table

package
v0.1.0-rc2 Latest Latest
Warning

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

 Go to latest Published: Nov 13, 2024 License: Apache-2.0 Imports: 29 Imported by: 1

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 (
	InitialSortOrderID  = 1
	UnsortedSortOrderID = 0
)
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 ApplyNameMapping 

func ApplyNameMapping(schemaWithoutIDs *iceberg.Schema, nameMapping NameMapping) (*iceberg.Schema, error)

func ArrowSchemaToIceberg 

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

func ArrowTypeToIceberg 

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

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(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) (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 VisitMappedFields 

func VisitMappedFields[S, T any](fields []MappedField, visitor NameMappingVisitor[S, T]) (res S, err error)

func VisitNameMapping 

func VisitNameMapping[S, T any](obj NameMapping, visitor NameMappingVisitor[S, T]) (res S, err error)

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 FileScanTask 

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

type Identifier 

type Identifier = []string

type MappedField 

type MappedField struct {
	Names []string `json:"names"`
	// iceberg spec says this is optional, but I don't see any examples
	// of this being left empty. Does pyiceberg need to be updated or should
	// the spec not say field-id is optional?
	FieldID *int          `json:"field-id,omitempty"`
	Fields  []MappedField `json:"fields,omitempty"`
}

func (*MappedField) Len 

func (m *MappedField) Len() int

func (*MappedField) String 

func (m *MappedField) String() string

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
	// 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
	// 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

	Equals(Metadata) bool
}

Metadata for an iceberg table as specified in the Iceberg spec

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

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 MetadataLogEntry 

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

type MetadataV1 

type MetadataV1 struct {
	Schema    iceberg.Schema           `json:"schema"`
	Partition []iceberg.PartitionField `json:"partition-spec"`
	// contains filtered or unexported fields
}

func (*MetadataV1) CurrentSchema 

func (c *MetadataV1) CurrentSchema() *iceberg.Schema

func (*MetadataV1) CurrentSnapshot 

func (c *MetadataV1) CurrentSnapshot() *Snapshot

func (*MetadataV1) DefaultPartitionSpec 

func (c *MetadataV1) DefaultPartitionSpec() int

func (*MetadataV1) Equals 

func (m *MetadataV1) Equals(other Metadata) bool

func (*MetadataV1) LastColumnID 

func (c *MetadataV1) LastColumnID() int

func (*MetadataV1) LastPartitionSpecID 

func (c *MetadataV1) LastPartitionSpecID() *int

func (*MetadataV1) LastUpdatedMillis 

func (c *MetadataV1) LastUpdatedMillis() int64

func (*MetadataV1) Location 

func (c *MetadataV1) Location() string

func (*MetadataV1) PartitionSpec 

func (c *MetadataV1) PartitionSpec() iceberg.PartitionSpec

func (*MetadataV1) PartitionSpecs 

func (c *MetadataV1) PartitionSpecs() []iceberg.PartitionSpec

func (*MetadataV1) Properties 

func (c *MetadataV1) Properties() iceberg.Properties

func (*MetadataV1) Schemas 

func (c *MetadataV1) Schemas() []*iceberg.Schema

func (*MetadataV1) SnapshotByID 

func (c *MetadataV1) SnapshotByID(id int64) *Snapshot

func (*MetadataV1) SnapshotByName 

func (c *MetadataV1) SnapshotByName(name string) *Snapshot

func (*MetadataV1) Snapshots 

func (c *MetadataV1) Snapshots() []Snapshot

func (*MetadataV1) SortOrder 

func (c *MetadataV1) SortOrder() SortOrder

func (*MetadataV1) SortOrders 

func (c *MetadataV1) SortOrders() []SortOrder

func (*MetadataV1) TableUUID 

func (c *MetadataV1) TableUUID() uuid.UUID

func (*MetadataV1) ToV2 

func (m *MetadataV1) ToV2() MetadataV2

func (*MetadataV1) UnmarshalJSON 

func (m *MetadataV1) UnmarshalJSON(b []byte) error

func (*MetadataV1) Version 

func (c *MetadataV1) Version() int

type MetadataV2 

type MetadataV2 struct {
	LastSequenceNumber int `json:"last-sequence-number"`
	// contains filtered or unexported fields
}

