuilder) RemoveProperties(keys []string) (*MetadataBuilder, error)

func (*MetadataBuilder) SetCurrentSchemaID added in v0.2.0 

func (b *MetadataBuilder) SetCurrentSchemaID(currentSchemaID int) (*MetadataBuilder, error)

func (*MetadataBuilder) SetDefaultSortOrderID added in v0.2.0 

func (b *MetadataBuilder) SetDefaultSortOrderID(defaultSortOrderID int) (*MetadataBuilder, error)

func (*MetadataBuilder) SetDefaultSpecID added in v0.2.0 

func (b *MetadataBuilder) SetDefaultSpecID(defaultSpecID int) (*MetadataBuilder, error)

func (*MetadataBuilder) SetFormatVersion added in v0.2.0 

func (b *MetadataBuilder) SetFormatVersion(formatVersion int) (*MetadataBuilder, error)

func (*MetadataBuilder) SetLastUpdatedMS added in v0.2.0 

func (b *MetadataBuilder) SetLastUpdatedMS() *MetadataBuilder

func (*MetadataBuilder) SetLoc added in v0.2.0 

func (b *MetadataBuilder) SetLoc(loc string) (*MetadataBuilder, error)

func (*MetadataBuilder) SetProperties added in v0.2.0 

func (b *MetadataBuilder) SetProperties(props iceberg.Properties) (*MetadataBuilder, error)

func (*MetadataBuilder) SetSnapshotRef added in v0.2.0 

func (b *MetadataBuilder) SetSnapshotRef(
	name string,
	snapshotID int64,
	refType RefType,
	options ...setSnapshotRefOption,
) (*MetadataBuilder, error)

func (*MetadataBuilder) SetUUID added in v0.2.0 

func (b *MetadataBuilder) SetUUID(uuid uuid.UUID) (*MetadataBuilder, error)

func (*MetadataBuilder) SnapshotByID added in v0.2.0 

func (b *MetadataBuilder) SnapshotByID(id int64) (*Snapshot, error)

func (*MetadataBuilder) TrimMetadataLogs added in v0.2.0 

func (b *MetadataBuilder) TrimMetadataLogs(maxEntries int) *MetadataBuilder

type MetadataLogEntry 

type MetadataLogEntry struct {
	MetadataFile string `json:"metadata-file"`
	TimestampMs  int64  `json:"timestamp-ms"`
}

type NullOrder 

type NullOrder string
const (
	NullsFirst NullOrder = "nulls-first"
	NullsLast  NullOrder = "nulls-last"
)

type Operation 

type Operation string
const (
	OpAppend    Operation = "append"
	OpReplace   Operation = "replace"
	OpOverwrite Operation = "overwrite"
	OpDelete    Operation = "delete"
)

func ValidOperation 

func ValidOperation(s string) (Operation, error)

ValidOperation ensures that a given string is one of the valid operation types: append,replace,overwrite,delete

type RefType 

type RefType string

RefType will be either a BranchRef or a TagRef 

const (
	BranchRef RefType = "branch"
	TagRef    RefType = "tag"
)

type Requirement added in v0.2.0 

type Requirement interface {
	// Validate checks that the current table metadata satisfies the requirement.
	Validate(Metadata) error
	GetType() string
}

A Requirement is a validation rule that must be satisfied before attempting to make and commit changes to a table. Requirements are used to ensure that the table is in a valid state before making changes.

func AssertCreate added in v0.2.0 

func AssertCreate() Requirement

AssertCreate creates a requirement that the table does not already exist.

func AssertCurrentSchemaID added in v0.2.0 

func AssertCurrentSchemaID(id int) Requirement

AssertCurrentSchemaId creates a requirement that the table's current schema ID matches the given id.

func AssertDefaultSortOrderID added in v0.2.0 

func AssertDefaultSortOrderID(id int) Requirement

AssertDefaultSortOrderID creates a requirement that the table's default sort order ID matches the given id.

