Media
1 item
Description
One-click export of Databook table column metadata to CSV, with a multi-table collection. Databook Column Extractor exports column metadata from Databook dataset pages to CSV — directly from Databook's own GraphQL response, so the output matches what you see in the Definition tab. HOW IT WORKS 1. Open any Databook dataset page (e.g. databook.uberinternal.com/datasets/hive/dwh.fact_trip). 2. Click "Extract Table Metadata" (floating button, bottom-right). 3. The page reloads once, metadata is captured, and the CSV downloads automatically. Each file is named <table_name>_columns.csv and includes column name, type, description, nesting, and related metadata. COLLECTION (MULTI-TABLE EXPORT) Open the toolbar icon to manage a local collection: • Download each table's CSV as you go (toggle on/off). • Add each extraction to a collection (toggle on/off). • Download one combined CSV across all collected tables — same format as the consolidate_databook_columns.py script. • Remove individual tables or clear the collection at any time. The collection lives in your browser only. Nothing is uploaded to any server. FALLBACK If automatic capture is blocked, use "Paste GraphQL response" in the popup: copy the Response JSON from DevTools → Network → graphql, paste it, and download. PRIVACY & DATA • Runs only on databook.uberinternal.com. • All processing happens locally in your browser. • No data is sent to external servers. • CSV files are saved via Chrome's download API to your machine. NOTE ON PERMISSIONS The extension uses the debugger permission to read Databook's GraphQL network response — the only reliable way to capture metadata from Databook's bundled client. Chrome may briefly show a "started debugging this browser" bar during capture; it disappears when the download starts. For Uber internal use on Databook dataset pages.