Files
model-platform/backend/tests/test_count_scripts.py
T
tao.chenandClaude bca239ed4b refactor(backend): split into api/ schemas/ services/ clients/ layers
4-phase restructuring of the previously flat backend/ package. Each
phase lands as a single squash commit so future bisects stay readable
per phase if needed.

## Phase 1 — move + shim (location-only, zero behavior change)
* git mv 14 files into api/ schemas/ services/ clients/ subpackages
  (history preserved via RM/R renames)
* New files: api/{admin,auth,dependencies,jupyter,platform,resources,
  scripts,storage}.py + api/schedules/{schedules,runs}.py
* New files: schemas/{auth,common,jupyter,platform,resources,
  schedules,scripts}.py
* New files: clients/{rclone,runtime,scheduler}.py
* Old paths kept as 1-line `from backend.<new> import *` shims so
  tests/main.py/importers kept working untouched
* schemas/__init__.py now re-exports from backend.schemas.<domain>

## Phase 2 — APIRouter prefix consolidation
* Every APIRouter() now carries its prefix (e.g. prefix="/api/v1/auth")
  and decorators are stripped of the redundant path prefix
* URL paths exposed to the frontend are byte-identical to before
* Affected: api/{auth,jupyter,admin,platform,resources,scripts,
  storage}.py + api/schedules/{schedules,runs}.py

## Phase 3 — first service-layer extraction
* backend.services.schedules.validate_dag moved out of api/
  (pure DAG validator, no Request/BackgroundTasks/DB)
* api/schedules/schedules.py now re-exports the symbol so existing
  4 callsites keep working unchanged
* Added backend/tests/test_validate_dag.py: 8 unit tests covering
  DAG_EMPTY, linear chain, diamond, cycle, self-edge, duplicate
  edge, orphan edge, multi-root ordering

## Phase 4 — delete shims + unify test imports
* Removed 14 flat shim files + schemas/__init__.py
* Migrated 5 test files (32 import sites) to new paths:
  backend.scripts.* → backend.api.scripts.*
  backend.resources.* → backend.api.resources.*
  backend.jupyter.* → backend.api.jupyter.*
  backend.runtime_client.* → backend.clients.runtime.*
  backend.schemas.UpdateScriptRequest → backend.schemas.scripts.*
* audit.py kept at backend.audit (main.py references it; not a
  shim, real code)

## Final structure
backend/src/backend/
  main.py, audit.py, __init__.py
  api/            (10 files: routes + 2 subpackage)
  schemas/        (7 files: Pydantic contracts)
  services/       (storage + schedules)
  clients/        (rclone, runtime, scheduler)

## Verification
* uv run python -m compileall backend/src backend/tests — clean
* uv run --package backend pytest backend/tests -q — 122 passed
  (114 → 114 → 122 → 122 across phases)
* grep -r 'from backend\.\(scripts\|resources\|...\)' backend/ — 0 hits
* git blame --follow still traces file origins through the renames

