Merge pull request #495 from fishtown-analytics/update/adapter-respon…

…se-dict

Add note about adapter_response in run_results.json, sources.json

website/docs/reference/artifacts/run-results-json.md

@@ -31,5 +31,5 @@ Each entry in `results` is a dictionary with the following keys:
 - `thread_id`: Which thread executed this node? E.g. `Thread-1`
 - `execution_time`: Total time spent executing this node
 - `timing`: Array that breaks down execution time into steps (often `compile` + `execute`)
-- `adapter_response`: Dictionary of information returned from the database, which varies by adapter. E.g. success `code`, number of `rows_affected`, total `bytes_processed`, etc.
+- `adapter_response`: Dictionary of information returned from the database, which varies by adapter. E.g. success `code`, number of `rows_affected`, total `bytes_processed`, etc. Not populated by tests, as of v0.19.0; we plan to fix in a future release ([dbt#2580](https://github.com/fishtown-analytics/dbt/issues/2580)).
 - `message`: How dbt will report this result on the CLI, based on information returned from the database

website/docs/reference/artifacts/sources-json.md

@@ -22,3 +22,4 @@ Each entry in `results` is a dictionary with the following keys:
 - `max_loaded_at_time_ago_in_s`: Interval between `max_loaded_at` and `snapshotted_at`, calculated in python to handle timezone complexity.
 - `criteria`: The freshness threshold(s) for this source, defined in the project.
 - `status`: The freshness status of this source, based on `max_loaded_at_time_ago_in_s` + `criteria`, reported on the CLI. One of `pass`, `warn`, or `error` if the query succeeds, `runtime error` if the query fails.
+- `adapter_response`: Dictionary of information returned from the database, which varies by adapter. Not populated by source freshness checks, as of v0.19.0; we plan to fix in a future release ([dbt#2580](https://github.com/fishtown-analytics/dbt/issues/2580)).