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

Dfml30 #1050

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Dfml30 #1050

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
7d22167
Add tests.
ghislainfourny Jul 1, 2021
8291d3f
Merge branch 'master' into DFML30
ghislainfourny Jul 2, 2021
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
121 changes: 75 additions & 46 deletions src/main/java/org/rumbledb/runtime/typing/ValidateTypeIterator.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,13 +24,11 @@
import org.rumbledb.types.FieldDescriptor;
import org.rumbledb.types.ItemType;

import sparksoniq.spark.DataFrameUtils;
import sparksoniq.spark.SparkSessionManager;

import java.sql.Date;
import java.sql.Timestamp;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.Map;
Expand Down Expand Up @@ -68,11 +66,11 @@ public JSoundDataFrame getDataFrame(DynamicContext context) {

if (inputDataIterator.isDataFrame()) {
JSoundDataFrame inputDataAsDataFrame = inputDataIterator.getDataFrame(context);
validateItemTypeAgainstDataFrame(
checkAnnotationPossibleOrThrowError(
this.itemType,
inputDataAsDataFrame.getDataFrame().schema()
inputDataAsDataFrame.getItemType()
);
return inputDataAsDataFrame;
return new JSoundDataFrame(inputDataAsDataFrame.getDataFrame(), this.itemType);
}

if (inputDataIterator.isRDDOrDataFrame()) {
Expand Down Expand Up @@ -444,63 +442,94 @@ private static Object getRowColumnFromItemUsingDataType(Item item, ItemType item
}


private static void validateItemTypeAgainstDataFrame(
ItemType itemType,
StructType dataFrameSchema
private static void checkAnnotationPossibleOrThrowError(
ItemType expectedType,
ItemType actualType
) {
StructType generatedSchema = convertToDataFrameSchema(itemType);
for (StructField column : dataFrameSchema.fields()) {
final String columnName = column.name();
final DataType columnDataType = column.dataType();

boolean columnMatched = false;
for (StructField structField : generatedSchema.fields()) {
String generatedColumnName = structField.name();
if (!generatedColumnName.equals(columnName)) {
continue;
}
if (expectedType.isObjectItemType()) {
if (!actualType.isObjectItemType()) {
throw new InvalidInstanceException(
"Unexpected type. Expected "
+ expectedType
+ " but actually encountered "
+ actualType
);
}
// StructType generatedSchema = convertToDataFrameSchema(itemType);
for (String columnName : actualType.getObjectContentFacet().keySet()) {
final ItemType actualColumnType = actualType.getObjectContentFacet().get(columnName).getType();

boolean columnMatched = false;
for (String generatedColumnName : expectedType.getObjectContentFacet().keySet()) {
if (!generatedColumnName.equals(columnName)) {
continue;
}

DataType generatedDataType = structField.dataType();
if (DataFrameUtils.isUserTypeApplicable(generatedDataType, columnDataType)) {
ItemType expectedColumnType = expectedType.getObjectContentFacet()
.get(generatedColumnName)
.getType();
checkAnnotationPossibleOrThrowError(expectedColumnType, actualColumnType);
columnMatched = true;
}

throw new InvalidInstanceException(
"Fields defined in schema must fully match the fields of input data: "
+ "expected '"
+ ItemParser.getItemTypeNameFromDataFrameDataType(columnDataType)
+ "' type for field '"
+ columnName
+ "', but found '"
+ ItemParser.getItemTypeNameFromDataFrameDataType(generatedDataType)
+ "'"
);
if (expectedType != null && expectedType.getClosedFacet()) {
if (!columnMatched) {
throw new InvalidInstanceException(
"Unexpected key in closed object type "
+ expectedType.getIdentifierString()
+ " : "
+ columnName
);
}
}
}

if (itemType != null && itemType.getClosedFacet()) {
if (!columnMatched) {
for (String generatedSchemaColumnName : expectedType.getObjectContentFacet().keySet()) {
boolean userColumnMatched = actualType.getObjectContentFacet()
.keySet()
.contains(generatedSchemaColumnName);

if (!userColumnMatched) {
throw new InvalidInstanceException(
"Unexpected key in closed object type "
+ itemType.getIdentifierString()
+ " : "
+ columnName
"Fields defined in schema must fully match the fields of input data: "
+ "redundant type information for non-existent field '"
+ generatedSchemaColumnName
+ "'."
);
}
}
}

for (String generatedSchemaColumnName : generatedSchema.fieldNames()) {
boolean userColumnMatched = Arrays.asList(dataFrameSchema.fieldNames()).contains(generatedSchemaColumnName);

