Tabular Operations

Tabular operations transform non-spatial table data: filtering, selecting, sorting, joining, combining, and aggregating. All operations are powered by DuckDB.

Part of the tabular domain.

Operations

Operation	Type	Description	Key Params
tabular_filter	table → table	Filter rows using a SQL WHERE expression.	where (string, required)
tabular_select	table → table	Select, rename, or compute columns. Pick a subset, exclude columns, or add computed expressions.	columns (array), exclude (array), expressions (object)
tabular_sort	table → table	Sort rows by one or more columns. Optionally limit to top-N.	by (array, required), limit (int)
tabular_join	table + table → table	Join two tables on matching columns. Supports inner, left, right, outer, and cross joins.	on, left_on/right_on, how
tabular_union	table[] → table	Stack rows from multiple tables vertically. All tables MUST have compatible schemas.	distinct (bool)
tabular_aggregate	table → table	Group rows and compute aggregate statistics using SQL functions.	group_by (array, required), aggregations (object, required), having, order_by, limit

Examples

Filter

layers:
  analysis/active-users:
    type: table
    compute:
      op: tabular_filter
      inputs:
        table: { layer: source/users }
      params:
        where: "status = 'active' AND last_login > '2024-01-01'"

Select and Compute Columns

layers:
  analysis/enriched:
    type: table
    compute:
      op: tabular_select
      inputs:
        table: { layer: source/parcels }
      params:
        columns: [parcel_id, area_sqft, zone_code]
        expressions:
          area_acres: "area_sqft / 43560.0"
          zone_label: "CASE WHEN zone_code = 'R' THEN 'Residential' ELSE 'Commercial' END"

Sort (Top-N)

layers:
  results/top-emitters:
    type: table
    compute:
      op: tabular_sort
      inputs:
        table: { layer: source/facilities }
      params:
        by: ["emissions DESC"]
        limit: 10
    style:
      table:
        columns:
          - { field: name, label: Facility }
          - { field: emissions, label: "CO2 (tons)", format: ",.0f" }

Join

layers:
  analysis/orders-with-customers:
    type: table
    compute:
      op: tabular_join
      inputs:
        left: { layer: source/orders }
        right: { layer: source/customers }
      params:
        on: customer_id
        how: left
        suffix: "_cust"

Union

layers:
  combined/all-observations:
    type: table
    compute:
      op: tabular_union
      inputs:
        tables:
          - { layer: source/observations-2023 }
          - { layer: source/observations-2024 }
      params:
        distinct: false

Aggregate

layers:
  results/sales-summary:
    type: table
    compute:
      op: tabular_aggregate
      inputs:
        table: { layer: source/transactions }
      params:
        group_by: [region, quarter]
        aggregations:
          total_revenue: "SUM(amount)"
          avg_order: "AVG(amount)"
          num_orders: "COUNT(*)"
        having: "SUM(amount) > 10000"
        order_by: ["total_revenue DESC"]
        limit: 20
    style:
      table:
        columns:
          - { field: region, label: Region }
          - { field: quarter, label: Quarter }
          - { field: total_revenue, label: Revenue, format: "$,.2f" }