priorityqueue

package
v1.1.0-beta.0...-bd07e09 Latest Latest
Warning

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

Go to latest
Published: Dec 16, 2024 License: Apache-2.0 Imports: 27 Imported by: 0

Documentation

Index

Constants

View Source
const (
	// EventNone represents no special event.
	EventNone = 0.0
	// EventNewIndex represents a special event for newly added indexes.
	EventNewIndex = 2.0
)
View Source
const NoRecord = -1

NoRecord is used to indicate that there is no related record in mysql.analyze_jobs.

Variables

View Source
var (
	// ErrHeapIsEmpty is returned when the heap is empty.
	ErrHeapIsEmpty = errors.New("heap is empty")
)

Functions

func GetAverageAnalysisDuration

func GetAverageAnalysisDuration(
	sctx sessionctx.Context,
	schema, tableName string,
	partitionNames ...string,
) (time.Duration, error)

GetAverageAnalysisDuration returns the average duration of the last 5 successful analyses for each specified partition. If there are no successful analyses, it returns 0.

func GetLastFailedAnalysisDuration

func GetLastFailedAnalysisDuration(
	sctx sessionctx.Context,
	schema, tableName string,
	partitionNames ...string,
) (time.Duration, error)

GetLastFailedAnalysisDuration returns the duration since the last failed analysis. If there is no failed analysis, it returns 0.

func GetPartitionStats

func GetPartitionStats(
	statsHandle statstypes.StatsHandle,
	tblInfo *model.TableInfo,
	defs []model.PartitionDefinition,
) map[PartitionIDAndName]*statistics.Table

GetPartitionStats gets the partition stats.

func IsDynamicPartitionedTableAnalysisJob

func IsDynamicPartitionedTableAnalysisJob(job AnalysisJob) bool

IsDynamicPartitionedTableAnalysisJob checks whether the job is a dynamic partitioned table analysis job.

Types

type AnalysisJob

type AnalysisJob interface {
	// ValidateAndPrepare validates if the analysis job can run and prepares it for execution.
	// Returns (true, "") if the job is valid and ready to run.
	// Returns (false, reason) if the job should be skipped, where reason explains why:
	// - Schema/table/partition doesn't exist
	// - Table is not partitioned (for partition jobs)
	// - Recent failed analysis (within 2x avg duration) to avoid queue blocking
	ValidateAndPrepare(
		sctx sessionctx.Context,
	) (bool, string)

	// Analyze executes the analyze statement within a transaction.
	Analyze(
		statsHandle statstypes.StatsHandle,
		sysProcTracker sysproctrack.Tracker,
	) error

	// SetWeight sets the weight of the job.
	SetWeight(weight float64)

	// GetWeight gets the weight of the job.
	GetWeight() float64

	// HasNewlyAddedIndex checks whether the job has newly added index.
	HasNewlyAddedIndex() bool

	// GetIndicators gets the indicators of the job.
	GetIndicators() Indicators

	// SetIndicators sets the indicators of the job.
	SetIndicators(indicators Indicators)

	// GetTableID gets the table ID of the job.
	GetTableID() int64

	// RegisterSuccessHook registers a successHook function that will be called after the job can be marked as successful.
	RegisterSuccessHook(hook SuccessJobHook)

	// RegisterFailureHook registers a failureHook function that will be called after the job is marked as failed.
	RegisterFailureHook(hook FailureJobHook)

	// AsJSON converts the job to a JSON object.
	AsJSON() statstypes.AnalysisJobJSON

	fmt.Stringer
}

AnalysisJob is the interface for the analysis job.

type AnalysisJobFactory

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

AnalysisJobFactory is responsible for creating different types of analysis jobs. NOTE: This struct is not thread-safe.

func NewAnalysisJobFactory

func NewAnalysisJobFactory(sctx sessionctx.Context, autoAnalyzeRatio float64, currentTs uint64) *AnalysisJobFactory