if (!userColumnMatched) {
if (expectedType.isArrayItemType()) {
if (!actualType.isArrayItemType()) {
throw new InvalidInstanceException(
"Fields defined in schema must fully match the fields of input data: "
+ "redundant type information for non-existent field '"
+ generatedSchemaColumnName
+ "'."
"Unexpected type. Expected "
+ expectedType
+ " but actually encountered "
+ actualType
);
}
}
if (expectedType.isAtomicItemType()) {
if (!actualType.isAtomicItemType()) {
throw new InvalidInstanceException(
"Unexpected type, and dataframe casts are not supported at this point (contact us to prioritize). Expected "
+ expectedType
+ " but actually encountered "
+ actualType
);
}
if (!actualType.isSubtypeOf(expectedType)) {
throw new InvalidInstanceException(
"Unexpected type, and dataframe casts are not supported at this point (contact us to prioritize). Expected "
+ expectedType
+ " but actually encountered "
+ actualType
);
}
}
if (expectedType.isUnionType()) {
throw new InvalidInstanceException("Union types are not supported at this point.");
}
}

@Override
Expand Down
13 changes: 0 additions & 13 deletions src/main/java/sparksoniq/spark/DataFrameUtils.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -246,17 +246,4 @@ public static void validateSchemaItemAgainstDataFrame(
}
}
}

public static boolean isUserTypeApplicable(
DataType userSchemaColumnDataType,
DataType columnDataType
) {
return userSchemaColumnDataType.equals(columnDataType)
||
(userSchemaColumnDataType.equals(ItemParser.decimalType) && columnDataType.equals(DataTypes.LongType))
||
(userSchemaColumnDataType.equals(DataTypes.DoubleType) && columnDataType.equals(DataTypes.FloatType))
||
(userSchemaColumnDataType.equals(DataTypes.IntegerType) && columnDataType.equals(DataTypes.ShortType));
}
}
6 changes: 3 additions & 3 deletions src/test/resources/test_files/runtime-native-flwor/udt/type-default.jq
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
(:JIQS: ShouldRun; Output="({ "foo" : "foo", "bar" : "foobar", "int" : 1 }, { "foo" : "foo", "bar" : "def", "int" : 42 }, { "foo" : "foo", "bar" : "foobar", "int" : 42 }, { "foo" : "foo", "bar" : "def", "int" : 42 }, { "bar" : "foobar", "int" : 42, "foo" : "foo" }, { "foo" : "foo", "bar" : "def", "int" : 42 }, { "bar" : "def", "int" : 42, "foo" : "foo" }, { "bar" : "foobar", "int" : 42, "foo" : "foo" }, { "bar" : "def", "int" : 42, "foo" : "foo" }, { "bar" : "foobar", "int" : 42, "foo" : "foo" })" :)
(:JIQS: ShouldRun; Output="({ "foo" : "foo", "bar" : "foobar", "int" : 1 }, { "foo" : "foo", "bar" : "def", "int" : 42 }, { "foo" : "foo", "bar" : "foobar", "int" : 42 }, { "foo" : "foo", "bar" : "def", "int" : 42 }, { "bar" : "foobar", "foo" : "foo", "int" : 42 }, { "foo" : "foo", "bar" : "def", "int" : 42 }, { "bar" : "def", "int" : 42, "foo" : "foo" }, { "bar" : "foobar", "int" : 42, "foo" : "foo" }, { "bar" : "def", "int" : 42, "foo" : "foo" }, { "bar" : "foobar", "int" : 42, "foo" : "foo" }, { "bar" : "foobar", "int" : 42, "foo" : "foo" })" :)
declare type local:a as { "!foo" : "string", "bar" : "string=def", "int" : "integer=42" };
declare type local:b as { "foo" : "string=def" };
declare type local:c as { "foo" : "string" };
Expand All @@ -11,6 +11,6 @@ validate type local:a { validate type local:a* { { "foo" : "foo", "bar" : "fooba
validate type local:a { validate type local:c* { { "foo" : "foo" } } },

validate type local:a* { { "foo" : "foo" }, { "foo" : "foo", "bar" : "foobar" } },
validate type local:a* { parallelize(({ "foo" : "foo" }, { "foo" : "foo", "bar" : "foobar" })) }(:,
validate type local:a* { validate type local:a* { { "foo" : "foo", "bar" : "foobar" } } },
validate type local:a* { parallelize(({ "foo" : "foo" }, { "foo" : "foo", "bar" : "foobar" })) },
validate type local:a* { validate type local:a* { { "foo" : "foo", "bar" : "foobar" } } }(:,
validate type local:a* { validate type local:c* { { "foo" : "foo" } } }:)