polygene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paul Merlin <paulmer...@apache.org>
Subject Re: Sorting out issues with Indexing SQL
Date Tue, 04 Apr 2017 06:58:36 GMT
Niclas,

Niclas Hedhman a écrit :
> I have fixed one problem related to the introduction of Identity in
> AbstractSQLQuerying.
>
> Object value;
> if( qName.type().equals( HasIdentity.class.getName() ) )
> {
>     columnName = DBNames.ENTITY_TABLE_IDENTITY_COLUMN_NAME;
>     value = predicate.value().toString();
> }
> else
> {
>     columnName = DBNames.QNAME_TABLE_VALUE_COLUMN_NAME;
>     value = predicate.value();
> }
>
> I can't push that, since I currently have other breaking changes
> regarding better Structural Error reporting.

I just pushed that. It fixes one of the 10 failing tests.
Thanks!

> I have also concluded that there are missing insertions into qname_*
> tables. 
> When executing
> org.apache.polygene.test.indexing.AbstractQueryTest#script29, qname_2
> should contain something but doesn't. I suspect that the problem is
> around how references are handled. And that the problem is showing
> itself
> at org/apache/polygene/index/sql/support/skeletons/AbstractSQLIndexing.java:713
>
> if( !qName.type().equals( HasIdentity.class.getName() ) )
> so that no insertions happen for HasIdentity. BUT I am not certain at
> all...

This indeed looks suspicious ...

/Paul



Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message