[jira] [Commented] (FLINK-7452) Add helper methods for all built-in Flink types to Types

["ASF GitHub Bot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-7452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16167868#comment-16167868 ] 

ASF GitHub Bot commented on FLINK-7452:
---------------------------------------

Github user fhueske commented on a diff in the pull request:

    https://github.com/apache/flink/pull/4612#discussion_r138919667
  
    --- Diff: flink-core/src/main/java/org/apache/flink/api/common/typeinfo/Types.java ---
    @@ -19,56 +19,408 @@
     package org.apache.flink.api.common.typeinfo;
     
     import org.apache.flink.annotation.PublicEvolving;
    +import org.apache.flink.api.common.functions.InvalidTypesException;
    +import org.apache.flink.api.java.typeutils.EitherTypeInfo;
    +import org.apache.flink.api.java.typeutils.EnumTypeInfo;
    +import org.apache.flink.api.java.typeutils.GenericTypeInfo;
    +import org.apache.flink.api.java.typeutils.ListTypeInfo;
    +import org.apache.flink.api.java.typeutils.MapTypeInfo;
    +import org.apache.flink.api.java.typeutils.ObjectArrayTypeInfo;
    +import org.apache.flink.api.java.typeutils.PojoField;
    +import org.apache.flink.api.java.typeutils.PojoTypeInfo;
     import org.apache.flink.api.java.typeutils.RowTypeInfo;
    +import org.apache.flink.api.java.typeutils.TupleTypeInfo;
    +import org.apache.flink.api.java.typeutils.TypeExtractor;
    +import org.apache.flink.api.java.typeutils.ValueTypeInfo;
     
    -import java.math.BigDecimal;
    -import java.sql.Date;
    -import java.sql.Time;
    -import java.sql.Timestamp;
    +import java.lang.reflect.Field;
    +import java.util.ArrayList;
    +import java.util.List;
    +import java.util.Map;
     
     /**
    - * This class gives access to the type information of the most most common types.
    + * This class gives access to the type information of the most common types for which Flink
    + * has built-in serializers and comparators.
    + *
    + * <p>In many cases, Flink tries to analyze generic signatures of functions to determine return
    + * types automatically. This class is intended for cases where the extraction is not possible
    + * (or inefficient) as well as cases where type information has to be supplied manually.
    + *
    + * <p>Depending on the API you are using (e.g. Scala API or Table API), there might be a more
    --- End diff --
    
    How about `Please note that the Scala API and Table API provide more specialized Types classes`, maybe linking to (or providing the full name) of these classes?


> Add helper methods for all built-in Flink types to Types
> --------------------------------------------------------
>
>                 Key: FLINK-7452
>                 URL: https://issues.apache.org/jira/browse/FLINK-7452
>             Project: Flink
>          Issue Type: Improvement
>          Components: Type Serialization System
>            Reporter: Timo Walther
>            Assignee: Timo Walther
>
> Sometimes it is very difficult to provide `TypeInformation` manually, in case some extraction fails or is not available. {{TypeHint}}s should be the preferred way but this methods can ensure correct types.
> I propose to add all built-in Flink types to the {{Types}}. Such as:
> {code}
> Types.POJO(MyPojo.class)
> Types.POJO(Map<String, TypeInformation>)
> Types.GENERIC(Object.class)
> Types.TUPLE(TypeInformation, ...)
> Types.MAP(TypeInformation, TypeInformation)
> {code}
> The methods should validate that the returned type is exactly the requested type. And especially in case of POJO should help creating  {{PojoTypeInfo}}.
> Once this is in place, we can deprecate the {{TypeInfoParser}}.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)