update robots.txt and sitemap.xml

- show minimal robots/sitemap if not in prod environment
- default to allow all in robots.txt; link to sitemap index files
- basic sitemap.xml without entity-level links

4 files changed, 52 insertions, 2 deletions

diff --git a/python/fatcat_web/routes.py b/python/fatcat_web/routes.py
index da2bb6cf..c66c7f0c 100644
--- a/python/fatcat_web/routes.py
+++ b/python/fatcat_web/routes.py
@@ -1157,7 +1157,18 @@ def page_rfc():
     return render_template('rfc.html')
 
 @app.route('/robots.txt', methods=['GET'])
-def robots():
+def page_robots_txt():
+    if conf.FATCAT_DOMAIN == "fatcat.wiki":
+        robots_path = "robots.txt"
+    else:
+        robots_path = "robots.deny_all.txt"
     return send_from_directory(os.path.join(app.root_path, 'static'),
-                               'robots.txt',
+                               robots_path,
                                mimetype='text/plain')
+
+@app.route('/sitemap.xml', methods=['GET'])
+def page_sitemap_xml():
+    if conf.FATCAT_DOMAIN == "fatcat.wiki":
+        return redirect('/sitemaps/sitemap.xml')
+    else:
+        abort(404)
diff --git a/python/fatcat_web/static/robots.deny_all.txt b/python/fatcat_web/static/robots.deny_all.txt
new file mode 100644
index 00000000..b88274b1
--- /dev/null
+++ b/python/fatcat_web/static/robots.deny_all.txt
@@ -0,0 +1,7 @@
+# Hello friends!
+
+# You have found a QA/development instance of the Fatcat catalog. The canonical
+# location is https://fatcat.wiki, please crawl and index that location instead.
+
+User-agent: *
+Disallow: /
diff --git a/python/fatcat_web/static/robots.txt b/python/fatcat_web/static/robots.txt
index a168f11b..e89af36e 100644
--- a/python/fatcat_web/static/robots.txt
+++ b/python/fatcat_web/static/robots.txt
@@ -1 +1,20 @@
 # Hello friends!
+# If you are considering large or automated crawling, you may want to look at
+# our API (https://api.fatcat.wiki) or bulk database snapshots instead.
+
+# by default, can crawl anything on this domain. HTTP 429 ("backoff") status
+# codes are used for rate-limiting instead of any crawl delay specified here.
+# Up to a handful concurrent requests should be fine.
+User-agent: *
+Allow: /
+
+# crawling search result pages is expensive, so we do specify a long crawl delay for those
+User-agent: *
+Allow: /release/search
+Allow: /container/search
+Allow: /coverage/search
+Crawl-delay: 5
+
+Sitemap: https://fatcat.wiki/sitemap.xml
+Sitemap: https://fatcat.wiki/sitemap-index-releases.xml
+Sitemap: https://fatcat.wiki/sitemap-index-containers.xml
diff --git a/python/fatcat_web/static/sitemap.xml b/python/fatcat_web/static/sitemap.xml
new file mode 100644
index 00000000..e6189aa4
--- /dev/null
+++ b/python/fatcat_web/static/sitemap.xml
@@ -0,0 +1,13 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+  <!-- basic site pages -->
+  <url><loc>https://fatcat.wiki/</loc></url>
+  <url><loc>https://fatcat.wiki/about</loc></url>
+  <url><loc>https://fatcat.wiki/rfc</loc></url>
+  <url><loc>https://fatcat.wiki/stats</loc></url>
+  <url><loc>https://fatcat.wiki/changelog</loc></url>
+  <url><loc>https://fatcat.wiki/release/lookup</loc></url>
+  <url><loc>https://fatcat.wiki/container/lookup</loc></url>
+  <url><loc>https://fatcat.wiki/file/lookup</loc></url>
+  <!-- additional entity-level URL lists are linked from robots.txt -->
+</urlset>