Hey everyone! Thank you for all the answers here. It really helps a lot. I was trying to learn more about the discovery process on the website and here using search. But cannot understand if it's possible to skip the reflection of indexes. (I'm not even sure if that's a part of the discovery process) I tried to use

capabilities:[]

but it doesn't help. It still takes a minute to go through the same process.

04:14:22 [info     ] /.../tap-postgres/venv/lib/python3.9/site-packages/singer_sdk/connectors/sql.py:424: SAWarning: Skipped unsupported reflection of expression-based index mkb_hyphen_analysis_output_old_index cmd_type=elb consumer=False name=tap-postgres producer=True stdio=stderr string_id=tap-postgres
04:14:22 [info     ]   for index_def in inspected.get_indexes(table_name, schema=schema_name) cmd_type=elb consumer=False name=tap-postgres producer=True stdio=stderr string_id=tap-postgres
...
04:15:20 [info     ] /.../tap-postgres/venv/lib/python3.9/site-packages/singer_sdk/connectors/sql.py:424: SAWarning: Skipped unsupported reflection of expression-based index enterprise_twilio_credentials_account_sid_idx cmd_type=elb consumer=False name=tap-postgres producer=True stdio=stderr string_id=tap-postgres
04:15:20 [info     ]   for index_def in inspected.get_indexes(table_name, schema=schema_name) cmd_type=elb consumer=False name=tap-postgres producer=True stdio=stderr string_id=tap-postgres

Hi @mykola_zavada! I think that reflection is required to detect potential primary keys. Other than the sqlalchemy warning, are you running into issue caused by that behavior?

It just takes extra time every run and I was trying to understand if it's possible to avoid.