Skip to content

PARQUET-1213: Column indexes: Limit index size #480

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
gszadovszky merged 2 commits into from
May 28, 2018
Merged

PARQUET-1213: Column indexes: Limit index size #480

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
15273fb
PARQUET-1213: Column indexes: Limit index size
Feb 14, 2018
d015feb
PARQUET-1213: fix review comments
gszadovszky May 24, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions ...rc/main/java/org/apache/parquet/internal/column/columnindex/BinaryColumnIndexBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,4 +102,9 @@ int compareMinValues(PrimitiveComparator<Binary> comparator, int index1, int ind
int compareMaxValues(PrimitiveComparator<Binary> comparator, int index1, int index2) {
return comparator.compare(maxValues.get(index1), maxValues.get(index2));
}

@Override
int sizeOf(Object value) {
return ((Binary) value).length();
}
}
5 changes: 5 additions & 0 deletions ...c/main/java/org/apache/parquet/internal/column/columnindex/BooleanColumnIndexBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,4 +103,9 @@ int compareMinValues(PrimitiveComparator<Binary> comparator, int index1, int ind
int compareMaxValues(PrimitiveComparator<Binary> comparator, int index1, int index2) {
return comparator.compare(maxValues.get(index1), maxValues.get(index2));
}

@Override
int sizeOf(Object value) {
return 1;
}
}
37 changes: 34 additions & 3 deletions ...lumn/src/main/java/org/apache/parquet/internal/column/columnindex/ColumnIndexBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -195,13 +195,19 @@ int compareMinValues(PrimitiveComparator<Binary> comparator, int index1, int ind
int compareMaxValues(PrimitiveComparator<Binary> comparator, int index1, int index2) {
return 0;
}

@Override
int sizeOf(Object value) {
return 0;
}
};

private static final Map<PrimitiveTypeName, ColumnIndexBuilder> BUILDERS = new EnumMap<>(PrimitiveTypeName.class);

private PrimitiveType type;
private final BooleanList nullPages = new BooleanArrayList();
private final LongList nullCounts = new LongArrayList();
private long minMaxSize;

/**
* @return a no-op builder that does not collect statistics objects and therefore returns {@code null} at
Expand Down Expand Up @@ -293,7 +299,11 @@ public static ColumnIndex build(
public void add(Statistics<?> stats) {
if (stats.hasNonNullValue()) {
nullPages.add(false);
addMinMax(stats.genericGetMin(), stats.genericGetMax());
Object min = stats.genericGetMin();
Object max = stats.genericGetMax();
addMinMax(min, max);
minMaxSize += sizeOf(min);
minMaxSize += sizeOf(max);
} else {
nullPages.add(true);
addMinMax(null, null);
Expand All @@ -316,7 +326,7 @@ private void fill(List<Boolean> nullPages, List<Long> nullCounts, List<ByteBuffe
nullPages.size(), nullCounts == null ? "null" : nullCounts.size(), minValues.size(), maxValues.size()));
}
this.nullPages.addAll(nullPages);
// Null counts is optional in the format
// Nullcounts is optional in the format
if (nullCounts != null) {
this.nullCounts.addAll(nullCounts);
}
Expand All @@ -325,7 +335,11 @@ private void fill(List<Boolean> nullPages, List<Long> nullCounts, List<ByteBuffe
if (nullPages.get(i)) {
addMinMaxFromBytes(null, null);
} else {
addMinMaxFromBytes(minValues.get(i), maxValues.get(i));
ByteBuffer min = minValues.get(i);
ByteBuffer max = maxValues.get(i);
addMinMaxFromBytes(min, max);
minMaxSize += min.remaining();
minMaxSize += max.remaining();
}
}
}
Expand Down Expand Up @@ -421,9 +435,26 @@ private void clear() {
nullPages.clear();
nullCounts.clear();
clearMinMax();
minMaxSize = 0;
}

abstract void clearMinMax();

abstract ColumnIndexBase createColumnIndex(PrimitiveType type);

abstract int sizeOf(Object value);

