Skip to content
pipetreepipetree

With problems

Terminal window
uv run python examples/with_problems/run.py --html graph.html

Same pipeline as the quickstart, two runs. Run 1 loads clean data. Run 2 loads a second delivery with four problems. Most tools let three of them pass silently. pipetree puts a note on the table.

The graph after run 2: three tables with a warning badge, gold.dim_customer failed, gold.fact_sales upstream_failed

Problem Table Note
Account 2 arrives three times in one batch (versions 1, 2 and 3) bronze.customer 2 duplicate row(s) dropped. The row with the highest sequence_by wins.
A row without employee_id bronze.employee 1 NULL-key row(s) dropped
A new channel column bronze.orders schema: added: channel (string). With schema_policy: evolve the column is added.
A column that does not exist gold.dim_customer failed. gold.fact_sales is upstream_failed.

A table with notes gets an amber badge in the graph and notes: ... in the digest. Its status does not change. The last row is a real failure: the run exits with 1.

Details: Notes and run log.