Skip to content

Drive Tools — Activity — QA Test Cases

Source: src/mcp_gee_sweet/tools/drive/activity.py

Fixtures: see docs/qa/setup.md. Substitute your {SPREADSHEET_ID}, {DOC_ID}, and {FOLDER_ID} from fixtures.local.md.

list_file_activity

TC-D163: Basic activity fetch returns timeline entries

Prompt

"Show me the activity history for file {DOC_ID}"

Checks - file_id matches {DOC_ID} - activities is a list (may be empty for brand-new fixtures) - Each entry has timestamp, action, and actors keys - action is one of: edit, create, move, rename, delete, restore, permission_change, comment, settings_change, system_event, unknown - No error key in result

Result (2026-06-24) ✅ PASS 52 activities returned. All entries have timestamp, action, and actors. Actions observed: edit, rename, permission_change, create. Actor types include user (known, is_current_user: true/false) and system. No error key.

Result (2026-09-04) ✅ PASS list_file_activity({DOC_ID}) -> file_id matches; activities is a list of 5; every entry has timestamp/action/actors; actions observed: edit, move, create (all in allowed set); no error key. drive.activity.readonly scope is granted.

TC-D164: Known-user actor structure

Prompt

"Show me the activity history for file {DOC_ID} — I want to see who made each change"

Checks - At least one activity entry has an actor with type: "user" - That actor has a person_name field (e.g. "people/12345") - is_current_user is a boolean

Result (2026-06-24) ✅ PASS Multiple user actors returned. Current-user entries have person_name: "people/101951097007377611160", is_current_user: true. A second user (people/114161724974780080071, is_current_user: false) also appears. A system actor appears on the permission_change entry with event: null.

Result (2026-09-04) ✅ PASS Each actor: type:"user", person_name:"people/108427788683920971958", is_current_user:true (boolean). At least one user actor with a person_name field present. (All activity on this fixture is by the current user; no second/system actor in the current 5-entry window.)

TC-D165: Pagination — next_page_token present when results exceed page_size

Prompt

"List file activity for {DOC_ID} with page_size=1"

Checks - At least 1 activity in activities (the Drive Activity API treats page_size as a hint and may return slightly more due to activity grouping — do not assert exactly 1) - next_page_token is present in the response

Result (2026-06-24) ✅ PASS page_size=1 returned 2 activities (Drive Activity API groups related events and does not hard-clip to the requested count). next_page_token present. Confirmed pagination works.

Result (2026-09-04) ✅ PASS list_file_activity({DOC_ID}, page_size=1) -> 2 activities returned (Drive Activity API groups related events, does not hard-clip to 1) and next_page_token present.

TC-D166: Invalid file ID returns error (unit test)

Checks (unit test) - When the Drive Activity API returns an HTTP 403 or 404, the tool returns {"error": "..."} rather than raising an exception

Result (2026-06-24) ✅ Unit test test_http_error_returns_error_dict confirms HTTP errors are caught and returned as {"error": str(e)}.

Result (2026-09-04) ⏭️ SKIP Unit-test-only check (HTTP 403/404 -> {"error": ...}); no live fixture that is readable-but-activity-forbidden. Separately, live HttpError propagation for the sibling sharing tools was confirmed clean in TC-D126/D129/D131/D235.

TC-D168: list_file_activity response-size cap — code path only, no dedicated live fixture (issue #242)

Background: list_file_activity is already Drive-API-paginated (page_size clamped 1–100) and low per-item size — the only realistic exposure is a single activity's actors list ballooning on a file with many collaborators. Per the #242 decision doc, this tool intentionally did NOT get a dedicated live-fixture verification (reproducing hundreds of real Drive Activity events isn't cheaply reproducible) — the cap was added for defense-in-depth only, verified by unit tests (tests/drive/test_activity.py::TestListFileActivity::test_oversized_result_raises, test_error_points_to_page_size_not_local_path).

Result (2026-07-03) ✅ N/A (by design) — sanity-checked live that the tool still functions normally post-change (list_file_activity(file_id={SPREADSHEET_ID}, page_size=5) returned a normal activity list, no regression). Cap-triggering behavior covered by unit tests only, not live-verified — documented scoping decision, not an oversight.

Result (2026-09-04) ✅ PASS N/A by design (#242 decision doc). Sanity: list_file_activity({SPREADSHEET_ID}, page_size=5) returned a normal 8-entry activity list, no cap ValueError, no regression. Cap-trigger path is unit-tested only.