I need to execute a subquery in a select clause with Apache Openjpa 2.
Does JPA support subqueries in SELECT clause?
My Query is something like this:
SELECT t.date, t.value,
(SELECT COUNT(DISTINCT t2.value2) FROM table t2 WHERE t2.date = t.date)
FROM table t
WHERE ...
When I execute my query, I get a class cast exception:
Exception in thread "main" <openjpa-2.1.1-SNAPSHOT-r422266:1141200 nonfatal user error> org.apache.openjpa.persistence.ArgumentException:
at org.apache.openjpa.kernel.QueryImpl.execute(QueryImpl.java:872)
at org.apache.openjpa.kernel.QueryImpl.execute(QueryImpl.java:794)
at org.apache.openjpa.kernel.DelegatingQuery.execute(DelegatingQuery.java:542)
at org.apache.openjpa.persistence.QueryImpl.execute(QueryImpl.java:315)
at org.apache.openjpa.persistence.QueryImpl.getResultList(QueryImpl.java:331)
Caused by: java.lang.ClassCastException: org.apache.openjpa.jdbc.sql.LogicalUnion$UnionSelect incompatible with org.apache.openjpa.jdbc.sql.SelectImpl
at org.apache.openjpa.jdbc.sql.SelectImpl.setParent(SelectImpl.java:579)
at org.apache.openjpa.jdbc.kernel.exps.SelectConstructor.newSelect(SelectConstructor.java:147)
at org.apache.openjpa.jdbc.kernel.exps.SelectConstructor.evaluate(SelectConstructor.java:87)
at org.apache.openjpa.jdbc.kernel.exps.SubQ.appendTo(SubQ.java:209)
at org.apache.openjpa.jdbc.kernel.exps.SubQ.appendTo(SubQ.java:203)
at org.apache.openjpa.jdbc.kernel.exps.SubQ.newSQLBuffer(SubQ.java:167)
at org.apache.openjpa.jdbc.kernel.exps.SubQ.selectColumns(SubQ.java:153)
at org.apache.openjpa.jdbc.kernel.exps.SubQ.select(SubQ.java:148)
at org.apache.openjpa.jdbc.kernel.exps.SelectConstructor.select(SelectConstructor.java:372)
at org.apache.openjpa.jdbc.kernel.JDBCStoreQuery.populateSelect(JDBCStoreQuery.java:295)
at org.apache.openjpa.jdbc.kernel.JDBCStoreQuery.access$100(JDBCStoreQuery.java:86)
at org.apache.openjpa.jdbc.kernel.JDBCStoreQuery$1.select(JDBCStoreQuery.java:267)
at org.apache.openjpa.jdbc.sql.LogicalUnion.select(LogicalUnion.java:297)
at org.apache.openjpa.jdbc.kernel.JDBCStoreQuery.populateUnion(JDBCStoreQuery.java:265)
at org.apache.openjpa.jdbc.kernel.JDBCStoreQuery.executeQuery(JDBCStoreQuery.java:211)
at org.apache.openjpa.kernel.ExpressionStoreQuery$DataStoreExecutor.executeQuery(ExpressionStoreQuery.java:782)
at org.apache.openjpa.datacache.QueryCacheStoreQuery$QueryCacheExecutor.executeQuery(QueryCacheStoreQuery.java:346)
at org.apache.openjpa.kernel.QueryImpl.execute(QueryImpl.java:1005)
at org.apache.openjpa.kernel.QueryImpl.execute(QueryImpl.java:863)
... 6 more
Is this possible or do I have to use NativeQuery / Single Queries?
No, it is not supported to use subqueries in SELECT clause. In JPA 2.0 specification this is told with following words:
Subqueries may be used in the WHERE or HAVING clauses.
Related
I am using Spark 3.0, in my Java program I am querying data from views which are in Oracle DB. I used the Java API JdbcRDD to query the views.
The problem I have is that the view doesn't contain any ID or timestamp columns. So, I am unable to construct my SQL query with lowerBound and upperBound values.
Please find below the example query I need to run in Spark. Here exp_stg.usr and exp_stg.prtcpnt are the two views exposed to me.