func AssertDefaultSpecID added in v0.2.0 

func AssertDefaultSpecID(id int) Requirement

AssertDefaultSpecID creates a requirement that the table's default partition spec ID matches the given id.

func AssertLastAssignedFieldID added in v0.2.0 

func AssertLastAssignedFieldID(id int) Requirement

AssertLastAssignedFieldID validates that the table's last assigned column ID matches the given id.

func AssertLastAssignedPartitionID added in v0.2.0 

func AssertLastAssignedPartitionID(id int) Requirement

AssertLastAssignedPartitionID creates a requriement that the table's last assigned partition ID matches the given id.

func AssertRefSnapshotID added in v0.2.0 

func AssertRefSnapshotID(ref string, id *int64) Requirement

AssertRefSnapshotID creates a requirement which ensures that the table branch or tag identified by the given ref must reference the given snapshot id. If the id is nil, the ref must not already exist.

func AssertTableUUID added in v0.2.0 

func AssertTableUUID(uuid uuid.UUID) Requirement

AssertTableUUID creates a requirement that the table UUID matches the given UUID.

type Scan 

type Scan struct {
	// contains filtered or unexported fields
}

func (*Scan) PlanFiles 

func (scan *Scan) PlanFiles(ctx context.Context) ([]FileScanTask, error)

PlanFiles orchestrates the fetching and filtering of manifests, and then building a list of FileScanTasks that match the current Scan criteria.

func (*Scan) Projection 

func (scan *Scan) Projection() (*iceberg.Schema, error)

func (*Scan) Snapshot 

func (scan *Scan) Snapshot() *Snapshot

func (*Scan) ToArrowRecords 

func (scan *Scan) ToArrowRecords(ctx context.Context) (*arrow.Schema, iter.Seq2[arrow.Record, error], error)

ToArrowRecords returns the arrow schema of the expected records and an interator that can be used with a range expression to read the records as they are available. If an error is encountered, during the planning and setup then this will return the error directly. If the error occurs while iterating the records, it will be returned by the iterator.

The purpose for returning the schema up front is to handle the case where there are no rows returned. The resulting Arrow Schema of the projection will still be known.

func (*Scan) ToArrowTable 

func (scan *Scan) ToArrowTable(ctx context.Context) (arrow.Table, error)

ToArrowTable calls ToArrowRecords and then gathers all of the records together and returns an arrow.Table make from those records.

func (*Scan) UseRef 

func (scan *Scan) UseRef(name string) (*Scan, error)

func (*Scan) UseRowLimit 

func (scan *Scan) UseRowLimit(n int64) *Scan

type ScanOption 

type ScanOption func(*Scan)

func WitMaxConcurrency 

func WitMaxConcurrency(n int) ScanOption

WitMaxConcurrency sets the maximum concurrency for table scan and plan operations. When unset it defaults to runtime.GOMAXPROCS.

func WithCaseSensitive 

func WithCaseSensitive(b bool) ScanOption

func WithLimit 

func WithLimit(n int64) ScanOption

func WithOptions 

func WithOptions(opts iceberg.Properties) ScanOption

func WithRowFilter 

func WithRowFilter(e iceberg.BooleanExpression) ScanOption

func WithSelectedFields 

func WithSelectedFields(fields ...string) ScanOption

func WithSnapshotID 

func WithSnapshotID(n int64) ScanOption

type Snapshot 

type Snapshot struct {
	SnapshotID       int64    `json:"snapshot-id"`
	ParentSnapshotID *int64   `json:"parent-snapshot-id,omitempty"`
	SequenceNumber   int64    `json:"sequence-number"`
	TimestampMs      int64    `json:"timestamp-ms"`
	ManifestList     string   `json:"manifest-list,omitempty"`
	Summary          *Summary `json:"summary,omitempty"`
	SchemaID         *int     `json:"schema-id,omitempty"`
}

