Merge pull request 'Updates the generation means of robots.txt and sitemap.xml to be based off siteconfig's website param.' (#5) from bugfix/fix_API_robots_txt into main

Reviewed-on: kfj001/kmtnflower#5

backend/app/api/sitemap.py

@@ -1,35 +1,66 @@
 """Sitemap & robots.txt API: tenant-aware SEO files served by the backend.
 
 Because this is a white-label app where each tenant has its own domain
-(set via KMTN_WEBSITE_URL), sitemap.xml and robots.txt are generated
+(stored in StationConfig.website), sitemap.xml and robots.txt are
-on-the-fly from the station config — no static files needed.
+generated on-the-fly from the database — no static files needed.
 
 This matches the reviewer feedback:
   'This is a whitelabel app - this file may need to be dynaimcally generated...'
 
 Follows the same pattern as theme_export.py: content is derived from
-KMTN_WEBSITE_URL at runtime, so every tenant gets their own correct URL.
+the station config at runtime, so every tenant gets their own correct URL.
 """
 
 from typing import Any, Dict
 
 from fastapi import APIRouter, Depends, Response, status as http_status
+from sqlalchemy import select
 
 from app.auth import get_current_admin_user
 from app.config import settings
+from app.database import async_session
+from app.models import StationConfig
 from app.sitemap_export import build_robots_txt, build_sitemap_xml
 from app.user_models import User
 
 router = APIRouter()
 
 
+def _build_base_url(website: str) -> str:
+    """Build the canonical base URL from the station website field.
+
+    The website column stores the bare domain (e.g. 'kmountainflower.org'),
+    so we prefix with https:// to produce a valid base URL.
+    """
+    domain = website.strip().removesuffix("/")
+    # The website field may or may not include a scheme; if it already
+    # has one, use it as-is, otherwise default to https.
+    if "://" in domain:
+        return domain
+    return f"https://{domain}"
+
+
+async def _get_base_url() -> str:
+    """Fetch the base URL from the station config in the database."""
+    async with async_session() as session:
+        result = await session.execute(
+            select(StationConfig).where(StationConfig.key == "default")
+        )
+        config = result.scalar_one_or_none()
+    if config is None:
+        # Fall back to env config if DB is empty (e.g. before seeding)
+        return settings.WEBSITE_URL
+    return _build_base_url(config.website)
+
+
 @router.get("/sitemap.xml")
 async def get_sitemap() -> Response:
     """Return a dynamically-generated sitemap.xml. Public endpoint.
 
-    Uses KMTN_WEBSITE_URL so each tenant gets the correct domain.
+    Uses StationConfig.website so each tenant gets the correct domain.
     """
-    xml = build_sitemap_xml(settings.WEBSITE_URL)
+    base_url = await _get_base_url()
+    xml = build_sitemap_xml(base_url)
     return Response(content=xml, media_type="application/xml")
 
 
@@ -37,10 +68,11 @@ async def get_sitemap() -> Response:
 async def get_robots() -> Response:
     """Return a dynamically-generated robots.txt. Public endpoint.
 
-    Uses KMTN_WEBSITE_URL so the Sitemap directive points to the
+    Uses StationConfig.website so the Sitemap directive points to the
     correct tenant domain.
     """
-    txt = build_robots_txt(settings.WEBSITE_URL)
+    base_url = await _get_base_url()
+    txt = build_robots_txt(base_url)
     return Response(content=txt, media_type="text/plain")
 
 
@@ -53,8 +85,9 @@ async def regenerate_sitemap(
     In the dynamic model this is a no-op (files are generated per-request),
     but kept for API consistency with the theme endpoint.
     """
+    base_url = await _get_base_url()
     return {
         "status": "ok",
-        "note": "sitemap and robots.txt are dynamically generated per-request using KMTN_WEBSITE_URL",
+        "note": "sitemap and robots.txt are dynamically generated per-request from StationConfig.website",
-        "base_url": settings.WEBSITE_URL,
+        "base_url": base_url,
     }

backend/tests/test_sitemap_api.py

@@ -1,12 +1,38 @@
-"""Tests for sitemap API endpoints — /api/sitemap/sitemap.xml, /api/sitemap/robots.txt, /api/sitemap/regenerate."""
+"""Tests for sitemap API endpoints — /api/sitemap/sitemap.xml, /api/sitemap/robots.txt, /api/sitemap/regenerate.
+
+Note: The sitemap and robots.txt endpoints now read the site URL from
+StationConfig.website in the database (falling back to KMTN_WEBSITE_URL
+if the config is missing). These tests assume a seeded database with the
+default station config (website = 'kmountainflower.org').
+"""
 
 import pytest
 from httpx import AsyncClient
 from starlette.testclient import TestClient
 
+from app.api.sitemap import _build_base_url
 from app.main import app
 
 
+class TestBuildBaseUrl:
+    """Tests for the _build_base_url helper."""
+
+    def test_bare_domain_gets_https(self):
+        assert _build_base_url("kmountainflower.org") == "https://kmountainflower.org"
+
+    def test_domain_with_trailing_slash(self):
+        assert _build_base_url("kmountainflower.org/") == "https://kmountainflower.org"
+
+    def test_domain_with_whitespace(self):
+        assert _build_base_url("  mystation.org  ") == "https://mystation.org"
+
+    def test_already_has_https_scheme(self):
+        assert _build_base_url("https://kmountainflower.org") == "https://kmountainflower.org"
+
+    def test_already_has_http_scheme(self):
+        assert _build_base_url("http://kmountainflower.org") == "http://kmountainflower.org"
+
+
 class TestSitemapEndpoints:
     """Tests for sitemap API routes."""