Separating test and release reporting

kmelby

Is the only way to be able to test changes to reports without making any changes to the reports the end-users are already using by having two iServer instances (one for "release" and one for "test")?

When I say test I mean a formal QA through the application, not a preview from the designer.

johnw

You can do this in either a single or multiple iServer strategy.

The way we have done it in the past is to have a promotion script that will migrate from a staging environment to production. This is done through a series of IDAPI calls. So you can have a source and target, and these can be either the same machine with different paths or 2 different instances of iServer.

The way I am partial to doing this now is with a Unit Test or Functional Test framework, and using a custom XML emitter that keeps the same structure as a formatted report output, it just strips the formatting. This way I can test the results and have the unit tests leverage the report engine API for testing the output. This should idealy be used in conjunction with a staging environment so that you are seperating your unit tests from your functional and acceptance tests.