1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
|
-module(fabric_doc).
-export([open_doc/3, open_revs/4, get_missing_revs/2, update_docs/3]).
-include("../../couch/src/couch_db.hrl").
-include("../../dynomite/include/membership.hrl").
open_doc(DbName, Id, Options) ->
Workers = fabric_util:submit_jobs(partitions:for_key(DbName,Id), open_doc,
[Id, Options]),
SuppressDeletedDoc = not lists:member(deleted, Options),
Acc0 = {length(Workers), couch_util:get_value(r, Options, 1), []},
case fabric_util:recv(Workers, #shard.ref, fun handle_open_doc/3, Acc0) of
{ok, {ok, #doc{deleted=true}}} when SuppressDeletedDoc ->
{not_found, deleted};
{ok, Else} ->
Else;
Error ->
Error
end.
open_revs(DbName, Id, Revs, Options) ->
Workers = fabric_util:submit_jobs(partitions:for_key(DbName,Id), open_revs,
[Id, Revs, Options]),
Acc0 = {length(Workers), couch_util:get_value(r, Options, 1), []},
fabric_util:recv(Workers, #shard.ref, fun handle_open_revs/3, Acc0).
update_docs(DbName, AllDocs, Options) ->
GroupedDocs = lists:map(fun({#shard{name=Name, node=Node} = Shard, Docs}) ->
Ref = rexi:cast(Node, {fabric_rpc, update_docs, [Name, Docs, Options]}),
{Shard#shard{ref=Ref}, Docs}
end, group_docs_by_shard(DbName, AllDocs)),
{Workers, _} = lists:unzip(GroupedDocs),
Acc0 = {length(Workers), length(AllDocs), couch_util:get_value(w, Options, 1),
GroupedDocs, dict:new()},
case fabric_util:recv(Workers, #shard.ref, fun handle_update_docs/3, Acc0) of
{ok, Results} ->
{ok, couch_util:reorder_results(AllDocs, Results)};
Else ->
Else
end.
get_missing_revs(_DbName, _IdsRevs) ->
ok.
handle_open_doc(_Worker, {rexi_DOWN, _, _, _}, Acc0) ->
skip_message(Acc0);
handle_open_doc(_Worker, {rexi_EXIT, _Reason}, Acc0) ->
skip_message(Acc0);
handle_open_doc(_Worker, Reply, {WaitingCount, R, Replies}) ->
case merge_read_reply(make_key(Reply), Reply, Replies) of
{_, KeyCount} when KeyCount =:= R ->
{stop, Reply};
{NewReplies, KeyCount} when KeyCount < R ->
if WaitingCount =:= 1 ->
% last message arrived, but still no quorum
repair_read_quorum_failure(NewReplies);
true ->
{ok, {WaitingCount-1, R, NewReplies}}
end
end.
handle_open_revs(_Worker, {rexi_DOWN, _, _, _}, Acc0) ->
skip_message(Acc0);
handle_open_revs(_Worker, {rexi_EXIT, _}, Acc0) ->
skip_message(Acc0);
handle_open_revs(_Worker, _Reply, {_WaitingCount, _R, _Replies}) ->
{stop, not_implemented}.
handle_update_docs(_Worker, {rexi_DOWN, _, _, _}, Acc0) ->
skip_message(Acc0);
handle_update_docs(_Worker, {rexi_EXIT, _}, Acc0) ->
skip_message(Acc0);
handle_update_docs(Worker, {ok, Replies}, Acc0) ->
{WaitingCount, DocCount, W, GroupedDocs, DocReplyDict0} = Acc0,
Docs = couch_util:get_value(Worker, GroupedDocs),
DocReplyDict = append_update_replies(Docs, Replies, DocReplyDict0),
case {WaitingCount, dict:size(DocReplyDict)} of
{1, _} ->
% last message has arrived, we need to conclude things
{W, Reply} = dict:fold(fun force_update_reply/3, {W,[]}, DocReplyDict),
{stop, Reply};
{_, DocCount} ->
% we've got at least one reply for each document, let's take a look
case dict:fold(fun maybe_update_reply/3, {stop,W,[]}, DocReplyDict) of
continue ->
{ok, {WaitingCount - 1, DocCount, W, GroupedDocs, DocReplyDict}};
{stop, W, FinalReplies} ->
{stop, FinalReplies}
end;
{_, N} when N < DocCount ->
% no point in trying to finalize anything yet
{ok, {WaitingCount - 1, DocCount, W, GroupedDocs, DocReplyDict}}
end.
force_update_reply(Doc, Replies, {W, Acc}) ->
% TODO make a real decision here
case Replies of
[] ->
{W, [{Doc, {error, internal_server_error}} | Acc]};
[Reply| _] ->
{W, [{Doc, Reply} | Acc]}
end.
maybe_update_reply(_, _, continue) ->
% we didn't meet quorum for all docs, so we're fast-forwarding the fold
continue;
maybe_update_reply(Doc, Replies, {stop, W, Acc}) ->
case update_quorum_met(W, Replies) of
{true, Reply} ->
{stop, W, [{Doc, Reply} | Acc]};
false ->
continue
end.
update_quorum_met(W, Replies) ->
% TODO make a real decision here
case length(Replies) >= W of
true ->
{true, hd(Replies)};
false ->
false
end.
-spec group_docs_by_shard(binary(), [#doc{}]) -> [{#shard{}, [#doc{}]}].
group_docs_by_shard(DbName, Docs) ->
dict:to_list(lists:foldl(fun(#doc{id=Id} = Doc, D0) ->
lists:foldl(fun(Shard, D1) ->
dict:append(Shard, Doc, D1)
end, D0, partitions:for_key(DbName,Id))
end, dict:new(), Docs)).
skip_message({1, _R, Replies}) ->
repair_read_quorum_failure(Replies);
skip_message({WaitingCount, R, Replies}) ->
{ok, {WaitingCount-1, R, Replies}}.
merge_read_reply(Key, Reply, Replies) ->
case lists:keyfind(Key, 1, Replies) of
false ->
{[{Key, Reply, 1} | Replies], 1};
{Key, _, N} ->
{lists:keyreplace(Key, 1, Replies, {Key, Reply, N+1}), N+1}
end.
append_update_replies([], [], DocReplyDict) ->
DocReplyDict;
append_update_replies([Doc|Rest1], [Reply|Rest2], Dict0) ->
% TODO what if the same document shows up twice in one update_docs call?
append_update_replies(Rest1, Rest2, dict:append(Doc, Reply, Dict0)).
make_key({ok, #doc{id=Id, revs=Revs}}) ->
{Id, Revs};
make_key({not_found, missing}) ->
{not_found, missing}.
repair_read_quorum_failure(Replies) ->
case [Doc || {_Key, {ok, Doc}, _Count} <- Replies] of
[] ->
{stop, {not_found, missing}};
[Doc|_] ->
% TODO merge docs to find the winner as determined by replication
{stop, {ok, Doc}}
end.
|