NewAnalysisJobFactory creates a new AnalysisJobFactory.

func (*AnalysisJobFactory) CalculateChangePercentage

func (f *AnalysisJobFactory) CalculateChangePercentage(tblStats *statistics.Table) float64

CalculateChangePercentage calculates the change percentage of the table based on the change count and the analysis count.

func (*AnalysisJobFactory) CalculateIndicatorsForPartitions

func (f *AnalysisJobFactory) CalculateIndicatorsForPartitions(
	globalStats *statistics.Table,
	partitionStats map[PartitionIDAndName]*statistics.Table,
) (
	avgChange float64,
	avgSize float64,
	avgLastAnalyzeDuration time.Duration,
	partitionIDs map[int64]struct{},
)

CalculateIndicatorsForPartitions calculates the average change percentage, average size and average last analyze duration for the partitions that meet the threshold. Change percentage is the ratio of the number of modified rows to the total number of rows. Size is the product of the number of rows and the number of columns. Last analyze duration is the duration since the last analyze.

func (*AnalysisJobFactory) CalculateTableSize

func (*AnalysisJobFactory) CalculateTableSize(tblStats *statistics.Table) float64

CalculateTableSize calculates the size of the table.

func (*AnalysisJobFactory) CheckIndexesNeedAnalyze

func (*AnalysisJobFactory) CheckIndexesNeedAnalyze(tblInfo *model.TableInfo, tblStats *statistics.Table) map[int64]struct{}

CheckIndexesNeedAnalyze checks if the indexes need to be analyzed.

func (*AnalysisJobFactory) CheckNewlyAddedIndexesNeedAnalyzeForPartitionedTable

func (*AnalysisJobFactory) CheckNewlyAddedIndexesNeedAnalyzeForPartitionedTable(
	tblInfo *model.TableInfo,
	partitionStats map[PartitionIDAndName]*statistics.Table,
) map[int64][]int64

CheckNewlyAddedIndexesNeedAnalyzeForPartitionedTable checks if the indexes of the partitioned table need to be analyzed. It returns a map from index name to the names of the partitions that need to be analyzed. NOTE: This is only for newly added indexes.

func (*AnalysisJobFactory) CreateDynamicPartitionedTableAnalysisJob

func (f *AnalysisJobFactory) CreateDynamicPartitionedTableAnalysisJob(
	globalTblInfo *model.TableInfo,
	globalTblStats *statistics.Table,
	partitionStats map[PartitionIDAndName]*statistics.Table,
) AnalysisJob

CreateDynamicPartitionedTableAnalysisJob creates a job for dynamic partitioned tables.

func (*AnalysisJobFactory) CreateNonPartitionedTableAnalysisJob

func (f *AnalysisJobFactory) CreateNonPartitionedTableAnalysisJob(
	tblInfo *model.TableInfo,
	tblStats *statistics.Table,
) AnalysisJob

CreateNonPartitionedTableAnalysisJob creates a job for non-partitioned tables.

func (*AnalysisJobFactory) CreateStaticPartitionAnalysisJob

func (f *AnalysisJobFactory) CreateStaticPartitionAnalysisJob(
	globalTblInfo *model.TableInfo,
	partitionID int64,
	partitionStats *statistics.Table,
) AnalysisJob

CreateStaticPartitionAnalysisJob creates a job for static partitions.

func (*AnalysisJobFactory) FindLastAnalyzeTime

func (f *AnalysisJobFactory) FindLastAnalyzeTime(tblStats *statistics.Table) time.Time

FindLastAnalyzeTime finds the last analyze time of the table. It uses `LastUpdateVersion` to find the last analyze time. The `LastUpdateVersion` is the version of the transaction that updates the statistics. It always not null(default 0), so we can use it to find the last analyze time.

func (*AnalysisJobFactory) GetTableLastAnalyzeDuration

func (f *AnalysisJobFactory) GetTableLastAnalyzeDuration(tblStats *statistics.Table) time.Duration

