memento: for not found timemap query, return empty timemap, instead of html query error page, closes #158

Author: ikreymer

pywb/framework/memento.py

@@ -181,14 +181,19 @@ def make_timemap(wbrequest, cdx_lines):
     mod = wbrequest.options.get('replay_mod', '')
 
     # get first memento as it'll be used for 'from' field
-    first_cdx = cdx_lines.next()
-    from_date = timestamp_to_http_date(first_cdx['timestamp'])
+    try:
+        first_cdx = cdx_lines.next()
+        from_date = timestamp_to_http_date(first_cdx['timestamp'])
+    except StopIteration:
+        first_cdx = None
 
-    # timemap link
-    timemap = ('<{0}>; rel="self"; ' +
-               'type="application/link-format"; from="{1}",\n')
-    yield timemap.format(prefix + wbrequest.wb_url.to_str(),
-                         from_date)
+
+    if first_cdx:
+        # timemap link
+        timemap = ('<{0}>; rel="self"; ' +
+                   'type="application/link-format"; from="{1}",\n')
+        yield timemap.format(prefix + wbrequest.wb_url.to_str(),
+                             from_date)
 
     # original link
     original = '<{0}>; rel="original",\n'
@@ -202,6 +207,12 @@ def make_timemap(wbrequest, cdx_lines):
 
     yield timegate.format(prefix + timegate_url)
 
+    if not first_cdx:
+        # terminating timemap link, no from
+        timemap = ('<{0}>; rel="self"; type="application/link-format"')
+        yield timemap.format(prefix + wbrequest.wb_url.to_str())
+        return
+
     # first memento link
     yield make_timemap_memento_link(first_cdx, prefix,
                             datetime=from_date, mod=mod)

pywb/webapp/handlers.py

@@ -176,7 +176,8 @@ def handle_not_found(self, wbrequest, nfe):
 
         # if capture query, just return capture page
         if wbrequest.wb_url.is_query():
-            return self.index_reader.make_cdx_response(wbrequest, [], 'html')
+            output = self.index_reader.get_output_type(wbrequest.wb_url)
+            return self.index_reader.make_cdx_response(wbrequest, iter([]), output)
         else:
             return self.not_found_view.render_response(status='404 Not Found',
                                                        wbrequest=wbrequest,

pywb/webapp/query_handler.py

@@ -46,11 +46,7 @@ def init_from_config(config,
 
         return QueryHandler(cdx_server, html_view, perms_policy)
 
-    def load_for_request(self, wbrequest):
-        wbrequest.normalize_post_query()
-
-        wb_url = wbrequest.wb_url
-
+    def get_output_type(self, wb_url):
         # cdx server only supports text and cdxobject for now
         if wb_url.mod == 'cdx_':
             output = 'text'
@@ -61,6 +57,14 @@ def load_for_request(self, wbrequest):
         else:
             output = 'cdxobject'
 
+        return output
+
+    def load_for_request(self, wbrequest):
+        wbrequest.normalize_post_query()
+
+        wb_url = wbrequest.wb_url
+        output = self.get_output_type(wb_url)
+
         # init standard params
         params = self.get_query_params(wb_url)
 

tests/test_memento.py

@@ -306,6 +306,44 @@ def test_timemap_2(self):
 
         assert len(lines) == 3 + 3
 
+
+    def test_timemap_not_found(self):
+        """
+        Test application/link-format timemap
+        """
+
+        resp = self.testapp.get('/pywb/timemap/*/http://example.com/blah/not_found')
+        assert resp.status_int == 200
+        assert resp.content_type == LINK_FORMAT
+
+        lines = resp.body.split('\n')
+
+        assert len(lines) == 3
+
+        assert lines[0] == '<http://example.com/blah/not_found>; rel="original",'
+
+        assert lines[1] == '<http://localhost:80/pywb/http://example.com/blah/not_found>; rel="timegate",'
+
+        assert lines[2] == '<http://localhost:80/pywb/timemap/*/http://example.com/blah/not_found>; \
+rel="self"; type="application/link-format"'
+
+
+    def test_timemap_2(self):
+        """
+        Test application/link-format timemap total count
+        """
+
+        resp = self.testapp.get('/pywb/timemap/*/http://example.com')
+        assert resp.status_int == 200
+        assert resp.content_type == LINK_FORMAT
+
+        lines = resp.body.split('\n')
+
+        assert len(lines) == 3 + 3
+
+
+
+
     # Below functions test pywb proxy mode behavior
     # They are designed to roughly conform to Memento protocol Pattern 1.3
     # with the exception that the original resource is not available