"SELECT a.participant,
a.desc,
b.firstname,
b.lastname,
b.dept,
b.telno,
b.emailaddr
FROM usr_stg.prtcpnt a
LEFT OUTER JOIN usr_stg.usr b
ON a.participant = b.participant
WHERE a.class = 'SpSession' "
I tried using temp tables in spark and join, but the query performance is bad as there are around ~13,000,000 rows in each view. Hence I tried to use the join query in the Oracle DB.
I was able to overcome the constraint using ROWNUM in the query. Using ROWNUM as the lowerBound and upperBound I am now able to get the data using JdbcRDD.
`
SELECT ROWNUM as id, a.participant,
a.desc,
b.firstname,
b.lastname,
b.dept,
b.telno,
b.emailaddr
FROM usr_stg.prtcpnt a
LEFT OUTER JOIN usr_stg.usr b
ON a.participant = b.participant
WHERE a.class = 'SpSession' and ?<=ROWNUM and ROWNUM<=?"`
I'm trying to execute the test query like this:
SELECT COUNT(CASE WHEN name IN (SELECT name FROM requiredProducts) THEN name END)
FROM myProducts
which throws the following exception:
java.lang.ClassCastException:
org.apache.spark.sql.execution.datasources.LogicalRelation cannot be cast to
org.apache.spark.sql.execution.SparkPlan
I have a suggestion that IN operator can not be used in CASE WHEN. Is it really so? Spark documentation is silent about this.
The IN operator using a subquery does not work in a projection regardless of whether it is contained in a CASE WHEN, it will only work in filters. It works fine if you specify values in the IN clause directly rather than using a subquery.
I am not sure how to generate the exact exception you got above, but when I attempt to run a similar query in Spark Scala, it returns a more descriptive error:
org.apache.spark.sql.AnalysisException: IN/EXISTS predicate sub-queries can only be used in a Filter: Project [CASE WHEN agi_label#5 IN (list#96 []) THEN 1 ELSE 0 END AS CASE WHEN (agi_label IN (listquery())) THEN 1 ELSE 0 END#97]
I have run into this issue in the past. Your best bet is probably to restructure it to use a left join to requiredProducts and then check for a null in the case statement. For example, something like this might work:
SELECT COUNT(CASE WHEN rp.name is not null THEN mp.name END)
FROM myProducts mp
LEFT JOIN requiredProducts rp ON mp.name = rp.name
I am trying to query the Cassandra database using SparkSQL terminal.
Query:
select * from keyspace.tablename
where user_id = e3a119e0-8744-11e5-a557-e789fe3b4cc1;
Error: java.lang.RuntimeException: [1.88] failure: ``union'' expected but identifier e5 found
Also tried:
user_id= UUID.fromString(\`e3a119e0-8744-11e5-a557-e789fe3b4cc1\`)")
user_id= \'e3a119e0-8744-11e5-a557-e789fe3b4cc1\'")
token(user_id)= token(`e3a119e0-8744-11e5-a557-e789fe3b4cc1`)
I am not sure how can I query data on timeuuid.
TimeUUIDs are not supported as a type in SparkSQL so you are only allowed to do direct string comparisons. Represent the TIMEUUID as a string
select * from keyspace.tablename where user_id = "e3a119e0-8744-11e5-a557-e789fe3b4cc1"
I'm trying to get an average for a count on a groupBy by joining with a subquery. Don't know if that the right way to go at all but I couldn't anything about subqueries other than the mysema doc.
Scenario:
How many orders per product did a customer do on average?
Meaning: A Customer orders products. So a customer ordered a specific product a number of times (count). What's the average number of orders that customer placed for any product?
Might sound a bit hypothetical, in fact it's just part of a prototype, but it made me wonder, how to get a reference to a custom column created within a subquery with the fancy QueryDSL from Mysema.
In SQL you just give the count column an alias and join using a second ID column. QueryDSL has the "as()" method as well but I have no Idea, how to retrieve that column plus I dont't see how it can join one query with anothers, since query.list() just gets a list but for some reason the join accepts it. Feels wrong...
Here's my code:
JPQLQuery query = createJPQLQuery();
QOrdering qOrdering = QOrdering.ordering;
QProduct qProduct = QProduct.product;
QCustomer qCustomer = QCustomer.customer;
// how many of each product did a customer order?
