memento/api: add a new /collinfo.json end-point, enabled with 'enable_coll_info' config setting, which returns

the value fo collinfo.json template. Default template returns an entry for each handler route,
including the route path (id), title (name) and memento timegate and timemap paths, to be used with
an aggregator. Using a custom 'info_json' template can specify a different collinfo template, alternative to #69 (local aggregation)
Closes #146

Author: ikreymer

pywb/default_config.yaml

@@ -23,6 +23,7 @@ paths:
         proxy_cert_download_html: proxy_cert_download.html
         proxy_select_html: proxy_select.html
 
+        info_json: collinfo.json
 
 templates_dirs:
     - templates
@@ -49,6 +50,7 @@ not_found_html: not_found.html
 proxy_cert_download_html: proxy_cert_download.html
 proxy_select_html: proxy_select.html
 
+info_json: collinfo.json
 
 static_default_prefix: &static_default_prefix static/__pywb
 static_shared_prefix: static/__shared

pywb/framework/archivalrouter.py

@@ -23,9 +23,12 @@ def __init__(self, routes, **kwargs):
 
         self.home_view = kwargs.get('home_view')
         self.error_view = kwargs.get('error_view')
+        self.info_view = kwargs.get('info_view')
 
-        self.urlrewriter_class = (kwargs.get('config', {}).
-                                  get('urlrewriter_class', UrlRewriter))
+        config = kwargs.get('config', {})
+        self.urlrewriter_class = config.get('urlrewriter_class', UrlRewriter)
+
+        self.enable_coll_info = config.get('enable_coll_info', False)
 
     def __call__(self, env):
         request_uri = self.ensure_rel_uri_set(env)
@@ -43,6 +46,13 @@ def __call__(self, env):
         if request_uri in ['/', '/index.html', '/index.htm']:
             return self.render_home_page(env)
 
+        if self.enable_coll_info and request_uri in ['/collinfo.json']:
+            params = env.get('pywb.template_params', {})
+            host = WbRequest.make_host_prefix(env)
+            return self.info_view.render_response(env=env, host=host, routes=self.routes,
+                                                  content_type='application/json',
+                                                  **params)
+
         return self.fallback(env, self) if self.fallback else None
 
     def parse_request(self, route, env, matcher, coll, request_uri,

pywb/templates/collinfo.json

@@ -0,0 +1,15 @@
+[
+{% for route in routes %}
+{% if route | is_wb_handler %}
+{{ ',' if notfirst else '' }}
+{
+    "id": "{{ route.path }}",
+    "name": "{{ route.user_metadata.title if route.user_metadata.title else route.path }}", 
+    "timegate": "{{ host }}/{{route.path}}/",
+    "timemap": "{{ host }}/{{route.path}}/timemap/*/"
+
+}
+{% set notfirst = true %}
+{% endif %}
+{% endfor %}
+]

pywb/webapp/pywb_init.py

@@ -381,5 +381,6 @@ def create_wb_router(passed_config=None):
         abs_path=config.get('absolute_paths', True),
         home_view=init_view(config, 'home_html'),
         error_view=init_view(config, 'error_html'),
+        info_view=init_view(config, 'info_json'),
         config=config
     )

tests/test_config.yaml

@@ -121,6 +121,10 @@ proxy_options:
     use_client_rewrite: true
     use_wombat: true
 
+
+#enable coll info JSON
+enable_coll_info: true
+
 # enable cdx server api for querying cdx directly (experimental)
 #enable_cdx_api: True
 # or specify suffix

tests/test_integration.py

@@ -477,7 +477,13 @@ def test_error(self):
         assert resp.status_int == 400
         assert 'Invalid Url: http://?abc' in resp.body
 
-    #def test_invalid_config(self):
+
+    def test_coll_info_json(self):
+        resp = self.testapp.get('/collinfo.json')
+        assert resp.content_type == 'application/json'
+        assert len(resp.json) == 9
+
+   #def test_invalid_config(self):
     #    with raises(IOError):
     #        init_app(create_wb_router,
     #                 load_yaml=True,