Support for reading legacy date in Hive for Parquet #24611
Draft
+1,465
−645
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Praveen2112
reviewed
Jan 3, 2025
| @@ -281,7 +281,7 @@ public static ReaderPageSource createPageSource( | |||
| exception -> handleException(dataSourceId, exception), | |||
| // We avoid using disjuncts of parquetPredicate for page pruning in ParquetReader as currently column indexes | |||
| // are not present in the Parquet files which are read with disjunct predicates. | |||
| parquetPredicates.size() == 1 ? Optional.of(parquetPredicates.get(0)) : Optional.empty(), | |||
| parquetPredicates.size() == 1 ? Optional.of(parquetPredicates.getFirst()) : Optional.empty(), | |||
| @@ -265,7 +264,7 @@ public static ReaderPageSource createPageSource( | |||
| List<HiveColumnHandle> baseColumns = readerProjections.map(projection -> | |||
| projection.get().stream() | |||
| .map(HiveColumnHandle.class::cast) | |||
| .collect(toUnmodifiableList())) | |||
| .toList()) | |||
| @@ -255,38 +255,27 @@ public Block getBlock(Page sourcePage, long startRowId) | |||
| } | |||
| } | |||
|
|
|||
| private static class SourceColumn | |||
| private record SourceColumn(int sourceChannel) | |||
There was a problem hiding this comment.
In general record can be used for POJO but this implements ColumnAdaptation which encapsulates function like details like getBlock which is something beyond POJO so do we need this record conversion ?
Comment on lines
298
to
299
| private record CoercedColumn(SourceColumn sourceColumn, TypeCoercer<?, ?> typeCoercer) | ||
| implements ColumnAdaptation |
| @@ -125,7 +125,7 @@ public ColumnReader create(PrimitiveField field, AggregatedMemoryContext aggrega | |||
| } | |||
| if (DATE.equals(type) && primitiveType == INT32) { | |||
| if (annotation == null || annotation instanceof DateLogicalTypeAnnotation) { | |||
| return createColumnReader(field, valueDecoders::getIntDecoder, INT_ADAPTER, memoryContext); | |||
| return createColumnReader(field, valueDecoders::getInt32Decoder, INT_ADAPTER, memoryContext); | |||
There was a problem hiding this comment.
This is for date type right ?
| @@ -36,6 +36,7 @@ public class ParquetReaderOptions | |||
| private final boolean useBloomFilter; | |||
| private final DataSize smallFileThreshold; | |||
| private final boolean vectorizedDecodingEnabled; | |||
| private final boolean hybridCalendarEnabled; | |||
| return julianDays; | ||
| } | ||
|
|
||
| public static int convertDaysToHybridCalendar(int prolepticDays) |
There was a problem hiding this comment.
Is this used only for testing ? In that case - can we move to test class ?
Comment on lines
29
to
34
| private static final int[] JULIAN_GREGORIAN_DIFFS = {2, 1, 0, -1, -2, -3, -4, -5, -6, -7, -8, -9, -10, 0}; | ||
| private static final int[] JULIAN_GREGORIAN_DIFF_SWITCH_DAY = { | ||
| -719164, -682945, -646420, -609895, -536845, | ||
| -500320, -463795, -390745, -354220, -317695, | ||
| -244645, -208120, -171595, -141427 | ||
| }; |
There was a problem hiding this comment.
Can we build a RangeMap and maintain it ? It would allow us to identify on which set of ranges which dates should we consider ?
Comment on lines
79
to
81
| else if (julianDays < LAST_SWITCH_JULIAN_DAY) { | ||
| return rebaseDays(JULIAN_GREGORIAN_DIFF_SWITCH_DAY, JULIAN_GREGORIAN_DIFFS, julianDays); | ||
| } |
There was a problem hiding this comment.
Do we need a dedicated conversion for rebase - Can we use convertHybridDaysToProlepticDays ?
There was a problem hiding this comment.
yes we can, however rebaseDays method will be faster
Comment on lines
136
to
141
| if (shouldUseHybridCalendarForDate) { | ||
| return createColumnReader(field, encoding -> getDateInt32Decoder(valueDecoders.getInt32Decoder(encoding)), INT_ADAPTER, memoryContext); | ||
| } | ||
| else { | ||
| return createColumnReader(field, valueDecoders::getInt32Decoder, INT_ADAPTER, memoryContext); | ||
| } |
There was a problem hiding this comment.
Can we implement it as a ColumnAdaptation ? In this case the Date/Timestamp conversion can be unified and we could use them for ORC as well.
marcinsbd
force-pushed
the
marcinsbd/support-legacy-date
branch
from
9c61f71 to
27ef0fc
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
#23904
Additional context and related issues
Release notes
( ) This is not user-visible or is docs only, and no release notes are required.
( ) Release notes are required. Please propose a release note for me.
( ) Release notes are required, with the following suggested text: