kocolosk commented on a change in pull request #1370: [5/5] Clustered Purge 
Implementation
URL: https://github.com/apache/couchdb/pull/1370#discussion_r194942869
 
 

 ##########
 File path: src/mem3/src/mem3_rep.erl
 ##########
 @@ -169,20 +200,131 @@ find_source_seq_int(#doc{body={Props}}, SrcNode0, 
TgtNode0, TgtUUID, TgtSeq) ->
     end.
 
 
-repl(#acc{db = Db} = Acc0) ->
-    erlang:put(io_priority, {internal_repl, couch_db:name(Db)}),
-    #acc{seq=Seq} = Acc1 = calculate_start_seq(Acc0),
-    case Seq >= couch_db:get_update_seq(Db) of
-        true ->
-            {ok, 0};
-        false ->
-            Fun = fun ?MODULE:changes_enumerator/2,
-            {ok, Acc2} = couch_db:fold_changes(Db, Seq, Fun, Acc1),
-            {ok, #acc{seq = LastSeq}} = replicate_batch(Acc2),
-            {ok, couch_db:count_changes_since(Db, LastSeq)}
+repl(#acc{db = Db0} = Acc0) ->
+    erlang:put(io_priority, {internal_repl, couch_db:name(Db0)}),
+    Acc1 = calculate_start_seq(Acc0),
+    try
+        Acc3 = case config:get_boolean("mem3", "replicate_purges", false) of
+            true ->
+                Acc2 = pull_purges(Acc1),
+                push_purges(Acc2);
+            false ->
+                Acc1
+        end,
+        push_changes(Acc3)
+    catch
+        throw:{finished, Count} ->
+            {ok, Count}
     end.
 
 
+pull_purges(#acc{} = Acc0) ->
+    #acc{
+        batch_size = Count,
+        seq = UpdateSeq,
+        target = Target
+    } = Acc0,
+    #shard{
+        node = TgtNode,
+        name = TgtDbName
+    } = Target,
+
+    with_src_db(Acc0, fun(Db) ->
 
 Review comment:
   So now every invocation of `repl/1` loops back through `couch_server` 3 
times to open the database? Is that really necessary? I can see where you  need 
to reopen the DB after executing the purge requests that you pulled from the 
target but otherwise I'd expect that the source DB is more easily preserved 
across these calls.
   
   Also it's a little weird now that `#acc.db` is set but is the wrong thing to 
use.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

Reply via email to