func (*MetadataV2) CurrentSchema 

func (c *MetadataV2) CurrentSchema() *iceberg.Schema

func (*MetadataV2) CurrentSnapshot 

func (c *MetadataV2) CurrentSnapshot() *Snapshot

func (*MetadataV2) DefaultPartitionSpec 

func (c *MetadataV2) DefaultPartitionSpec() int

func (*MetadataV2) Equals 

func (m *MetadataV2) Equals(other Metadata) bool

func (*MetadataV2) LastColumnID 

func (c *MetadataV2) LastColumnID() int

func (*MetadataV2) LastPartitionSpecID 

func (c *MetadataV2) LastPartitionSpecID() *int

func (*MetadataV2) LastUpdatedMillis 

func (c *MetadataV2) LastUpdatedMillis() int64

func (*MetadataV2) Location 

func (c *MetadataV2) Location() string

func (*MetadataV2) PartitionSpec 

func (c *MetadataV2) PartitionSpec() iceberg.PartitionSpec

func (*MetadataV2) PartitionSpecs 

func (c *MetadataV2) PartitionSpecs() []iceberg.PartitionSpec

func (*MetadataV2) Properties 

func (c *MetadataV2) Properties() iceberg.Properties

func (*MetadataV2) Schemas 

func (c *MetadataV2) Schemas() []*iceberg.Schema

func (*MetadataV2) SnapshotByID 

func (c *MetadataV2) SnapshotByID(id int64) *Snapshot

func (*MetadataV2) SnapshotByName 

func (c *MetadataV2) SnapshotByName(name string) *Snapshot

func (*MetadataV2) Snapshots 

func (c *MetadataV2) Snapshots() []Snapshot

func (*MetadataV2) SortOrder 

func (c *MetadataV2) SortOrder() SortOrder

func (*MetadataV2) SortOrders 

func (c *MetadataV2) SortOrders() []SortOrder

func (*MetadataV2) TableUUID 

func (c *MetadataV2) TableUUID() uuid.UUID

func (*MetadataV2) UnmarshalJSON 

func (m *MetadataV2) UnmarshalJSON(b []byte) error

func (*MetadataV2) Version 

func (c *MetadataV2) Version() int

type NameMapping 

type NameMapping []MappedField

func (NameMapping) String 

func (nm NameMapping) String() string

type NameMappingAccessor 

type NameMappingAccessor struct{}

func (NameMappingAccessor) FieldPartner 

func (n NameMappingAccessor) FieldPartner(partnerStruct *MappedField, _ int, fieldName string) *MappedField

func (NameMappingAccessor) ListElementPartner 

func (n NameMappingAccessor) ListElementPartner(partnerList *MappedField) *MappedField

func (NameMappingAccessor) MapKeyPartner 

func (n NameMappingAccessor) MapKeyPartner(partnerMap *MappedField) *MappedField

func (NameMappingAccessor) MapValuePartner 

func (n NameMappingAccessor) MapValuePartner(partnerMap *MappedField) *MappedField

func (NameMappingAccessor) SchemaPartner 

func (NameMappingAccessor) SchemaPartner(partner *MappedField) *MappedField

type NameMappingVisitor 

type NameMappingVisitor[S, T any] interface {
	Mapping(nm NameMapping, fieldResults S) S
	Fields(st []MappedField, fieldResults []T) S
	Field(field MappedField, fieldResult S) T
}

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 Scan 

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

func (*Scan) PlanFiles 

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

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 io.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 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 (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 Summary 

type Summary struct {
	Operation  Operation
	Properties map[string]string
}

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, location string, fs io.IO) *Table

func NewFromLocation 

func NewFromLocation(ident Identifier, metalocation string, fsys io.IO) (*Table, 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) Metadata 

func (t Table) Metadata() Metadata

func (Table) MetadataLocation 

func (t Table) MetadataLocation() string

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

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.