GetTableLastAnalyzeDuration gets the last analyze duration of the table.

type AnalysisPriorityQueue

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

AnalysisPriorityQueue is a priority queue for TableAnalysisJobs. Testing shows that keeping all jobs in memory is feasible. Memory usage for one million tables is approximately 300 to 500 MiB, which is acceptable. Typically, not many tables need to be analyzed simultaneously.

func NewAnalysisPriorityQueue

func NewAnalysisPriorityQueue(handle statstypes.StatsHandle) *AnalysisPriorityQueue

NewAnalysisPriorityQueue creates a new AnalysisPriorityQueue2.

func (*AnalysisPriorityQueue) Close

func (pq *AnalysisPriorityQueue) Close()

Close closes the priority queue. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) GetLastFetchTimestamp

func (pq *AnalysisPriorityQueue) GetLastFetchTimestamp() uint64

GetLastFetchTimestamp returns the last fetch timestamp of DML updates. Note: This function is thread-safe. Exported for testing.

func (*AnalysisPriorityQueue) GetRunningJobs

func (pq *AnalysisPriorityQueue) GetRunningJobs() map[int64]struct{}

GetRunningJobs returns the running jobs. Note: This function is thread-safe. Exported for testing.

func (*AnalysisPriorityQueue) HandleDDLEvent

func (pq *AnalysisPriorityQueue) HandleDDLEvent(_ context.Context, sctx sessionctx.Context, event *notifier.SchemaChangeEvent) (err error)

HandleDDLEvent handles DDL events for the priority queue.

func (*AnalysisPriorityQueue) Initialize

func (pq *AnalysisPriorityQueue) Initialize() error

Initialize initializes the priority queue. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) IsEmpty

func (pq *AnalysisPriorityQueue) IsEmpty() (bool, error)

IsEmpty checks whether the priority queue is empty. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) IsInitialized

func (pq *AnalysisPriorityQueue) IsInitialized() bool

IsInitialized checks if the priority queue is initialized.

func (*AnalysisPriorityQueue) Len

func (pq *AnalysisPriorityQueue) Len() (int, error)

Len returns the number of jobs in the priority queue. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) Peek

func (pq *AnalysisPriorityQueue) Peek() (AnalysisJob, error)

Peek peeks the top job from the priority queue.

func (*AnalysisPriorityQueue) Pop

Pop pops a job from the priority queue and marks it as running. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) ProcessDMLChanges

func (pq *AnalysisPriorityQueue) ProcessDMLChanges()

ProcessDMLChanges processes DML changes. Note: This function is thread-safe. Performance: To scan all table stats and process the DML changes, it takes about less than 100ms for 1m tables.

func (*AnalysisPriorityQueue) Push

func (pq *AnalysisPriorityQueue) Push(job AnalysisJob) error

Push pushes a job into the priority queue. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) Rebuild

func (pq *AnalysisPriorityQueue) Rebuild() error

Rebuild rebuilds the priority queue. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) RefreshLastAnalysisDuration

func (pq *AnalysisPriorityQueue) RefreshLastAnalysisDuration()

RefreshLastAnalysisDuration refreshes the last analysis duration of all jobs in the priority queue. Note: This function is thread-safe.

func (*AnalysisPriorityQueue) RequeueMustRetryJobs

func (pq *AnalysisPriorityQueue) RequeueMustRetryJobs()

RequeueMustRetryJobs requeues the must retry jobs.

func (*AnalysisPriorityQueue) Snapshot

func (pq *AnalysisPriorityQueue) Snapshot() (
	snapshot statstypes.PriorityQueueSnapshot,
	err error,
)

Snapshot returns a snapshot of all the jobs in the priority queue.

type AutoAnalysisTimeWindow

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

AutoAnalysisTimeWindow is a struct that contains the start and end time of the auto analyze time window.

