Re: Problem with schema reference not getting pickedup in raw sql

Sorry for the slow dely.  It got behind on the –devel list and missed that I needed to reply to a couple emails.

The code is currently not checking for parameters inside strings, so you’ll have to use the workaround for now.  I’ll add it as a feature that should be implemented after/during a refactoring I a starting on this week., so it should be an option in 1.10.

Nathan

Hey all,

We are using postgres with the postgis spatial extensions.  We have 2 schemas within a database that we keep in sync (one is a test schema and one is production).  We have specified the defaultSchemaName in our pom.xml for doing the migrate, however the raw sql does not insert the schema name into the sql to be executed.  As a result, if we are creating a spatial index (using GIST), it fails on the second pass as it is trying to create it on the first schema not the second.

Our changeset looks like this:

      <changeSet id="15" author="MDG">

            <comment>Adding index to upper_tier table</comment>

            <sql>create index upper_tier_boundary_idx on upper_tier using gist (boundary GIST_GEOMETRY_OPS)</sql>

      </changeSet>

We have noted that we can work around it by doing the following:

      <changeSet id="15" author="MDG">

            <comment>Adding index to upper_tier table</comment>

            <createIndex indexName="upper_tier_boundary_idx" tableName="upper_tier using gist" >

                  <column name="boundary GIST_GEOMETRY_OPS"/>

            </createIndex>

      </changeSet>

My question is, is there a way to refer to the schema in the xml file for cases like these?  Ie. using a ${defaultSchemaName} within the xml itself?

-mike

“I don't think that the right time to tell a dog not to pee on the carpet is semiannually.”

House MD on performance reviews