Skip to content

PARQUET-1246: Ignore float/double statistics in case of NaN #461

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

Closed
gszadovszky wants to merge 2 commits into from
Closed

PARQUET-1246: Ignore float/double statistics in case of NaN #461

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
3447938
PARQUET-1246: Ignore float/double statistics in case of NaN
Mar 13, 2018
20e9332
PARQUET-1246: Changes according to zi's comments
Mar 14, 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
Prev Previous commit
PARQUET-1246: Changes according to zi's comments
  • Loading branch information
Gabor Szadovszky committed Mar 14, 2018
commit 20e93329904b88bce88495e916118d1eab4f100b
18 changes: 10 additions & 8 deletions parquet-column/src/main/java/org/apache/parquet/column/statistics/Statistics.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,12 +92,13 @@ public Statistics<?> build() {
((Statistics<?>) stats).hasNonNullValue = false;
} else {
// Updating min to -0.0 and max to +0.0 to ensure that no 0.0 values would be skipped
if (min == 0.0f) {
stats.setMinMax(-0.0f, max);
if (Float.compare(min, 0.0f) == 0) {
min = -0.0f;
Copy link
Contributor

@zivanfi zivanfi Mar 14, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

(nit) In my opinion,

min = -0.0f;
stats.setMinMax(min, max);

would be cleaner. Similar for the max case (although max is not used any more in this function).

stats.setMinMax(min, max);
}
if (max == -0.0f) {
stats.setMinMax(min, 0.0f);
if (Float.compare(max, -0.0f) == 0) {
max = 0.0f;
stats.setMinMax(min, max);
}
}
}
Expand All @@ -124,12 +125,13 @@ public Statistics<?> build() {
((Statistics<?>) stats).hasNonNullValue = false;
} else {
// Updating min to -0.0 and max to +0.0 to ensure that no 0.0 values would be skipped
if (min == 0.0) {
stats.setMinMax(-0.0, max);
if (Double.compare(min, 0.0) == 0) {
min = -0.0;
stats.setMinMax(min, max);
}
if (max == -0.0) {
stats.setMinMax(min, 0.0);
if (Double.compare(max, -0.0) == 0) {
max = 0.0;
stats.setMinMax(min, max);
}
}
}
Expand Down
36 changes: 36 additions & 0 deletions parquet-column/src/test/java/org/apache/parquet/column/statistics/TestStatistics.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -724,6 +724,24 @@ public void testSpecBuilderForFloat() {
assertTrue(stats.isNumNullsSet());
assertEquals(42, stats.getNumNulls());
assertFalse(stats.hasNonNullValue());

builder = Statistics.getBuilderForReading(type);
stats = builder.withMin(intToBytes(floatToIntBits(0.0f)))
.withMax(intToBytes(floatToIntBits(42.0f))).build();
assertEquals(0, Float.compare(-0.0f, (Float) stats.genericGetMin()));
assertEquals(0, Float.compare(42.0f, (Float) stats.genericGetMax()));

builder = Statistics.getBuilderForReading(type);
stats = builder.withMin(intToBytes(floatToIntBits(-42.0f)))
.withMax(intToBytes(floatToIntBits(-0.0f))).build();
assertEquals(0, Float.compare(-42.0f, (Float) stats.genericGetMin()));
assertEquals(0, Float.compare(0.0f, (Float) stats.genericGetMax()));

builder = Statistics.getBuilderForReading(type);
stats = builder.withMin(intToBytes(floatToIntBits(0.0f)))
.withMax(intToBytes(floatToIntBits(-0.0f))).build();
assertEquals(0, Float.compare(-0.0f, (Float) stats.genericGetMin()));
assertEquals(0, Float.compare(0.0f, (Float) stats.genericGetMax()));
}

@Test
Expand All @@ -749,5 +767,23 @@ public void testSpecBuilderForDouble() {
assertTrue(stats.isNumNullsSet());
assertEquals(42, stats.getNumNulls());
assertFalse(stats.hasNonNullValue());

builder = Statistics.getBuilderForReading(type);
stats = builder.withMin(longToBytes(doubleToLongBits(0.0)))
.withMax(longToBytes(doubleToLongBits(42.0))).build();
assertEquals(0, Double.compare(-0.0, (Double) stats.genericGetMin()));
assertEquals(0, Double.compare(42.0, (Double) stats.genericGetMax()));

builder = Statistics.getBuilderForReading(type);
stats = builder.withMin(longToBytes(doubleToLongBits(-42.0)))
.withMax(longToBytes(doubleToLongBits(-0.0))).build();
assertEquals(0, Double.compare(-42.0, (Double) stats.genericGetMin()));
assertEquals(0, Double.compare(0.0, (Double) stats.genericGetMax()));

builder = Statistics.getBuilderForReading(type);
stats = builder.withMin(longToBytes(doubleToLongBits(0.0)))
.withMax(longToBytes(doubleToLongBits(-0.0))).build();
assertEquals(0, Double.compare(-0.0, (Double) stats.genericGetMin()));
assertEquals(0, Double.compare(0.0, (Double) stats.genericGetMax()));
}
}