Package org.elasticsearch.xpack.esql.datasources.spi

Class FormatReadContext.Builder

java.lang.Object

org.elasticsearch.xpack.esql.datasources.spi.FormatReadContext.Builder

Enclosing class:

FormatReadContext

public static final class FormatReadContext.Builder
extends Object

Method Summary

Modifier and Type	Method	Description
FormatReadContext.Builder	batchSize(int batchSize)
FormatReadContext	build()
FormatReadContext.Builder	errorPolicy(ErrorPolicy errorPolicy)
FormatReadContext.Builder	firstSplit(boolean firstSplit)
FormatReadContext.Builder	informationalWarningSink(Consumer<String> informationalWarningSink)	See FormatReadContext.informationalWarningSink(); pass null for direct-to-HeaderWarning emission.
FormatReadContext.Builder	lastSplit(boolean lastSplit)
FormatReadContext.Builder	maxRecordBytes(int maxRecordBytes)
FormatReadContext.Builder	projectedColumns(List<String> projectedColumns)
FormatReadContext.Builder	readSchema(List<Attribute> readSchema)	See FormatReadContext.readSchema(); pass null to fall back to per-file inference.
FormatReadContext.Builder	recordAligned(boolean recordAligned)	Marks the split as starting at a record boundary so line-oriented readers can skip the "trim leading partial record" workaround used for byte-range macro-splits.
FormatReadContext.Builder	rowLimit(int rowLimit)
FormatReadContext.Builder	splitStartByte(long splitStartByte)	See FormatReadContext.splitStartByte(); the file-global byte offset of this split's start.
FormatReadContext.Builder	stats(long baseOffset, long stripeSize, boolean fileFinal)	Canonical-stripe addressing for per-stripe stats capture: baseOffset is this read's first byte in file/decompressed coordinates, stripeSize the grid (<= 0 disables), fileFinal whether this read reaches the file's true end (only the final read may mark its last stripe complete + terminal).
FormatReadContext.Builder	statsColumnScope(StripeColumnScope statsColumnScope)	Sets how much per-stripe statistics the read harvests (see FormatReadContext.statsColumnScope()).

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Details

projectedColumns

public FormatReadContext.Builder projectedColumns(List<String> projectedColumns)

batchSize

public FormatReadContext.Builder batchSize(int batchSize)

rowLimit

public FormatReadContext.Builder rowLimit(int rowLimit)

errorPolicy

public FormatReadContext.Builder errorPolicy(ErrorPolicy errorPolicy)

firstSplit

public FormatReadContext.Builder firstSplit(boolean firstSplit)

lastSplit

public FormatReadContext.Builder lastSplit(boolean lastSplit)

recordAligned

public FormatReadContext.Builder recordAligned(boolean recordAligned)

Marks the split as starting at a record boundary so line-oriented readers can skip the "trim leading partial record" workaround used for byte-range macro-splits.

readSchema

public FormatReadContext.Builder readSchema(@Nullable
 List<Attribute> readSchema)

See FormatReadContext.readSchema(); pass null to fall back to per-file inference.

splitStartByte

public FormatReadContext.Builder splitStartByte(long splitStartByte)

See FormatReadContext.splitStartByte(); the file-global byte offset of this split's start.

maxRecordBytes

public FormatReadContext.Builder maxRecordBytes(int maxRecordBytes)

stats

public FormatReadContext.Builder stats(long baseOffset,
 long stripeSize,
 boolean fileFinal)

Canonical-stripe addressing for per-stripe stats capture: baseOffset is this read's first byte in file/decompressed coordinates, stripeSize the grid (<= 0 disables), fileFinal whether this read reaches the file's true end (only the final read may mark its last stripe complete + terminal). A pure stats overlay; never affects chunking or splitting.

statsColumnScope

public FormatReadContext.Builder statsColumnScope(@Nullable
 StripeColumnScope statsColumnScope)

Sets how much per-stripe statistics the read harvests (see FormatReadContext.statsColumnScope()). null restores the StripeColumnScope.PROJECTED default. Orthogonal to stats(long, long, boolean).

informationalWarningSink

public FormatReadContext.Builder informationalWarningSink(@Nullable
 Consumer<String> informationalWarningSink)

See FormatReadContext.informationalWarningSink(); pass null for direct-to-HeaderWarning emission.

build

public FormatReadContext build()