[
https://issues.apache.org/jira/browse/DRILL-6361?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16461349#comment-16461349
]
ASF GitHub Bot commented on DRILL-6361:
---------------------------------------
paul-rogers commented on a change in pull request #1242: DRILL-6361: Revised
typeOf() function versions
URL: https://github.com/apache/drill/pull/1242#discussion_r185576220
##########
File path:
exec/java-exec/src/main/java/org/apache/drill/exec/expr/fn/impl/UnionFunctions.java
##########
@@ -146,16 +139,131 @@ private static int getTypeValue(MinorType type) {
@Inject
DrillBuf buf;
+ @Override
public void setup() {}
+ @Override
public void eval() {
- byte[] type;
+ String typeName;
if (input.isSet()) {
- type = input.getType().getMinorType().name().getBytes();
+ typeName = input.getType().getMinorType().name();
} else {
- type =
org.apache.drill.common.types.TypeProtos.MinorType.NULL.name().getBytes();
+ typeName =
org.apache.drill.common.types.TypeProtos.MinorType.NULL.name();
}
+ byte[] type = typeName.getBytes();
+ buf = buf.reallocIfNeeded(type.length);
+ buf.setBytes(0, type);
+ out.buffer = buf;
+ out.start = 0;
+ out.end = type.length;
+ }
+ }
+
+ @FunctionTemplate(name = "sqlTypeOf",
+ scope = FunctionTemplate.FunctionScope.SIMPLE,
+ nulls = NullHandling.INTERNAL)
+ public static class GetSqlType implements DrillSimpleFunc {
+
+ @Param
+ FieldReader input;
+ @Output
+ VarCharHolder out;
+ @Inject
+ DrillBuf buf;
+
+ @Override
+ public void setup() {}
+
+ @Override
+ public void eval() {
+
+ org.apache.drill.common.types.TypeProtos.MajorType type =
input.getType();
+
+ // Note: extendType is a static function because the byte code magic
+ // for UDFS can't handle switch statements.
+
+ String typeName =
+ org.apache.drill.exec.expr.fn.impl.UnionFunctions.extendType(
+ type,
+ org.apache.drill.common.types.Types.getBaseSqlTypeName(type));
+
+ org.apache.drill.exec.expr.fn.impl.StringFunctionHelpers.varCharOutput(
Review comment:
True. However, since this function is not performance critical, the
convenience of a single call vs. multiple lines of code, is a win.
Further, I believe experiments have been done (I did some myself) to show
that scalar replacement adds no value: The Java 8 JIT compiler does it
automatically so there is no need for Drill to do it itself.
That said, if you prefer me to inline the code, I will do so.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Provide sqlTypeOf() and modeOf() functions
> ------------------------------------------
>
> Key: DRILL-6361
> URL: https://issues.apache.org/jira/browse/DRILL-6361
> Project: Apache Drill
> Issue Type: Improvement
> Affects Versions: 1.13.0
> Reporter: Paul Rogers
> Assignee: Paul Rogers
> Priority: Minor
> Labels: doc-impacting
> Fix For: 1.14.0
>
>
> Drill provides a {{typeof()}} function to return the type of a column. The
> returned string, however, has only the base data type. A Drill data type (a
> "major type") also includes a cardinality (a "mode"). For example, {{Optional
> Int}} or {{Required VarChar}}.
> This type information is useful for handling data conversions. For example,
> if I could tell that a column value was a {{Nullable Int}}, I could guess
> that it is one Drill invented, and I could merge it, by hand, with the type
> from another file that had actual values.
> The two options are equivalent. Either provide a {{modeOf()}} to just return
> cardinality, or a {{dataTypeOf()}} that returns both. (Maybe the {{modeOf()}}
> might be more useful.)
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
