Ricardo J. Méndez

Hi Jonathan, It doesn't. From an initial inspection it seems that what's passed to the compile function you modified on postgresql.clj is the DataTypeClause contained in the :data-type section of...

For what it's worth, there _are_ tests for alter table in postgresql and they are currently failing: ERROR in (test-alter-table-postgresql) (QueryExecutorImpl.java:2198) Uncaught exception, not in assertion. expected: nil actual: org.postgresql.util.PSQLException:...

I don't believe I saw the second one... I'll try it later today. Cheers!

Hi Jonathan, No time to test yet - busy on an unrelated client project - but I'll give it a shot in a few days and report back.

Confirmed, #75 fixes it.

Yes, just the commands would be fine. It would save a lot of set up on repeated manual experiments, like requires and initial defs.

> I figured it out, If I return a vector, the first entry being the boolean result of the decision and the second entry being the :representation map, then it...

Thanks @graue, I was indeed wondering if a future release would end up breaking this workaround. I can live with returning the content-type.

> or if someone else wants to come along and do the work. I'd say that's a reasonable expectation if it's an issue, but for a PR, either it's merged...

I am now looking into node more for Macchiato. What's the status of this PR?