content-find-agent-new/tests/test_query_next_page_removed.py

"""M8E:query_next_page 边全链路无残留——硬编码集合 2 边、图 9 边、profile/源码/前端无残留。"""

import json
from pathlib import Path

from content_agent.business_modules import walk_engine
from content_agent.integrations.walk_graph_json import WalkGraphStore
from scripts.validate_v4_config_contract import M4_WALK_GATE_EDGES

ROOT = Path(__file__).resolve().parents[1]
_TWO_EDGES = {"hashtag_to_query", "author_to_works"}


def test_hardcoded_gate_edge_sets_are_two_edges():
    # 删边的"必崩点":四处硬编码集合必须全是 2 边(不含 query_next_page)。
    assert walk_engine.V4_WALK_GATE_EDGES == _TWO_EDGES
    assert M4_WALK_GATE_EDGES == _TWO_EDGES


def test_walk_graph_has_no_query_next_page_edge():
    graph = WalkGraphStore().load_graph()
    edge_ids = {edge["edge_id"] for edge in graph["edges"]}
    assert "query_next_page" not in edge_ids


def test_walk_policy_gate_applies_to_two_edges():
    policy = WalkGraphStore().load_policy()
    gate = policy.get("v4_walk_gate") or {}
    assert set(gate.get("applies_to_edges") or []) == _TWO_EDGES


def test_platform_profiles_have_no_query_next_page():
    profile_dir = ROOT / "tech_documents" / "数据接口与来源" / "platform_profiles"
    for path in profile_dir.glob("*.json"):
        data = json.loads(path.read_text(encoding="utf-8"))
        assert "query_next_page" not in json.dumps(data), f"{path.name} still references query_next_page"


def test_engine_and_frontend_source_have_no_query_next_page():
    for rel in [
        "content_agent/business_modules/walk_engine.py",
        "content_agent/flow_ledger_service.py",
        "content_agent/dashboard_service.py",
    ]:
        text = (ROOT / rel).read_text(encoding="utf-8")
        assert "query_next_page" not in text, f"{rel} still references query_next_page"