func NewAutoAnalysisTimeWindow

func NewAutoAnalysisTimeWindow(start, end time.Time) AutoAnalysisTimeWindow

NewAutoAnalysisTimeWindow creates a new AutoAnalysisTimeWindow.

func (AutoAnalysisTimeWindow) IsWithinTimeWindow

func (a AutoAnalysisTimeWindow) IsWithinTimeWindow(currentTime time.Time) bool

IsWithinTimeWindow checks if the current time is within the time window. If the auto analyze time window is not set or the current time is not in the window, return false.

type DynamicPartitionedTableAnalysisJob

type DynamicPartitionedTableAnalysisJob struct {
	GlobalTableID     int64
	PartitionIDs      map[int64]struct{}
	PartitionIndexIDs map[int64][]int64

	// Some indicators to help us decide whether we need to analyze this table.
	Indicators
	// Analyze table with this version of statistics.
	TableStatsVer int
	// Weight is used to calculate the priority of the job.
	Weight float64

	// Lazy initialized.
	SchemaName      string
	GlobalTableName string
	// This will analyze all indexes and columns of the specified partitions.
	PartitionNames []string
	// Only set when partitions's indexes need to be analyzed.
	// It looks like: {"indexName": ["partitionName1", "partitionName2"]}
	// This is only for newly added indexes.
	// The reason why we need to record the partition names is that we need to analyze partitions in batch mode
	// and we don't want to analyze the same partition multiple times.
	// For example, the user may analyze some partitions manually, and we don't want to analyze them again.
	PartitionIndexNames map[string][]string
	// contains filtered or unexported fields
}

DynamicPartitionedTableAnalysisJob is a TableAnalysisJob for analyzing dynamic pruned partitioned table.

func NewDynamicPartitionedTableAnalysisJob

func NewDynamicPartitionedTableAnalysisJob(
	tableID int64,
	partitionIDs map[int64]struct{},
	partitionIndexIDs map[int64][]int64,
	tableStatsVer int,
	changePercentage float64,
	tableSize float64,
	lastAnalysisDuration time.Duration,
) *DynamicPartitionedTableAnalysisJob

NewDynamicPartitionedTableAnalysisJob creates a new job for analyzing a dynamic partitioned table's partitions.

func (*DynamicPartitionedTableAnalysisJob) Analyze

func (j *DynamicPartitionedTableAnalysisJob) Analyze(
	statsHandle statstypes.StatsHandle,
	sysProcTracker sysproctrack.Tracker,
) error

Analyze analyzes the partitions or partition indexes.

func (*DynamicPartitionedTableAnalysisJob) AsJSON

AsJSON converts the job to a JSON object.

func (*DynamicPartitionedTableAnalysisJob) GetIndicators

GetIndicators returns the indicators of the table.

func (*DynamicPartitionedTableAnalysisJob) GetTableID

func (j *DynamicPartitionedTableAnalysisJob) GetTableID() int64

GetTableID gets the table ID of the job.

func (*DynamicPartitionedTableAnalysisJob) GetWeight

GetWeight gets the weight of the job.

func (*DynamicPartitionedTableAnalysisJob) HasNewlyAddedIndex

func (j *DynamicPartitionedTableAnalysisJob) HasNewlyAddedIndex() bool

HasNewlyAddedIndex checks whether the job has newly added index.

func (*DynamicPartitionedTableAnalysisJob) RegisterFailureHook

func (j *DynamicPartitionedTableAnalysisJob) RegisterFailureHook(hook FailureJobHook)

RegisterFailureHook registers a failureHook function that will be called after the job can be marked as failed.

func (*DynamicPartitionedTableAnalysisJob) RegisterSuccessHook

func (j *DynamicPartitionedTableAnalysisJob) RegisterSuccessHook(hook SuccessJobHook)

RegisterSuccessHook registers a successHook function that will be called after the job can be marked as successful.

