The Search API

Day 3 Lab

1. Build a Search API request: a query string plus a page_size.
2. Add a facet filter to scope by datasource (e.g. only Drive).
3. Run it and read each result's title, URL, and snippet from the typed response.
4. Confirm results respect the caller's permissions — same filtering as the UI.

Working Example: A Filtered Search Query

The Search API takes a query plus optional facet filters and returns typed results with snippets that show why each matched. Verified against api_clients.

import os
from glean import Glean

client = Glean(
    instance=os.environ["GLEAN_INSTANCE"],
    api_token=os.environ["GLEAN_API_TOKEN"],
    actas_user=os.environ["END_USER_EMAIL"],
)

response = client.client.search.query(
    query="incident runbook",
    page_size=10,
    request_options={
        "facet_filters": [
            {"field_name": "datasource", "values": [{"value": "gdrive"}]}
        ]
    },
)

for r in response.results:
    print(r.title)
    print("  ", r.url)
    print("  snippet:", r.snippets[0].text if r.snippets else "(none)")

The snippets field is the "why it matched" — the same matched text the UI highlights. Verified against api_clients.