Avoiding schemaName from generateChangeLog

General Discussion
1

Is there a way to avoid the schemaName to be output to the database changelog file when running generateChangeLog.

Why I am asking this is that we have an existing database that we want to retrieve a master changelog file from. We then want to use this changelog file for other database schemas depending what we want to do. Those database schemas have different schema names.

When applying databasechangelog files to a database, does the schemaName attribute have any meaning?

The question also applies to the baseTableSchemaName attribute on the addForeignKeyConstraint. What is this attribute used for? Does it default to the schema given as the --username parameter if not given in the changeSet?

Best regards /Lasse

2

There isn’t currently.  I think we used to not include it, but then people that needed to target the schema and/or run multiple schemas had probles. 

The easiest approach for right now would be to write an XML parser that would read in the generated changelog and strip them out. 

If you wanted to create an issue at liquibase.jira.com, I’ll look into it more for the 2.1 release.

Nathan

3

I have the same problem. Is there any way to configure to include or not schemaName in changelog?