func (*DynamicPartitionedTableAnalysisJob) SetIndicators

func (j *DynamicPartitionedTableAnalysisJob) SetIndicators(indicators Indicators)

SetIndicators sets the indicators of the table.

func (*DynamicPartitionedTableAnalysisJob) SetWeight

func (j *DynamicPartitionedTableAnalysisJob) SetWeight(weight float64)

SetWeight sets the weight of the job.

func (*DynamicPartitionedTableAnalysisJob) String

String implements fmt.Stringer interface.

func (*DynamicPartitionedTableAnalysisJob) ValidateAndPrepare

func (j *DynamicPartitionedTableAnalysisJob) ValidateAndPrepare(
	sctx sessionctx.Context,
) (bool, string)

ValidateAndPrepare validates if the analysis job can run and prepares it for execution. For dynamic partitioned tables, it checks: - Schema exists - Table exists and is partitioned - All specified partitions exist - No recent failed analysis for any partition to avoid queue blocking

type FailureJobHook

type FailureJobHook func(job AnalysisJob, mustRetry bool)

FailureJobHook is the failureHook function that will be called after the job is failed.

type Indicators

type Indicators struct {
	// ChangePercentage is the percentage of the changed rows.
	// Usually, the more the changed rows, the higher the priority.
	// It is calculated by modifiedCount / last time analysis count.
	ChangePercentage float64
	// TableSize is the table size in rows * len(columns).
	TableSize float64
	// LastAnalysisDuration is the duration from the last analysis to now.
	LastAnalysisDuration time.Duration
}

Indicators contains some indicators to evaluate the table priority.

type NonPartitionedTableAnalysisJob

type NonPartitionedTableAnalysisJob struct {
	TableID  int64
	IndexIDs map[int64]struct{}

	Indicators
	TableStatsVer int
	Weight        float64

	// Lazy initialized.
	SchemaName string
	TableName  string
	IndexNames []string
	// contains filtered or unexported fields
}

NonPartitionedTableAnalysisJob is a TableAnalysisJob for analyzing the physical table.

func NewNonPartitionedTableAnalysisJob

func NewNonPartitionedTableAnalysisJob(
	tableID int64,
	indexIDs map[int64]struct{},
	tableStatsVer int,
	changePercentage float64,
	tableSize float64,
	lastAnalysisDuration time.Duration,
) *NonPartitionedTableAnalysisJob

NewNonPartitionedTableAnalysisJob creates a new TableAnalysisJob for analyzing the physical table.

func (*NonPartitionedTableAnalysisJob) Analyze

func (j *NonPartitionedTableAnalysisJob) Analyze(
	statsHandle statstypes.StatsHandle,
	sysProcTracker sysproctrack.Tracker,
) error

Analyze analyzes the table or indexes.

func (*NonPartitionedTableAnalysisJob) AsJSON

AsJSON converts the job to a JSON object.

func (*NonPartitionedTableAnalysisJob) GenSQLForAnalyzeIndex

func (j *NonPartitionedTableAnalysisJob) GenSQLForAnalyzeIndex(index string) (string, []any)

GenSQLForAnalyzeIndex generates the SQL for analyzing the specified index.

func (*NonPartitionedTableAnalysisJob) GenSQLForAnalyzeTable

func (j *NonPartitionedTableAnalysisJob) GenSQLForAnalyzeTable() (string, []any)

GenSQLForAnalyzeTable generates the SQL for analyzing the specified table.

func (*NonPartitionedTableAnalysisJob) GetIndicators

func (j *NonPartitionedTableAnalysisJob) GetIndicators() Indicators

GetIndicators returns the indicators of the table.

func (*NonPartitionedTableAnalysisJob) GetTableID

func (j *NonPartitionedTableAnalysisJob) GetTableID() int64

GetTableID gets the table ID of the job.

func (*NonPartitionedTableAnalysisJob) GetWeight

func (j *NonPartitionedTableAnalysisJob) GetWeight() float64

GetWeight gets the weight of the job.

