Github user patricker commented on the issue:
https://github.com/apache/nifi/pull/952
Paul is on vacation, but I worked with him on this.
After looking over the syntax I can see why this worked for us but not for
others. It looks like we used a syntax for Teradata that allows you to MERGE
single rows, `USING VALUES`, that isn't done the same way on other platforms.
We derived it from the standard MERGE syntax, but didn't notice when we
deviated from the standard.
For reference: the syntax for MS SQL Server seems to be
http://stackoverflow.com/a/5981300/328968.
I don't see any issues with moving this into a `DatabaseAdapter`, but it
would also make sense to me to have a default, maybe something closer to the MS
SQL one(?) that could be used if no `DatabaseAdapter` is found. I'm not sure
where it makes sense for that default implementation to live.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
