WangShen007
54b949b392
The upstream project name is gone from code, assets, docs and build
output: brand strings, domains (replypier.site and its app/api/docs/help
subdomains), env prefixes, image and network names, the widget's public
JS API (window.ReplyPier, replypierId/BaseUrl/Config/Token), CSS tokens
(--rp-*), file and directory names, and the WordPress plugin package.
- widget loader is now webclient/replypier.min.js; both bundles rebuilt
from source (backend/assets/widget.{js,css}, frontend/public/webclient)
- NOTICE, llms.txt, CONTRIBUTING, HELP_CENTER_INFRA and config.dev.js
rewritten rather than string-swapped; the demo config no longer ships
the previous vendor's Firebase/Razorpay/Google keys
- new migration replypier_ai_model_001 adds REPLYPIER to the managed
AIModelType and rewrites pre-rename rows
- local compose project, database, role and volume renamed; LOCAL_DEPLOY
documents carrying the old local data over
105 lines
4.0 KiB
Python
105 lines
4.0 KiB
Python
"""
|
|
Copyright 2024-2026 ReplyPier
|
|
|
|
Licensed under the Apache License, Version 2.0 (the "License");
|
|
you may not use this file except in compliance with the License.
|
|
You may obtain a copy of the License at
|
|
|
|
http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
Unless required by applicable law or agreed to in writing, software
|
|
distributed under the License is distributed on an "AS IS" BASIS,
|
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
See the License for the specific language governing permissions and
|
|
limitations under the License.
|
|
|
|
The API origin's crawler policy. Search Console reported dozens of
|
|
/api/v1/... endpoints as "Not found (404)" because this host answered
|
|
robots.txt with a 404 of its own, so these run against the real wired app
|
|
rather than a stand-in — the wiring is the part that was missing.
|
|
"""
|
|
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
from fastapi.testclient import TestClient
|
|
|
|
import app.main # noqa: F401 — the robots route and middleware are wired on import
|
|
from app.core.application import app
|
|
from app.core.config import settings
|
|
from app.core.crawler_policy import api_robots_txt
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
def client():
|
|
return TestClient(app)
|
|
|
|
|
|
class TestApiRobotsTxt:
|
|
def test_served_as_plain_text_not_json(self, client):
|
|
response = client.get("/robots.txt")
|
|
|
|
assert response.status_code == 200
|
|
assert response.headers["content-type"].startswith("text/plain")
|
|
|
|
def test_disallows_everything(self, client):
|
|
lines = client.get("/robots.txt").text.splitlines()
|
|
|
|
assert "User-agent: *" in lines
|
|
assert "Disallow: /" in lines
|
|
|
|
def test_path_mode_keeps_the_self_hosted_help_center_crawlable(self, monkeypatch):
|
|
"""Path dispatch serves public help-center pages from this same origin,
|
|
and those are meant to be indexed."""
|
|
monkeypatch.setattr(settings, "HELP_CENTER_PUBLIC_MODE", "path")
|
|
|
|
assert "Allow: /help/" in api_robots_txt().splitlines()
|
|
|
|
def test_path_mode_allow_precedes_the_blanket_disallow(self, monkeypatch):
|
|
"""A crawler that takes the first match instead of the most specific
|
|
one must not read "Disallow: /" before reaching the carve-out."""
|
|
monkeypatch.setattr(settings, "HELP_CENTER_PUBLIC_MODE", "path")
|
|
lines = api_robots_txt().splitlines()
|
|
|
|
assert lines.index("Allow: /help/") < lines.index("Disallow: /")
|
|
|
|
def test_subdomain_mode_has_nothing_to_carve_out(self, monkeypatch):
|
|
monkeypatch.setattr(settings, "HELP_CENTER_PUBLIC_MODE", "subdomain")
|
|
lines = api_robots_txt().splitlines()
|
|
|
|
assert "Disallow: /" in lines
|
|
assert not any(line.startswith("Allow:") for line in lines)
|
|
|
|
|
|
class TestNoIndexHeader:
|
|
"""robots.txt stops the fetch; this stops a URL discovered elsewhere from
|
|
being indexed regardless."""
|
|
|
|
def test_api_success_response_is_noindex(self, client):
|
|
probe = Path("uploads") / "agents" / "crawler-policy-probe.txt"
|
|
probe.parent.mkdir(parents=True, exist_ok=True)
|
|
probe.write_text("probe")
|
|
try:
|
|
response = client.get("/api/v1/uploads/agents/crawler-policy-probe.txt")
|
|
finally:
|
|
probe.unlink(missing_ok=True)
|
|
|
|
assert response.status_code == 200
|
|
assert response.headers["x-robots-tag"] == "noindex"
|
|
|
|
def test_api_404_is_noindex(self, client):
|
|
"""The 404s are exactly what Search Console indexed, so they need it
|
|
most. Deliberately a path no router claims: the 404 is then raised
|
|
before any endpoint dependency runs, so this asserts the middleware
|
|
and not some unrelated route's database access."""
|
|
response = client.get("/api/v1/no-such-endpoint")
|
|
|
|
assert response.status_code == 404
|
|
assert response.headers["x-robots-tag"] == "noindex"
|
|
|
|
def test_non_api_paths_are_left_alone(self, client):
|
|
response = client.get("/health")
|
|
|
|
assert response.status_code == 200
|
|
assert "x-robots-tag" not in response.headers
|