Segments

Segments are named filters you can reuse across queries.

Segments are named filters you can apply in queries without repeating filter SQL everywhere.

Runnable Example (Executed in CI)

models:
  - name: orders
    sql: |
      select * from (values
        (1, 120.00, 'completed'),
        (2, 80.00,  'completed'),
        (3, 50.00,  'pending'),
        (4, 200.00, 'completed')
      ) as t(id, amount, status)
    primary_key: id
    dimensions:
      - name: status
        type: categorical
        sql: status
    metrics:
      - name: revenue
        agg: sum
        sql: amount
    segments:
      - name: completed
        sql: "{model}.status = 'completed'"

from pathlib import Path
from sidemantic import SemanticLayer
from sidemantic.loaders import load_from_directory

layer = SemanticLayer()
load_from_directory(layer, Path("./models"))

res = layer.query(metrics=["orders.revenue"], segments=["orders.completed"])
rows = res.fetchall()
print(f"{rows[0][0]:.2f}")

400.00

YAML

models:
  - name: orders
    segments:
      - name: completed
        sql: "{model}.status = 'completed'"

      - name: high_value
        sql: "{model}.amount >= 500"

Python

from sidemantic import Segment

completed = Segment(name="completed", sql="{model}.status = 'completed'")

Using Segments

Programmatic query:

result = layer.query(
    metrics=["orders.revenue"],
    dimensions=["orders.status"],
    segments=["orders.completed"]
)

Gotchas

• Segments are filters; they don’t change grouping. Grouping is derived from dimensions.