func (*NonPartitionedTableAnalysisJob) HasNewlyAddedIndex

func (j *NonPartitionedTableAnalysisJob) HasNewlyAddedIndex() bool

HasNewlyAddedIndex checks whether the table has newly added indexes.

func (*NonPartitionedTableAnalysisJob) RegisterFailureHook

func (j *NonPartitionedTableAnalysisJob) RegisterFailureHook(hook FailureJobHook)

RegisterFailureHook registers a failureHook function that will be called after the job can be marked as failed.

func (*NonPartitionedTableAnalysisJob) RegisterSuccessHook

func (j *NonPartitionedTableAnalysisJob) RegisterSuccessHook(hook SuccessJobHook)

RegisterSuccessHook registers a successHook function that will be called after the job can be marked as successful.

func (*NonPartitionedTableAnalysisJob) SetIndicators

func (j *NonPartitionedTableAnalysisJob) SetIndicators(indicators Indicators)

SetIndicators sets the indicators of the table.

func (*NonPartitionedTableAnalysisJob) SetWeight

func (j *NonPartitionedTableAnalysisJob) SetWeight(weight float64)

SetWeight sets the weight of the job.

func (*NonPartitionedTableAnalysisJob) String

String implements fmt.Stringer interface.

func (*NonPartitionedTableAnalysisJob) ValidateAndPrepare

func (j *NonPartitionedTableAnalysisJob) ValidateAndPrepare(
	sctx sessionctx.Context,
) (bool, string)

ValidateAndPrepare validates if the analysis job can run and prepares it for execution. For non-partitioned tables, it checks: - Schema exists - Table exists - No recent failed analysis to avoid queue blocking

type PartitionIDAndName

type PartitionIDAndName struct {
	Name string
	ID   int64
}

PartitionIDAndName is a struct that contains the ID and name of a partition. Exported for testing purposes. Do not use it in other packages.

func NewPartitionIDAndName

func NewPartitionIDAndName(name string, id int64) PartitionIDAndName

NewPartitionIDAndName creates a new PartitionIDAndName.

type PriorityCalculator

type PriorityCalculator struct{}

PriorityCalculator implements the WeightCalculator interface.

func NewPriorityCalculator

func NewPriorityCalculator() *PriorityCalculator

NewPriorityCalculator creates a new PriorityCalculator.

For more information, please visit: https://github.com/pingcap/tidb/blob/master/docs/design/2023-11-29-priority-queue-for-auto-analyze.md

func (*PriorityCalculator) CalculateWeight

func (pc *PriorityCalculator) CalculateWeight(job AnalysisJob) float64

CalculateWeight calculates the weight based on the given rules. - Table Change Ratio (Change Ratio): Accounts for 60% - Table Size (Size): Accounts for 10% - Analysis Interval (Analysis Interval): Accounts for 30% priority_score calculates the priority score based on the following formula:

priority_score = (0.6 * math.Log10(1 + ChangeRatio) +
                  0.1 * (1 - math.Log10(1 + TableSize)) +
                  0.3 * math.Log10(1 + math.Sqrt(AnalysisInterval)) +
                  special_event[event])

func (*PriorityCalculator) GetSpecialEvent

func (*PriorityCalculator) GetSpecialEvent(job AnalysisJob) float64

GetSpecialEvent returns the special event weight. Exported for testing purposes.

type StaticPartitionedTableAnalysisJob

type StaticPartitionedTableAnalysisJob struct {
	GlobalTableID     int64
	StaticPartitionID int64
	IndexIDs          map[int64]struct{}

	Indicators
	TableStatsVer int
	Weight        float64

	// Lazy initialized.
	SchemaName          string
	GlobalTableName     string
	StaticPartitionName string
	IndexNames          []string
	// contains filtered or unexported fields
}

StaticPartitionedTableAnalysisJob is a job for analyzing a static partitioned table.

