Skip to content
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

Jump to bottom

[SEDONA-607] Return geometry with ST functions with exceptions #1525

Merged
merged 12 commits into from
Jul 23, 2024
Merged

[SEDONA-607] Return geometry with ST functions with exceptions #1525

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
02494dd
[SEDONA-607] New function ST_SafeGeom to return geometry with exceptions
zhangfengcdt Jul 17, 2024
ab3dae4
Merge branch 'master' of github.com:zhangfengcdt/sedona into sedona-6…
zhangfengcdt Jul 17, 2024
05db1ac
[SEDONA-607] Include Geometry in ST Function Exceptions
zhangfengcdt Jul 17, 2024
b9d3b3d
throw exception on serialization of geometry
zhangfengcdt Jul 17, 2024
2e5b91a
refactor the code
zhangfengcdt Jul 17, 2024
15872ce
fix unit tests
zhangfengcdt Jul 18, 2024
59181ac
fix additional tests
zhangfengcdt Jul 18, 2024
b9eb22a
Merge branch 'master' of github.com:zhangfengcdt/sedona into sedona-6…
zhangfengcdt Jul 18, 2024
0df050c
refactor to add exception handling in InferredExpression
zhangfengcdt Jul 19, 2024
9734f98
add tests and comments
zhangfengcdt Jul 19, 2024
4acf6dc
Refactor InferredExpression to include the input row in the exception…
zhangfengcdt Jul 19, 2024
dcb6402
reformat
zhangfengcdt Jul 20, 2024
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
167 changes: 99 additions & 68 deletions spark/common/src/main/scala/org/apache/spark/sql/sedona_sql/expressions/Constructors.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -160,16 +160,21 @@ case class ST_GeomFromWKB(inputExpressions: Seq[Expression])
override def nullable: Boolean = true

override def eval(inputRow: InternalRow): Any = {
(inputExpressions.head.eval(inputRow)) match {
case (geomString: UTF8String) => {
// Parse UTF-8 encoded wkb string
Constructors.geomFromText(geomString.toString, FileDataSplitter.WKB).toGenericArrayData
}
case (wkb: Array[Byte]) => {
// convert raw wkb byte array to geometry
Constructors.geomFromWKB(wkb).toGenericArrayData
try {
(inputExpressions.head.eval(inputRow)) match {
case (geomString: UTF8String) => {
// Parse UTF-8 encoded wkb string
Constructors.geomFromText(geomString.toString, FileDataSplitter.WKB).toGenericArrayData
}
case (wkb: Array[Byte]) => {
// convert raw wkb byte array to geometry
Constructors.geomFromWKB(wkb).toGenericArrayData
}
case null => null
}
case null => null
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(inputRow, inputExpressions, e)
}
}

Expand All @@ -196,16 +201,21 @@ case class ST_GeomFromEWKB(inputExpressions: Seq[Expression])
override def nullable: Boolean = true

override def eval(inputRow: InternalRow): Any = {
(inputExpressions.head.eval(inputRow)) match {
case (geomString: UTF8String) => {
// Parse UTF-8 encoded wkb string
Constructors.geomFromText(geomString.toString, FileDataSplitter.WKB).toGenericArrayData
}
case (wkb: Array[Byte]) => {
// convert raw wkb byte array to geometry
Constructors.geomFromWKB(wkb).toGenericArrayData
try {
(inputExpressions.head.eval(inputRow)) match {
case (geomString: UTF8String) => {
// Parse UTF-8 encoded wkb string
Constructors.geomFromText(geomString.toString, FileDataSplitter.WKB).toGenericArrayData
}
case (wkb: Array[Byte]) => {
// convert raw wkb byte array to geometry
Constructors.geomFromWKB(wkb).toGenericArrayData
}
case null => null
}
case null => null
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(inputRow, inputExpressions, e)
}
}

