Skip to content

[SPARK-21552][SQL] Add DecimalType support to ArrowWriter. #18754

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
ueshin wants to merge 6 commits into from
Closed

[SPARK-21552][SQL] Add DecimalType support to ArrowWriter. #18754

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
9e60762
Add DecimalType support to ArrowWriter.
ueshin Jul 26, 2017
1fccaf8
Merge branch 'master' into issues/SPARK-21552
ueshin Dec 22, 2017
4a697b3
Use Arrow 0.8 APIs and fix a test.
ueshin Dec 12, 2017
bcadac8
Fix Python side.
ueshin Dec 18, 2017
e29f833
Modify tests for unsupported types.
ueshin Dec 18, 2017
025f298
Check the return value of `changePrecision()` for the case of overflow.
ueshin Dec 25, 2017
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
Next Next commit
Use Arrow 0.8 APIs and fix a test.
  • Loading branch information
@ueshin
ueshin committed Dec 22, 2017
commit 4a697b3048218df5ae8167ba88ff9a1e9bcba60b
11 changes: 4 additions & 7 deletions sql/core/src/main/scala/org/apache/spark/sql/execution/arrow/ArrowWriter.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@ object ArrowWriter {
case (LongType, vector: BigIntVector) => new LongWriter(vector)
case (FloatType, vector: Float4Vector) => new FloatWriter(vector)
case (DoubleType, vector: Float8Vector) => new DoubleWriter(vector)
case (DecimalType.Fixed(precision, scale), vector: NullableDecimalVector) =>
case (DecimalType.Fixed(precision, scale), vector: DecimalVector) =>
new DecimalWriter(vector, precision, scale)
case (StringType, vector: VarCharVector) => new StringWriter(vector)
case (BinaryType, vector: VarBinaryVector) => new BinaryWriter(vector)
Expand Down Expand Up @@ -217,21 +217,18 @@ private[arrow] class DoubleWriter(val valueVector: Float8Vector) extends ArrowFi
}

private[arrow] class DecimalWriter(
val valueVector: NullableDecimalVector,
val valueVector: DecimalVector,
precision: Int,
scale: Int) extends ArrowFieldWriter {

override def valueMutator: NullableDecimalVector#Mutator = valueVector.getMutator()

override def setNull(): Unit = {
valueMutator.setNull(count)
valueVector.setNull(count)
}

override def setValue(input: SpecializedGetters, ordinal: Int): Unit = {
valueMutator.setIndexDefined(count)
val decimal = input.getDecimal(ordinal, precision, scale)
decimal.changePrecision(precision, scale)
Copy link
Member

@BryanCutler BryanCutler Dec 22, 2017

Choose a reason for hiding this comment

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

Is it necessary to call changePrecision even though getDecimal already takes the precision/scale as input - is it not guaranteed to return a decimal with that scale?

Copy link
Member Author

@ueshin ueshin Dec 25, 2017

Choose a reason for hiding this comment

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

Unfortunately, it depends on the implementation of getDecimal for now.
Btw, I guess we need to check the return value of changePrecision() and set null if the value is false, which means overflow.

DecimalUtility.writeBigDecimalToArrowBuf(decimal.toJavaBigDecimal, valueVector.getBuffer, count)
valueVector.setSafe(count, decimal.toJavaBigDecimal)
}
}

Expand Down
50 changes: 16 additions & 34 deletions sql/core/src/test/scala/org/apache/spark/sql/execution/arrow/ArrowConvertersSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -311,40 +311,22 @@ class ArrowConvertersSuite extends SharedSQLContext with BeforeAndAfterAll {
| "schema" : {
| "fields" : [ {
| "name" : "a_d",
| "nullable" : true,
| "type" : {
| "name" : "decimal",
| "precision" : 38,
| "scale" : 18
| },
| "children" : [ ],
| "typeLayout" : {
| "vectors" : [ {
| "type" : "VALIDITY",
| "typeBitWidth" : 1
| }, {
| "type" : "DATA",
| "typeBitWidth" : 64
| } ]
| }
| "nullable" : true,
| "children" : [ ]
| }, {
| "name" : "b_d",
| "nullable" : true,
| "type" : {
| "name" : "decimal",
| "precision" : 38,
| "scale" : 18
| },
| "children" : [ ],
| "typeLayout" : {
| "vectors" : [ {
| "type" : "VALIDITY",
| "typeBitWidth" : 1
| }, {
| "type" : "DATA",
| "typeBitWidth" : 64
| } ]
| }
| "nullable" : true,
| "children" : [ ]
| } ]
| },
| "batches" : [ {
Expand All @@ -354,23 +336,23 @@ class ArrowConvertersSuite extends SharedSQLContext with BeforeAndAfterAll {
| "count" : 6,
| "VALIDITY" : [ 1, 1, 1, 1, 1, 1 ],
| "DATA" : [
| 1.000000000000000000,
| 2.000000000000000000,
| 0.010000000000000000,
| 200.000000000000000000,
| 0.000100000000000000,
| 20000.000000000000000000 ]
| "1000000000000000000",
| "2000000000000000000",
| "10000000000000000",
| "200000000000000000000",
| "100000000000000",
| "20000000000000000000000" ]
| }, {
| "name" : "b_d",
| "count" : 6,
| "VALIDITY" : [ 1, 0, 0, 1, 0, 1 ],
| "DATA" : [
| 1.100000000000000000,
| 0E-18,
| 0E-18,
| 2.200000000000000000,
| 0E-18,
| 3.300000000000000000 ]
| "1100000000000000000",
| "0",
| "0",
| "2200000000000000000",
| "0",
| "3300000000000000000" ]
| } ]
| } ]
|}
Expand Down