func (Snapshot) Equals 

func (s Snapshot) Equals(other Snapshot) bool

func (Snapshot) Manifests 

func (s Snapshot) Manifests(fio iceio.IO) ([]iceberg.ManifestFile, error)

func (Snapshot) String 

func (s Snapshot) String() string

type SnapshotLogEntry 

type SnapshotLogEntry struct {
	SnapshotID  int64 `json:"snapshot-id"`
	TimestampMs int64 `json:"timestamp-ms"`
}

type SnapshotRef 

type SnapshotRef struct {
	SnapshotID         int64   `json:"snapshot-id"`
	SnapshotRefType    RefType `json:"type"`
	MinSnapshotsToKeep *int    `json:"min-snapshots-to-keep,omitempty"`
	MaxSnapshotAgeMs   *int64  `json:"max-snapshot-age-ms,omitempty"`
	MaxRefAgeMs        *int64  `json:"max-ref-age-ms,omitempty"`
}

SnapshotRef represents the reference information for a specific snapshot

func (*SnapshotRef) Equals 

func (s *SnapshotRef) Equals(rhs SnapshotRef) bool

func (*SnapshotRef) UnmarshalJSON 

func (s *SnapshotRef) UnmarshalJSON(b []byte) error

type SnapshotSummaryCollector added in v0.2.0 

type SnapshotSummaryCollector struct {
	// contains filtered or unexported fields
}

type SortDirection 

type SortDirection string
const (
	SortASC  SortDirection = "asc"
	SortDESC SortDirection = "desc"
)

type SortField 

type SortField struct {
	// SourceID is the source column id from the table's schema
	SourceID int `json:"source-id"`
	// Transform is the tranformation used to produce values to be
	// sorted on from the source column.
	Transform iceberg.Transform `json:"transform"`
	// Direction is an enum indicating ascending or descending direction.
	Direction SortDirection `json:"direction"`
	// NullOrder describes the order of null values when sorting
	// should be only either nulls-first or nulls-last enum values.
	NullOrder NullOrder `json:"null-order"`
}

SortField describes a field used in a sort order definition.

func (*SortField) MarshalJSON 

func (s *SortField) MarshalJSON() ([]byte, error)

func (*SortField) String 

func (s *SortField) String() string

func (*SortField) UnmarshalJSON 

func (s *SortField) UnmarshalJSON(b []byte) error

type SortOrder 

type SortOrder struct {
	OrderID int         `json:"order-id"`
	Fields  []SortField `json:"fields"`
}

SortOrder describes how the data is sorted within the table.

Data can be sorted within partitions by columns to gain performance. The order of the sort fields within the list defines the order in which the sort is applied to the data.

func AssignFreshSortOrderIDs added in v0.2.0 

func AssignFreshSortOrderIDs(sortOrder SortOrder, old, fresh *iceberg.Schema) (SortOrder, error)

AssignFreshSortOrderIDs updates and reassigns the field source IDs from the old schema to the corresponding fields in the fresh schema, while also giving the Sort Order a fresh ID of 0 (the initial Sort Order ID).

func AssignFreshSortOrderIDsWithID added in v0.2.0 

func AssignFreshSortOrderIDsWithID(sortOrder SortOrder, old, fresh *iceberg.Schema, sortOrderID int) (SortOrder, error)

AssignFreshSortOrderIDsWithID is like AssignFreshSortOrderIDs but allows specifying the id of the returned SortOrder.