/**
* @return the number of pages added so far to this builder
*/
public int getPageCount() {
return nullPages.size();
}

/**
* @return the sum of size in bytes of the min/max values added so far to this builder
*/
public long getMinMaxSize() {
return minMaxSize;
}
}
7 changes: 6 additions & 1 deletion ...rc/main/java/org/apache/parquet/internal/column/columnindex/DoubleColumnIndexBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ private static double convert(ByteBuffer buffer) {
}

private static ByteBuffer convert(double value) {
return ByteBuffer.allocate(Double.SIZE / 8).order(LITTLE_ENDIAN).putDouble(0, value);
return ByteBuffer.allocate(Double.BYTES).order(LITTLE_ENDIAN).putDouble(0, value);
}

@Override
Expand Down Expand Up @@ -105,4 +105,9 @@ int compareMinValues(PrimitiveComparator<Binary> comparator, int index1, int ind
int compareMaxValues(PrimitiveComparator<Binary> comparator, int index1, int index2) {
return comparator.compare(maxValues.get(index1), maxValues.get(index2));
}

@Override
int sizeOf(Object value) {
return Double.BYTES;
}
}
7 changes: 6 additions & 1 deletion ...src/main/java/org/apache/parquet/internal/column/columnindex/FloatColumnIndexBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ private static float convert(ByteBuffer buffer) {
}

private static ByteBuffer convert(float value) {
return ByteBuffer.allocate(Float.SIZE / 8).order(LITTLE_ENDIAN).putFloat(0, value);
return ByteBuffer.allocate(Float.BYTES).order(LITTLE_ENDIAN).putFloat(0, value);
}

@Override
Expand Down Expand Up @@ -105,4 +105,9 @@ int compareMinValues(PrimitiveComparator<Binary> comparator, int index1, int ind
int compareMaxValues(PrimitiveComparator<Binary> comparator, int index1, int index2) {
return comparator.compare(maxValues.get(index1), maxValues.get(index2));
}

@Override
int sizeOf(Object value) {
return Float.BYTES;
}
}
7 changes: 6 additions & 1 deletion ...n/src/main/java/org/apache/parquet/internal/column/columnindex/IntColumnIndexBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ private static int convert(ByteBuffer buffer) {
}

private static ByteBuffer convert(int value) {
return ByteBuffer.allocate(Integer.SIZE / 8).order(LITTLE_ENDIAN).putInt(0, value);
return ByteBuffer.allocate(Integer.BYTES).order(LITTLE_ENDIAN).putInt(0, value);
}

@Override
Expand Down Expand Up @@ -105,4 +105,9 @@ int compareMinValues(PrimitiveComparator<Binary> comparator, int index1, int ind
int compareMaxValues(PrimitiveComparator<Binary> comparator, int index1, int index2) {
return comparator.compare(maxValues.get(index1), maxValues.get(index2));
}

@Override
int sizeOf(Object value) {
return Integer.BYTES;
}
}
7 changes: 6 additions & 1 deletion .../src/main/java/org/apache/parquet/internal/column/columnindex/LongColumnIndexBuilder.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -67,7 +67,7 @@ private static long convert(ByteBuffer buffer) {
}

private static ByteBuffer convert(long value) {
return ByteBuffer.allocate(Long.SIZE / 8).order(LITTLE_ENDIAN).putLong(0, value);
return ByteBuffer.allocate(Long.BYTES).order(LITTLE_ENDIAN).putLong(0, value);
}

@Override
Expand Down Expand Up @@ -105,4 +105,9 @@ int compareMinValues(PrimitiveComparator<Binary> comparator, int index1, int ind
int compareMaxValues(PrimitiveComparator<Binary> comparator, int index1, int index2) {
return comparator.compare(maxValues.get(index1), maxValues.get(index2));
}

@Override
int sizeOf(Object value) {
return Long.BYTES;
}
}
Loading