[
https://issues.apache.org/jira/browse/CALCITE-3108?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16853916#comment-16853916
]
Muhammad Gelbana edited comment on CALCITE-3108 at 6/2/19 12:31 PM:
--------------------------------------------------------------------
My branches got mixed up and I added this feature to an unmurged PR that
supports the {{::}} operator. Please refer to the description for the updated
sample query.
I added {{TEXT}} as a keyword so the returned result set column type name would
be {{TEXT}} as returned from a PostgreSQL DB.
I don't usually like touching the parser so you're suggestion sounds more Java
oriented and I prefer that.
I managed to support the TEXT type by adding the type to the root schema:
{code:java}
Connection connection = DriverManager.getConnection("jdbc:calcite:", info);
connection.unwrap(CalciteConnection.class).getRootSchema().unwrap(CalciteSchema.class).add("TEXT",
new RelProtoDataType() {
@Override
public RelDataType apply(RelDataTypeFactory factory) {
return
factory.createTypeWithNullability(factory.createJavaType(String.class), false);
// return
factory.createTypeWithNullability(factory.createSqlType(SqlTypeName.VARCHAR),
false);
// return factory.createJavaType(String.class); // Failes:
Cannot be nullable
}
});
{code}
And it worked but the returned column type name is {{VARCHAR}}, not {{TEXT}} as
what would be returned form a real PostgreSQL. How could I direct Calcite to
use the type name I provided to the .add method instead of the concluded one
(ie. {{VARCHAR}}) ?
was (Author: mgelbana):
My branches got mixed up and I added this feature to an unmurged PR that
supports the {{::}} operator. Please refer to the description for the updated
sample query.
I added {{TEXT}} as a keyword so the returned result set column type name would
be {{TEXT}} as returned from a PostgreSQL DB.
I don't usually like touching the parser so you're suggestion sounds more Java
oriented and I prefer that.
I found I can do what you suggested by exposing that type through the exposed
schemas. But I can't find a way to do so for the root schema to get the example
query to work.
> Babel parser should parse the PostgreSQL TEXT type
> --------------------------------------------------
>
> Key: CALCITE-3108
> URL: https://issues.apache.org/jira/browse/CALCITE-3108
> Project: Calcite
> Issue Type: Bug
> Components: babel, core
> Affects Versions: 1.19.0
> Reporter: Muhammad Gelbana
> Assignee: Muhammad Gelbana
> Priority: Minor
> Labels: pull-request-available
> Fix For: 1.20.0
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Casting to PostgreSQL TEXT (ie. VARCHAR) isn't currently supported. The
> following query would fail in parsing and execution.
> {code:sql}SELECT CAST(EXPR$0 AS text) FROM (VALUES (1, 2, 3)){code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