func (SortOrder) Equals 

func (s SortOrder) Equals(rhs SortOrder) bool

func (SortOrder) String 

func (s SortOrder) String() string

func (*SortOrder) UnmarshalJSON 

func (s *SortOrder) UnmarshalJSON(b []byte) error

type StagedTable added in v0.2.0 

type StagedTable struct {
	*Table
}

func (*StagedTable) Refresh added in v0.2.0 

func (s *StagedTable) Refresh(ctx context.Context) (*Table, error)

func (*StagedTable) Scan added in v0.2.0 

func (s *StagedTable) Scan(opts ...ScanOption) *Scan

type Summary 

type Summary struct {
	Operation  Operation
	Properties iceberg.Properties
}

Summary stores the summary information for a snapshot indicating the operation that created the snapshot, and various properties which might exist in the summary.

func (*Summary) Equals 

func (s *Summary) Equals(other *Summary) bool

func (*Summary) MarshalJSON 

func (s *Summary) MarshalJSON() ([]byte, error)

func (*Summary) String 

func (s *Summary) String() string

func (*Summary) UnmarshalJSON 

func (s *Summary) UnmarshalJSON(b []byte) (err error)

type Table 

type Table struct {
	// contains filtered or unexported fields
}

func New 

func New(ident Identifier, meta Metadata, metadataLocation string, fs io.IO, cat CatalogIO) *Table

func NewFromLocation 

func NewFromLocation(ident Identifier, metalocation string, fsys io.IO, cat CatalogIO) (*Table, error)

func (Table) AllManifests added in v0.2.0 

func (t Table) AllManifests() iter.Seq2[iceberg.ManifestFile, error]

func (Table) CurrentSnapshot 

func (t Table) CurrentSnapshot() *Snapshot

func (Table) Equals 

func (t Table) Equals(other Table) bool

func (Table) FS 

func (t Table) FS() io.IO

func (Table) Identifier 

func (t Table) Identifier() Identifier

func (Table) Location 

func (t Table) Location() string

func (Table) LocationProvider added in v0.2.0 

func (t Table) LocationProvider() (LocationProvider, error)

func (Table) Metadata 

func (t Table) Metadata() Metadata

func (Table) MetadataLocation 

func (t Table) MetadataLocation() string

func (Table) NameMapping added in v0.2.0 

func (t Table) NameMapping() iceberg.NameMapping

func (Table) NewTransaction added in v0.2.0 

func (t Table) NewTransaction() *Transaction

func (Table) Properties 

func (t Table) Properties() iceberg.Properties

func (Table) Scan 

func (t Table) Scan(opts ...ScanOption) *Scan

func (Table) Schema 

func (t Table) Schema() *iceberg.Schema

func (Table) Schemas 

func (t Table) Schemas() map[int]*iceberg.Schema

func (Table) SnapshotByID 

func (t Table) SnapshotByID(id int64) *Snapshot

func (Table) SnapshotByName 

func (t Table) SnapshotByName(name string) *Snapshot

func (Table) SortOrder 

func (t Table) SortOrder() SortOrder

func (Table) Spec 

func (t Table) Spec() iceberg.PartitionSpec

type Transaction added in v0.2.0 

type Transaction struct {
	// contains filtered or unexported fields
}

func (*Transaction) AddFiles added in v0.2.0 

func (t *Transaction) AddFiles(files []string, snapshotProps iceberg.Properties, ignoreDuplicates bool) error

func (*Transaction) Append added in v0.2.0 

func (t *Transaction) Append(rdr array.RecordReader, snapshotProps iceberg.Properties) error

func (*Transaction) Commit added in v0.2.0 

func (t *Transaction) Commit(ctx context.Context) (*Table, error)

func (*Transaction) Scan added in v0.2.0 

func (t *Transaction) Scan(opts ...ScanOption) (*Scan, error)

func (*Transaction) SetProperties added in v0.2.0 

func (t *Transaction) SetProperties(props iceberg.Properties) error

func (*Transaction) StagedTable added in v0.2.0 

func (t *Transaction) StagedTable() (*StagedTable, error)

type Update added in v0.2.0 

type Update interface {
	// Action returns the name of the action that the update represents.
	Action() string
	// Apply applies the update to the given metadata builder.
	Apply(*MetadataBuilder) error
}

