... start at 1. Tests break randomly under Postgres due to the expectations for IDs in the test.
FIX Do not use hardcoded IDs in tests, as records from fixture might …
…not start at 1
For what it's worth, this looks like a reasonable change to me. It's a pretty common thing we've run up against when running tests outside of MySQL.
ya makes sense too me thanks guys ;). To be honest I'm surprised the module works at all with Postgres lol.