hashserv: server: Support searching upstream for outhash

Use the new get-outhash message to perform a read-only query against an
upstream server (if present) when a reported taskhash/outhash
combination is not found in the current database. If a matching entry is
found upstream it is copied into the current database so it can be found
by future queries.

Signed-off-by: Paul Barker <pbarker@konsulko.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

2 files changed, 32 insertions, 0 deletions

diff --git a/lib/hashserv/server.py b/lib/hashserv/server.py
index 2770c2360..9ade988e5 100644
--- a/lib/hashserv/server.py
+++ b/lib/hashserv/server.py
@@ -130,6 +130,18 @@ async def copy_from_upstream(client, db, method, taskhash):
         d = {k: v for k, v in d.items() if k in TABLE_COLUMNS}
         keys = sorted(d.keys())
 
+        with closing(db.cursor()) as cursor:
+            insert_task(cursor, d)
+            db.commit()
+
+    return d
+
+async def copy_outhash_from_upstream(client, db, method, outhash, taskhash):
+    d = await client.get_outhash(method, outhash, taskhash)
+    if d is not None:
+        # Filter out unknown columns
+        d = {k: v for k, v in d.items() if k in TABLE_COLUMNS}
+        keys = sorted(d.keys())
 
         with closing(db.cursor()) as cursor:
             insert_task(cursor, d)
@@ -359,6 +371,14 @@ class ServerClient(object):
 
             row = cursor.fetchone()
 
+            if row is None and self.upstream_client:
+                # Try upstream
+                row = await copy_outhash_from_upstream(self.upstream_client,
+                                                       self.db,
+                                                       data['method'],
+                                                       data['outhash'],
+                                                       data['taskhash'])
+
             # If no matching outhash was found, or one *was* found but it
             # wasn't an exact match on the taskhash, a new entry for this
             # taskhash should be added
diff --git a/lib/hashserv/tests.py b/lib/hashserv/tests.py
index 6f04e30d6..1a696481e 100644
--- a/lib/hashserv/tests.py
+++ b/lib/hashserv/tests.py
@@ -246,6 +246,18 @@ class HashEquivalenceCommonTests(object):
         self.assertClientGetHash(side_client, taskhash4, unihash4)
         self.assertClientGetHash(self.client, taskhash4, None)
 
+        # Test that reporting a unihash in the downstream is able to find a
+        # match which was previously reported to the upstream server
+        taskhash5 = '35788efcb8dfb0a02659d81cf2bfd695fb30faf9'
+        outhash5 = '2765d4a5884be49b28601445c2760c5f21e7e5c0ee2b7e3fce98fd7e5970796f'
+        unihash5 = 'f46d3fbb439bd9b921095da657a4de906510d2cd'
+        result = self.client.report_unihash(taskhash5, self.METHOD, outhash5, unihash5)
+
+        taskhash6 = '35788efcb8dfb0a02659d81cf2bfd695fb30fafa'
+        unihash6 = 'f46d3fbb439bd9b921095da657a4de906510d2ce'
+        result = down_client.report_unihash(taskhash6, self.METHOD, outhash5, unihash6)
+        self.assertEqual(result['unihash'], unihash5, 'Server failed to copy unihash from upstream')
+
     def test_ro_server(self):
         (ro_client, ro_server) = self.start_server(dbpath=self.server.dbpath, read_only=True)