Expand Down Expand Up @@ -238,21 +248,26 @@ case class ST_LineFromWKB(inputExpressions: Seq[Expression])
if (inputExpressions.length > 1) inputExpressions(1).eval(inputRow).asInstanceOf[Int]
else -1

wkb match {
case geomString: UTF8String =>
// Parse UTF-8 encoded WKB string
val geom = Constructors.lineStringFromText(geomString.toString, "wkb")
if (geom.getGeometryType == "LineString") {
(if (srid != -1) Functions.setSRID(geom, srid) else geom).toGenericArrayData
} else {
null
}

case wkbArray: Array[Byte] =>
// Convert raw WKB byte array to geometry
Constructors.lineFromWKB(wkbArray, srid).toGenericArrayData

case _ => null
try {
wkb match {
case geomString: UTF8String =>
// Parse UTF-8 encoded WKB string
val geom = Constructors.lineStringFromText(geomString.toString, "wkb")
if (geom.getGeometryType == "LineString") {
(if (srid != -1) Functions.setSRID(geom, srid) else geom).toGenericArrayData
} else {
null
}

case wkbArray: Array[Byte] =>
// Convert raw WKB byte array to geometry
Constructors.lineFromWKB(wkbArray, srid).toGenericArrayData

case _ => null
}
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(inputRow, inputExpressions, e)
}
}

Expand Down Expand Up @@ -287,21 +302,26 @@ case class ST_LinestringFromWKB(inputExpressions: Seq[Expression])
if (inputExpressions.length > 1) inputExpressions(1).eval(inputRow).asInstanceOf[Int]
else -1

wkb match {
case geomString: UTF8String =>
// Parse UTF-8 encoded WKB string
val geom = Constructors.lineStringFromText(geomString.toString, "wkb")
if (geom.getGeometryType == "LineString") {
(if (srid != -1) Functions.setSRID(geom, srid) else geom).toGenericArrayData
} else {
null
}

case wkbArray: Array[Byte] =>
// Convert raw WKB byte array to geometry
Constructors.lineFromWKB(wkbArray, srid).toGenericArrayData

case _ => null
try {
wkb match {
case geomString: UTF8String =>
// Parse UTF-8 encoded WKB string
val geom = Constructors.lineStringFromText(geomString.toString, "wkb")
if (geom.getGeometryType == "LineString") {
(if (srid != -1) Functions.setSRID(geom, srid) else geom).toGenericArrayData
} else {
null
}

case wkbArray: Array[Byte] =>
// Convert raw WKB byte array to geometry
Constructors.lineFromWKB(wkbArray, srid).toGenericArrayData

case _ => null
}
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(inputRow, inputExpressions, e)
}
}

Expand Down Expand Up @@ -336,21 +356,26 @@ case class ST_PointFromWKB(inputExpressions: Seq[Expression])
if (inputExpressions.length > 1) inputExpressions(1).eval(inputRow).asInstanceOf[Int]
else -1

wkb match {
case geomString: UTF8String =>
// Parse UTF-8 encoded WKB string
val geom = Constructors.pointFromText(geomString.toString, "wkb")
if (geom.getGeometryType == "Point") {
(if (srid != -1) Functions.setSRID(geom, srid) else geom).toGenericArrayData
} else {
null
}

case wkbArray: Array[Byte] =>
// Convert raw WKB byte array to geometry
Constructors.pointFromWKB(wkbArray, srid).toGenericArrayData

case _ => null
try {
wkb match {
case geomString: UTF8String =>
// Parse UTF-8 encoded WKB string
val geom = Constructors.pointFromText(geomString.toString, "wkb")
if (geom.getGeometryType == "Point") {
(if (srid != -1) Functions.setSRID(geom, srid) else geom).toGenericArrayData
} else {
null
}

case wkbArray: Array[Byte] =>
// Convert raw WKB byte array to geometry
Constructors.pointFromWKB(wkbArray, srid).toGenericArrayData

case _ => null
}
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(inputRow, inputExpressions, e)
}
}

Expand Down Expand Up @@ -387,12 +412,18 @@ case class ST_GeomFromGeoJSON(inputExpressions: Seq[Expression])

override def eval(inputRow: InternalRow): Any = {
val geomString = inputExpressions.head.eval(inputRow).asInstanceOf[UTF8String].toString
val geometry = Constructors.geomFromText(geomString, FileDataSplitter.GEOJSON)
// If the user specify a bunch of attributes to go with each geometry, we need to store all of them in this geometry
if (inputExpressions.length > 1) {
geometry.setUserData(generateUserData(minInputLength, inputExpressions, inputRow))
try {

val geometry = Constructors.geomFromText(geomString, FileDataSplitter.GEOJSON)
// If the user specify a bunch of attributes to go with each geometry, we need to store all of them in this geometry
if (inputExpressions.length > 1) {
geometry.setUserData(generateUserData(minInputLength, inputExpressions, inputRow))
}
GeometrySerializer.serialize(geometry)
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(inputRow, inputExpressions, e)
}
GeometrySerializer.serialize(geometry)
}