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2026-08-21 15:32:04 +08:00

146 lines
5.3 KiB
Python

"""Unit tests for GET /api/v1/scripts/count endpoint.
Verifies the count endpoint matches the (workspace-wide) listing scope of
``list_scripts(parent_path="")``: workspace + active scripts across every
owner's ``StorageObjects.relative_path``, narrowed by visibility for
non-admin (admin short-circuits). This avoids under/over-reporting on the
dashboard — the count is the size of the set list_scripts would return if
it weren't lazy.
"""
from __future__ import annotations
from types import SimpleNamespace
from unittest.mock import AsyncMock, MagicMock
import pytest
from backend.api.scripts import count_scripts
def _ctx(
user_id: str = "U001",
workspace_id: str = "W001",
*,
is_admin: bool = False,
is_system_admin: bool = False,
) -> SimpleNamespace:
return SimpleNamespace(
request_id="test",
user=SimpleNamespace(user_id=user_id),
workspace=SimpleNamespace(workspace_id=workspace_id),
role=SimpleNamespace(role_code="admin" if is_admin else "developer"),
is_system_admin=is_system_admin,
# ``count_scripts`` now consults ``context.is_admin`` directly
# (matching list_scripts / list_resources); SimpleNamespace needs
# it as a plain attribute.
is_admin=is_admin or is_system_admin,
)
def _compile(stmt) -> str:
from sqlalchemy.dialects import mysql as mysql_dialect
return str(
stmt.compile(
dialect=mysql_dialect.dialect(),
compile_kwargs={"literal_binds": True},
)
)
async def test_count_scripts_returns_scalar_int() -> None:
captured = []
mock_session = MagicMock()
mock_session.scalar = AsyncMock(
side_effect=lambda stmt: (captured.append(stmt), 7)[1]
)
result = await count_scripts(context=_ctx(), session=mock_session)
assert result["data"] == {"total": 7}
assert result["meta"] == {}
assert result["request_id"] == "test"
# Exactly one COUNT(*) query issued.
assert len(captured) == 1
stmt = captured[0]
sql = _compile(stmt).lower()
# JOIN to StorageObjects so orphaned scripts (no joinable row) are
# excluded — matches list_scripts INNER JOIN behaviour.
assert "inner join storage_objects" in sql
# Scope: workspace_id + active status + workspace-wide prefix.
assert "scripts.workspace_id" in sql
assert "scripts.status" in sql
assert "like 'workspace/%%'" in sql
# Non-admin (default) narrows by visibility.
assert "scripts.owner_user_id = 'u001'" in sql
assert "scripts.visibility in ('workspace', 'public')" in sql
async def test_count_scripts_handles_null_result() -> None:
"""MySQL COUNT(*) on empty result returns 0, not NULL — but defensively
coerce NULL to 0 to keep the response shape consistent."""
mock_session = MagicMock()
mock_session.scalar = AsyncMock(return_value=None)
result = await count_scripts(context=_ctx(), session=mock_session)
assert result["data"] == {"total": 0}
async def test_count_scripts_workspace_wide_not_user_scoped() -> None:
"""The prefix is workspace-wide (``workspace/%`` — no embedded user_id),
so different users count the same physical tree; the only per-user
difference is the non-admin visibility predicate (owner_user_id = me)."""
captured = []
mock_session = MagicMock()
mock_session.scalar = AsyncMock(
side_effect=lambda stmt: (captured.append(stmt), 3)[1]
)
await count_scripts(context=_ctx(user_id="alice"), session=mock_session)
sql_alice = _compile(captured[-1]).lower()
await count_scripts(context=_ctx(user_id="bob"), session=mock_session)
sql_bob = _compile(captured[-1]).lower()
# Both count the same workspace-wide subtree.
assert "like 'workspace/%%'" in sql_alice
assert "like 'workspace/%%'" in sql_bob
# Neither embeds the user_id in the path prefix.
assert "workspace/alice/%" not in sql_alice
assert "workspace/bob/%" not in sql_bob
# Per-user narrowing happens via the visibility predicate.
assert "scripts.owner_user_id = 'alice'" in sql_alice
assert "scripts.owner_user_id = 'bob'" in sql_bob
async def test_count_scripts_admin_skips_visibility_filter() -> None:
"""Admin short-circuits the visibility predicate and counts every
active script in the workspace (dashboard '全部脚本' / '工作副本')."""
captured = []
mock_session = MagicMock()
mock_session.scalar = AsyncMock(
side_effect=lambda stmt: (captured.append(stmt), 42)[1]
)
result = await count_scripts(
context=_ctx(user_id="alice", is_admin=True), session=mock_session
)
assert result["data"] == {"total": 42}
sql = _compile(captured[0]).lower()
assert "like 'workspace/%%'" in sql
# visibility / owner_user_id still appear in the SELECT projection, but
# the visibility WHERE predicate must be absent for admins.
assert "scripts.visibility in ('workspace', 'public')" not in sql
async def test_count_scripts_route_declared_before_script_id_route() -> None:
"""Static check: the `/api/v1/scripts/count` route MUST be declared in
scripts.py before `/api/v1/scripts/{script_id}/...`, otherwise FastAPI's
declaration-order matching will interpret `count` as a script_id."""
from backend.api.scripts import count_scripts, get_script
assert callable(count_scripts)
assert callable(get_script)