func NewStaticPartitionTableAnalysisJob

func NewStaticPartitionTableAnalysisJob(
	globalTableID int64,
	partitionID int64,
	indexIDs map[int64]struct{},
	tableStatsVer int,
	changePercentage float64,
	tableSize float64,
	lastAnalysisDuration time.Duration,
) *StaticPartitionedTableAnalysisJob

NewStaticPartitionTableAnalysisJob creates a job for analyzing a static partitioned table.

func (*StaticPartitionedTableAnalysisJob) Analyze

func (j *StaticPartitionedTableAnalysisJob) Analyze(
	statsHandle statstypes.StatsHandle,
	sysProcTracker sysproctrack.Tracker,
) error

Analyze analyzes the specified static partition or indexes.

func (*StaticPartitionedTableAnalysisJob) AsJSON

AsJSON converts the job to a JSON object.

func (*StaticPartitionedTableAnalysisJob) GenSQLForAnalyzeStaticPartition

func (j *StaticPartitionedTableAnalysisJob) GenSQLForAnalyzeStaticPartition() (string, []any)

GenSQLForAnalyzeStaticPartition generates the SQL for analyzing the specified static partition.

func (*StaticPartitionedTableAnalysisJob) GenSQLForAnalyzeStaticPartitionIndex

func (j *StaticPartitionedTableAnalysisJob) GenSQLForAnalyzeStaticPartitionIndex(index string) (string, []any)

GenSQLForAnalyzeStaticPartitionIndex generates the SQL for analyzing the specified static partition index.

func (*StaticPartitionedTableAnalysisJob) GetIndicators

func (j *StaticPartitionedTableAnalysisJob) GetIndicators() Indicators

GetIndicators implements AnalysisJob.

func (*StaticPartitionedTableAnalysisJob) GetTableID

func (j *StaticPartitionedTableAnalysisJob) GetTableID() int64

GetTableID gets the table ID of the job.

func (*StaticPartitionedTableAnalysisJob) GetWeight

GetWeight implements AnalysisJob.

func (*StaticPartitionedTableAnalysisJob) HasNewlyAddedIndex

func (j *StaticPartitionedTableAnalysisJob) HasNewlyAddedIndex() bool

HasNewlyAddedIndex implements AnalysisJob.

func (*StaticPartitionedTableAnalysisJob) RegisterFailureHook

func (j *StaticPartitionedTableAnalysisJob) RegisterFailureHook(hook FailureJobHook)

RegisterFailureHook registers a failureHook function that will be called after the job can be marked as failed.

func (*StaticPartitionedTableAnalysisJob) RegisterSuccessHook

func (j *StaticPartitionedTableAnalysisJob) RegisterSuccessHook(hook SuccessJobHook)

RegisterSuccessHook registers a successHook function that will be called after the job can be marked as successful.

func (*StaticPartitionedTableAnalysisJob) SetIndicators

func (j *StaticPartitionedTableAnalysisJob) SetIndicators(indicators Indicators)

SetIndicators implements AnalysisJob.

func (*StaticPartitionedTableAnalysisJob) SetWeight

func (j *StaticPartitionedTableAnalysisJob) SetWeight(weight float64)

SetWeight implements AnalysisJob.

func (*StaticPartitionedTableAnalysisJob) String

String implements fmt.Stringer interface.

func (*StaticPartitionedTableAnalysisJob) ValidateAndPrepare

func (j *StaticPartitionedTableAnalysisJob) ValidateAndPrepare(
	sctx sessionctx.Context,
) (bool, string)

ValidateAndPrepare validates if the analysis job can run and prepares it for execution. For static partitioned tables, it checks: - Schema exists - Table exists and is partitioned - Specified partition exists - No recent failed analysis to avoid queue blocking

type SuccessJobHook

type SuccessJobHook func(job AnalysisJob)

SuccessJobHook is the successHook function that will be called after the job is completed.

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL