Databook
activeopen governmentAn award-winning transparency tool that combines over 50 official NYC open datasets — agencies, people, jobs, capital projects, schools, districts and procurement — into one searchable picture of how city government works.
Adds planning documentation for bringing the building lookup to Who-Owns-What parity and creating a dedicated landlord portfolio view. Includes a phased implementation plan, name-normalization specifications, and SQL/endpoint specifications for Phase 1 API additions without making any code or data changes.
Added zoned-school lookups, accountability chips, and ownership sections to building profiles. Integrated a landlord portfolio view and updated documentation with Phase 1 and 2 API specifications, including Who-Owns-What parity plans and portfolio endpoint details.
This update fixes a bug where leading and trailing whitespace in CSV header names propagated to the API and broke the frontend. A new helper function now trims outer whitespace from column names during both network and local CSV ingestion.
Updated the CSV ingestion process to trim leading and trailing whitespace from column names. Fixed a bug in the match-saving process by deduping in-batch source texts to prevent duplicate database entries.
This pull request fixes a pipeline crash caused by duplicate source texts within the same batch violating a database unique constraint. It resolves the issue by deduplicating source texts in the save_matches function before the upsert loop.
It also corrects the DuckDB data path from a stale container directory to the active host directory.