HibernateSubQuery subQuery = new HibernateSubQuery();
subQuery.from(qOrdering).innerJoin(qOrdering.product,qProduct).innerJoin(qOrdering.customer, qCustomer);
subQuery.groupBy(qCustomer,qProduct).list(qCustomer.id,qProduct.id,qProduct.count());
// get the average number of orders per product for each customer
query.from(qCustomer);
query.innerJoin(subQuery.list(qCustomer.id,qOrdering.count().as("count_orders")));
query.groupBy(qCustomer.id);
return (List<Object[]>) query.list(qCustomer.firstname,subQuery.count_orders.avg());
Again: How do I join with a subquery?
How do I get the aliased "count" column to do more aggregation like avg (is my group right btw?)
Might be that I have some other errors in this, so any help appreciated!
Thanks!
Edit:
That's kind of the native SQL I'd like to see QueryDSL produce:
Select avg(numOrders) as average, cust.lastname from
customer cust
inner join
(select count(o.product_id) as numOrders, c.id as cid, p.name
from ordering o
inner join product p on o.product_id=p.id
inner join customer c on o.customer_id=c.id
group by o.customer_id, o.product_id) as numprods
on cust.id = numprods.cid
group by numprods.cid
order by cust.lastname;
Using subqueries in the join clause is not allowed. in JPQL, subqueries are only allowed in the WHERE and HAVING part. The join method signatures in Querydsl JPA queries are too wide.
As this query needs two levels of grouping, maybe it can't be expressed with JPQL / Querydsl JPA.
I'd suggest to write this query using the Querydsl JPA Native query support.
As Querydsl JPA uses JPQL internally, it is restricted by the expressiveness of JPQL.
I know that this question is old and already has an accepted answer, but judging from this question, it seems to still be troubling guys. See my answer in the same question. The use of JoinFlag in the join() section and Expression.path() is able to achieve left-joining a subquery. Hope this helps someone.
QueryDsl does not support subQuery in join but you can achieve this via following way:
We wanted to achieve the following query:
select A.* from A join (select aid from B group by aid) b on b.aid=A.id;
Map a View or SQL query to JPA entity:
import lombok.Setter;
import org.hibernate.annotations.Subselect;
import org.hibernate.annotations.Synchronize;
import javax.persistence.Entity;
import javax.persistence.Id;
#Entity
#Getter
#Setter
#Subselect("select aid from B group by aid")
#Synchronize("B")
public class BGroupByAid {
#Id
private Integer aId;
}
then use the equivalent QueryDSl entity in the class just like the regular entity:
JPAQuery<QAsset> query = new JPAQuery<>(entityManager);
QBGroupByAid bGroupById = QBGroupByAid.bGroupByAid;
List<A> tupleOfAssets =
query.select(A)
.from(A).innerJoin(bGroupById).on(bGroupById.aId.eq(A.aId))
.fetchResults()
.getResults();
You can also use blazebit which supports also subquery in join. I have try it and it is working. You can create SubQueryExpression f.e like this
SubQueryExpression<Tuple> sp2 = getQueryFactory().select(entity.id,
JPQLNextExpressions.rowNumber().over().partitionBy(entity.folId).orderBy(entity.creationDate.desc()).as(rowNumber))
.from(entity)
.where(Expressions.path(Integer.class, rowNumber).eq(1));
and then just join it like this:
return getBlazeQueryFactory()
.select(entity1, entity)
.from(entity1)
.leftJoin(sp2, entity).on(entity.id.eq(entity1.id)).fetch();
I have put here just simple example. So maybe it doesn't make a perfect sense but maybe can be helpful.
Also don't be confused it will can produce union in the generated select. This is just for naming columns from subquery you can read better explanation about this union here: Blaze-Persistence GROUP BY in LEFT JOIN SUBQUERY with COALESCE in root query
Is it possible to do a join to a nested select statement in SubSonic 2.2? I am having trouble with this because I have to do a group by...
I.E.
select conn_company.company_name, SUM(t.quote_grand_total) as 'SUM' from conn_company
inner join
(SELECT *
FROM [dbo].[QuoteDBAll]
where [dbo].[QuoteDBAll].[quote_status_id] = 1
AND [dbo].[QuoteDBAll].[quote_ca_modified_date] BETWEEN '12/1/2000' AND '12/1/2009'
) t
on conn_company.company_id = t.company_id
group by company_name
having company_name = 'JACK'
Unfortunately you're not going to be able to convert that SQL into SubSonic 2 syntax. I would suggest you create a sql view based on your query and get SubSonic to generate a model from that instead.