Query rows with filters & projection

query_rows is the default way to read data: pick columns, filter, sort, and paginate. Reach for aggregate or execute_sql only when you need something it can't express.

Query

{
  "table": "eddytor.cfg_xxx.<uuid>_products",
  "columns": ["product_id", "name", "price"],
  "filter": "status = 'active' AND price > 50",
  "order_by": "price DESC",
  "limit": 20
}

eddytor query "SELECT product_id, name, price FROM \`eddytor\`.\`sales\`.\`orders\` WHERE status = 'active' ORDER BY price DESC" --limit 20

Filter & order syntax

• Filter - standard SQL WHERE: =, !=, LIKE, IN (...), IS NULL, BETWEEN.
• Order - "price DESC" or "category ASC, price DESC".

Habits that keep it fast

• Select only the columns you need - faster than fetching everything.
• Default to limit: 100 - raise it only when you need more.
• Push filters to the engine rather than filtering client-side.

Exploring unfamiliar data

1. get_table_schema - columns and types.
2. query_rows with limit: 10 - see sample rows.
3. profile_table - distributions, nulls, ranges.

Next

• Aggregate & summarize · Run arbitrary SQL · Profile a table.