Stage data adds duplicate key

[jmgilman]

During the migration, stage-specific data is added. Using the dev cluster as an example, the SQL files that get imported are here. The test fund event that is added in this section uses a row_id of 0 which conflicts with the historical data that was added in a previous step.

The result is the migration fails with:

Adding stage specific data from ./stage_data/dev/00001_testfund_event.sql
psql:./stage_data/dev/00001_testfund_event.sql:54: ERROR:  duplicate key value violates unique constraint "event_pkey"

We should give the test event a very high row ID (i.e., 9999). If we do a UPSERT, it will end up overwriting some of the historical data, and the API will return incorrect data.

[kukkok3]

@jmgilman could this be due to my PR with the fund 11 data that was not merged yet when you did the migration?

[kukkok3]

Pr with fund11 entry #612
right now this is failing for what I think is a bug in this earthfile where stage data are copied even when data=test. Not sure how to handle this, I think I'll move coping the stage data with the historic data. A ARG array type would be perfect to manage this but it is not supported (yet earthly/earthly#1890) by earthly

[jmgilman]

There's still an issue here, as at the moment when the migration is run on a eventdb with existing data, it fails because of duplicate row IDs. The SQL script needs to be modified to handle this situation.

[FelipeRosa]

Fixed that in #624