[
https://issues.apache.org/jira/browse/IGNITE-10195?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexey Platonov updated IGNITE-10195:
-------------------------------------
Description:
Cannot create caches with different names but with same indexed types and
schema name. For example, such code will throw exception
"javax.cache.CacheException: Table already exists: PERSON".
{code:java}
node.createCache(new CacheConfiguration<PersonKey, Person>()
.setName("PERSON_1")
.setIndexedTypes(Key.class, Person.class)
.setSqlSchema(QueryUtils.DFLT_SCHEMA));
node.createCache(new CacheConfiguration<PersonKey, Person>()
.setName("PERSON_2")
.setIndexedTypes(Key.class, Person.class)
.setSqlSchema(QueryUtils.DFLT_SCHEMA));
{code}
If I set table name manually by setQueryEntities(...) then
"javax.cache.CacheException: Index already exists: PERSON_ORGID_IDX" wil be
thrown (Value has field with "origId" and annotation @QuerySqlField(index =
true)). Here is definition of Person class:
{code:java}
public static class PersonKey {
@QuerySqlField
public long id;
/**
* Constructor.
*
* @param id ID.
*/
PersonKey(long id) {
this.id = id;
}
/** {@inheritDoc} */
@Override public int hashCode() {
return (int)id;
}
/** {@inheritDoc} */
@Override public boolean equals(Object obj) {
return obj != null && obj instanceof PersonKey && (F.eq(id,
((PersonKey)obj).id));
}
}{code}
Such behavior seems to be usability bug. Why I cannot create two caches with
different names but with same indexed values?
was:
Cannot create caches with different names but with same indexed types and
schema name. For example, such code will throw exception
"javax.cache.CacheException: Table already exists: PERSON".
{code:java}
node.createCache(new CacheConfiguration<PersonKey, Person>()
.setName("PERSON_1")
.setIndexedTypes(Key.class, Person.class)
.setSqlSchema(QueryUtils.DFLT_SCHEMA));
node.createCache(new CacheConfiguration<PersonKey, Person>()
.setName("PERSON_2")
.setIndexedTypes(Key.class, Person.class)
.setSqlSchema(QueryUtils.DFLT_SCHEMA));
{code}
If I set table name manually by setQueryEntities(...) then
"javax.cache.CacheException: Index already exists: PERSON_ORGID_IDX" wil be
thrown (Value has field with "origId" and annotation @QuerySqlField(index =
true)). Here is definition of Person class:
{code:java}
public static class PersonKey {
@QuerySqlField public long id; /** * Constructor. * * @param id ID. */
PersonKey(long id) {
this.id = id; }
/** {@inheritDoc{color:#629755}} */
@Override public int hashCode() {
return (int)id; }
/** {@inheritDoc{color:#629755}} */
@Override public boolean equals(Object obj) {
return obj != null && obj instanceof PersonKey && (F.eq(id,
((PersonKey)obj).id)); }
}{code}
Such behavior seems to be usability bug. Why I cannot create two caches with
different names but with same indexed values?
> Cannot create caches with different names but with same indexed types and
> schema name
> -------------------------------------------------------------------------------------
>
> Key: IGNITE-10195
> URL: https://issues.apache.org/jira/browse/IGNITE-10195
> Project: Ignite
> Issue Type: Improvement
> Reporter: Alexey Platonov
> Priority: Major
>
> Cannot create caches with different names but with same indexed types and
> schema name. For example, such code will throw exception
> "javax.cache.CacheException: Table already exists: PERSON".
>
> {code:java}
> node.createCache(new CacheConfiguration<PersonKey, Person>()
> .setName("PERSON_1")
> .setIndexedTypes(Key.class, Person.class)
> .setSqlSchema(QueryUtils.DFLT_SCHEMA));
> node.createCache(new CacheConfiguration<PersonKey, Person>()
> .setName("PERSON_2")
> .setIndexedTypes(Key.class, Person.class)
> .setSqlSchema(QueryUtils.DFLT_SCHEMA));
> {code}
>
> If I set table name manually by setQueryEntities(...) then
> "javax.cache.CacheException: Index already exists: PERSON_ORGID_IDX" wil be
> thrown (Value has field with "origId" and annotation @QuerySqlField(index =
> true)). Here is definition of Person class:
>
> {code:java}
> public static class PersonKey {
> @QuerySqlField
> public long id;
> /**
> * Constructor.
> *
> * @param id ID.
> */
> PersonKey(long id) {
> this.id = id;
> }
> /** {@inheritDoc} */
> @Override public int hashCode() {
> return (int)id;
> }
> /** {@inheritDoc} */
> @Override public boolean equals(Object obj) {
> return obj != null && obj instanceof PersonKey && (F.eq(id,
> ((PersonKey)obj).id));
> }
> }{code}
> Such behavior seems to be usability bug. Why I cannot create two caches with
> different names but with same indexed values?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