Update represents a change to a table's metadata.

func NewAddPartitionSpecUpdate added in v0.2.0 

func NewAddPartitionSpecUpdate(spec *iceberg.PartitionSpec, initial bool) Update

NewAddPartitionSpecUpdate creates a new update that adds the given partition spec to the table metadata. If the initial flag is set to true, the spec is considered the initial spec of the table, and all other previously added specs in the metadata builder are removed.

func NewAddSchemaUpdate added in v0.2.0 

func NewAddSchemaUpdate(schema *iceberg.Schema, lastColumnID int, initial bool) Update

NewAddSchemaUpdate creates a new update that adds the given schema and last column ID to the table metadata. If the initial flag is set to true, the schema is considered the initial schema of the table, and all previously added schemas in the metadata builder are removed.

func NewAddSnapshotUpdate added in v0.2.0 

func NewAddSnapshotUpdate(snapshot *Snapshot) Update

NewAddSnapshotUpdate creates a new update that adds the given snapshot to the table metadata.

func NewAddSortOrderUpdate added in v0.2.0 

func NewAddSortOrderUpdate(sortOrder *SortOrder, initial bool) Update

NewAddSortOrderUpdate creates a new update that adds the given sort order to the table metadata. If the initial flag is set to true, the sort order is considered the initial sort order of the table, and all previously added sort orders in the metadata builder are removed.

func NewAssignUUIDUpdate added in v0.2.0 

func NewAssignUUIDUpdate(uuid uuid.UUID) Update

NewAssignUUIDUpdate creates a new update to assign a UUID to the table metadata.

func NewRemovePropertiesUpdate added in v0.2.0 

func NewRemovePropertiesUpdate(removals []string) Update

NewRemovePropertiesUpdate creates a new update that removes properties from the table metadata. The properties are identified by their names, and if a property with the given name does not exist, it is ignored.

func NewRemoveSnapshotsUpdate added in v0.2.0 

func NewRemoveSnapshotsUpdate(ids []int64) Update

NewRemoveSnapshotsUpdate creates a new update that removes all snapshots from the table metadata with the given snapshot IDs.

func NewSetCurrentSchemaUpdate added in v0.2.0 

func NewSetCurrentSchemaUpdate(id int) Update

NewSetCurrentSchemaUpdate creates a new update that sets the current schema of the table metadata to the given schema ID.

func NewSetDefaultSortOrderUpdate added in v0.2.0 

func NewSetDefaultSortOrderUpdate(id int) Update

NewSetDefaultSortOrderUpdate creates a new update that sets the default sort order of the table metadata to the given sort order ID.

func NewSetDefaultSpecUpdate added in v0.2.0 

func NewSetDefaultSpecUpdate(id int) Update

NewSetDefaultSpecUpdate creates a new update that sets the default partition spec of the table metadata to the given spec ID.

func NewSetLocationUpdate added in v0.2.0 

func NewSetLocationUpdate(loc string) Update

NewSetLocationUpdate creates a new update that sets the location of the table metadata.

func NewSetSnapshotRefUpdate added in v0.2.0 

func NewSetSnapshotRefUpdate(
	name string,
	snapshotID int64,
	refType RefType,
	maxRefAgeMs, maxSnapshotAgeMs int64,
	minSnapshotsToKeep int,
) Update

NewSetSnapshotRefUpdate creates a new update that sets the given snapshot reference as the current snapshot of the table metadata. MaxRefAgeMs, MaxSnapshotAgeMs, and MinSnapshotsToKeep are optional, and any non-positive values are ignored.

func NewUpgradeFormatVersionUpdate added in v0.2.0 

func NewUpgradeFormatVersionUpdate(formatVersion int) Update

NewUpgradeFormatVersionUpdate creates a new update that upgrades the format version of the table metadata to the given formatVersion.

Source Files

View all Source files

Directories

Path Synopsis

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.