override def dataType: DataType = GeometryUDT
Expand Down
80 changes: 51 additions & 29 deletions spark/common/src/main/scala/org/apache/spark/sql/sedona_sql/expressions/Functions.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -332,13 +332,18 @@ case class ST_IsValidDetail(children: Seq[Expression])
throw new IllegalArgumentException(s"Invalid number of arguments: $nArgs")
}

if (validDetail.location == null) {
return InternalRow.fromSeq(Seq(validDetail.valid, null, null))
}
try {
if (validDetail.location == null) {
return InternalRow.fromSeq(Seq(validDetail.valid, null, null))
}

val serLocation = GeometrySerializer.serialize(validDetail.location)
InternalRow.fromSeq(
Seq(validDetail.valid, UTF8String.fromString(validDetail.reason), serLocation))
val serLocation = GeometrySerializer.serialize(validDetail.location)
InternalRow.fromSeq(
Seq(validDetail.valid, UTF8String.fromString(validDetail.reason), serLocation))
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(input, children, e)
}
}

protected def withNewChildrenInternal(newChildren: IndexedSeq[Expression]): Expression = {
Expand Down Expand Up @@ -609,14 +614,20 @@ case class ST_MinimumBoundingRadius(inputExpressions: Seq[Expression])

override def eval(input: InternalRow): Any = {
val expr = inputExpressions(0)
val geometry = expr match {
case s: SerdeAware => s.evalWithoutSerialization(input)
case _ => expr.toGeometry(input)
}

geometry match {
case geometry: Geometry => getMinimumBoundingRadius(geometry)
case _ => null
try {
val geometry = expr match {
case s: SerdeAware => s.evalWithoutSerialization(input)
case _ => expr.toGeometry(input)
}

geometry match {
case geometry: Geometry => getMinimumBoundingRadius(geometry)
case _ => null
}
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(input, inputExpressions, e)
}
}

Expand Down Expand Up @@ -910,17 +921,23 @@ case class ST_SubDivideExplode(children: Seq[Expression]) extends Generator with
override def eval(input: InternalRow): TraversableOnce[InternalRow] = {
val geometryRaw = children.head
val maxVerticesRaw = children(1)
geometryRaw.toGeometry(input) match {
case geom: Geometry =>
ArrayData.toArrayData(
Functions.subDivide(geom, maxVerticesRaw.toInt(input)).map(_.toGenericArrayData))
Functions
.subDivide(geom, maxVerticesRaw.toInt(input))
.map(_.toGenericArrayData)
.map(InternalRow(_))
case _ => new Array[InternalRow](0)
try {
geometryRaw.toGeometry(input) match {
case geom: Geometry =>
ArrayData.toArrayData(
Functions.subDivide(geom, maxVerticesRaw.toInt(input)).map(_.toGenericArrayData))
Functions
.subDivide(geom, maxVerticesRaw.toInt(input))
.map(_.toGenericArrayData)
.map(InternalRow(_))
case _ => new Array[InternalRow](0)
}
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(input, children, e)
}
}

override def elementSchema: StructType = {
new StructType()
.add("geom", GeometryUDT, true)
Expand Down Expand Up @@ -978,13 +995,18 @@ case class ST_MaximumInscribedCircle(children: Seq[Expression])
with CodegenFallback {

override def eval(input: InternalRow): Any = {
val geometry = children.head.toGeometry(input)
var inscribedCircle: InscribedCircle = null
inscribedCircle = Functions.maximumInscribedCircle(geometry)

val serCenter = GeometrySerializer.serialize(inscribedCircle.center)
val serNearest = GeometrySerializer.serialize(inscribedCircle.nearest)
InternalRow.fromSeq(Seq(serCenter, serNearest, inscribedCircle.radius))
try {
val geometry = children.head.toGeometry(input)
var inscribedCircle: InscribedCircle = null
inscribedCircle = Functions.maximumInscribedCircle(geometry)

val serCenter = GeometrySerializer.serialize(inscribedCircle.center)
val serNearest = GeometrySerializer.serialize(inscribedCircle.nearest)
InternalRow.fromSeq(Seq(serCenter, serNearest, inscribedCircle.radius))
} catch {
case e: Exception =>
InferredExpression.throwExpressionInferenceException(input, children, e)
}
}

protected def withNewChildrenInternal(newChildren: IndexedSeq[Expression]): Expression = {
Expand Down
Loading
Loading