[
https://issues.apache.org/jira/browse/ASTERIXDB-1520?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15414522#comment-15414522
]
Yingyi Bu commented on ASTERIXDB-1520:
--------------------------------------
The query plan displays correctly:
{noformat}
distribute result [%0->$$6]
-- DISTRIBUTE_RESULT |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
project ([$$6])
-- STREAM_PROJECT |PARTITIONED|
assign [$$6] <- [function-call: asterix:open-record-constructor,
Args:[AString: {linenum1}, %0->$$18, AString: {count1}, %0->$$24, AString:
{t2info}, %0->$$17]]
-- ASSIGN |PARTITIONED|
exchange
-- SORT_MERGE_EXCHANGE [$$18(ASC) ] |PARTITIONED|
order (ASC, %0->$$18)
-- STABLE_SORT [$$18(ASC)] |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
project ([$$17, $$18, $$24])
-- STREAM_PROJECT |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
group by ([$$30 := %0->$$19]) decor ([%0->$$18; %0->$$24]) {
aggregate [$$17] <- [function-call:
asterix:listify, Args:[function-call: asterix:open-record-constructor,
Args:[AString: {linenum2}, %0->$$25, AString: {count2}, %0->$$23]]]
-- AGGREGATE |LOCAL|
select (function-call: algebricks:not,
Args:[function-call: algebricks:is-missing, Args:[%0->$$21]])
-- STREAM_SELECT |LOCAL|
nested tuple source
-- NESTED_TUPLE_SOURCE |LOCAL|
}
-- PRE_CLUSTERED_GROUP_BY[$$19] |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
order (ASC, %0->$$19) (ASC, %0->$$25)
-- STABLE_SORT [$$19(ASC), $$25(ASC)] |PARTITIONED|
exchange
-- HASH_PARTITION_EXCHANGE [$$19] |PARTITIONED|
select (function-call: algebricks:eq, Args:[%0->$$24,
%0->$$23])
-- STREAM_SELECT |PARTITIONED|
project ([$$18, $$19, $$21, $$23, $$24, $$25])
-- STREAM_PROJECT |PARTITIONED|
assign [$$25, $$23] <- [function-call:
asterix:field-access-by-name, Args:[%0->$$1, AString: {linenum}],
function-call: asterix:field-access-by-name, Args:[%0->$$1, AString: {count2}]]
-- ASSIGN |PARTITIONED|
project ([$$1, $$18, $$19, $$21, $$24])
-- STREAM_PROJECT |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
left-outer-unnest-map [$$21, $$1, $$22] <-
function-call: asterix:index-search, Args:[AString: {Book}, AInt32: {0},
AString: {test}, AString: {Book}, ABoolean: {true}, ABoolean: {false}, AInt32:
{1}, %0->$$35, AInt32: {1}, %0->$$35, TRUE, TRUE, TRUE]
-- BTREE_SEARCH |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
order (ASC, %0->$$35)
-- STABLE_SORT [$$35(ASC)]
|PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
project ([$$18, $$19, $$35, $$24])
-- STREAM_PROJECT |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE
|PARTITIONED|
left-outer-unnest-map [$$34,
$$35] <- function-call: asterix:index-search, Args:[AString: {NumIndex4},
AInt32: {0}, AString: {test}, AString: {Book}, ABoolean: {true}, ABoolean:
{true}, AInt32: {1}, %0->$$24, AInt32: {1}, %0->$$24, TRUE, TRUE, TRUE]
-- BTREE_SEARCH |PARTITIONED|
exchange
-- BROADCAST_EXCHANGE
|PARTITIONED|
project ([$$18, $$19, $$24])
-- STREAM_PROJECT
|PARTITIONED|
assign [$$24] <-
[function-call: asterix:field-access-by-name, Args:[%0->$$0, AString: {count1}]]
-- ASSIGN |PARTITIONED|
select (function-call:
algebricks:lt, Args:[%0->$$18, AInt64: {10}])
-- STREAM_SELECT
|PARTITIONED|
assign [$$18] <-
[function-call: asterix:field-access-by-name, Args:[%0->$$0, AString:
{linenum}]]
-- ASSIGN
|PARTITIONED|
project ([$$0,
$$19])
-- STREAM_PROJECT
|PARTITIONED|
exchange
--
ONE_TO_ONE_EXCHANGE |PARTITIONED|
unnest-map
[$$19, $$0, $$20] <- function-call: asterix:index-search, Args:[AString:
{Book}, AInt32: {0}, AString: {test}, AString: {Book}, ABoolean: {false},
ABoolean: {false}, AInt32: {1}, %0->$$33, AInt32: {1}, %0->$$33, TRUE, TRUE,
TRUE]
-- BTREE_SEARCH
|PARTITIONED|
exchange
--
ONE_TO_ONE_EXCHANGE |PARTITIONED|
order (ASC,
%0->$$33)
--
STABLE_SORT [$$33(ASC)] |PARTITIONED|
exchange
--
ONE_TO_ONE_EXCHANGE |PARTITIONED|
project
([$$33])
--
STREAM_PROJECT |PARTITIONED|
exchange
--
ONE_TO_ONE_EXCHANGE |PARTITIONED|
unnest-map [$$32, $$33] <- function-call: asterix:index-search, Args:[AString:
{NumIndex2}, AInt32: {0}, AString: {test}, AString: {Book}, ABoolean: {false},
ABoolean: {false}, AInt32: {0}, AInt32: {1}, %0->$$31, TRUE, FALSE, FALSE]
--
BTREE_SEARCH |PARTITIONED|
exchange
-- ONE_TO_ONE_EXCHANGE |PARTITIONED|
assign [$$31] <- [AInt32: {10}]
-- ASSIGN |PARTITIONED|
empty-tuple-source
-- EMPTY_TUPLE_SOURCE |PARTITIONED|
{noformat}
In the plan, "-- PRE_CLUSTERED_GROUP_BY[$$19] |PARTITIONED|" is the physical
operator for "group by ([$$30 := %0->$$19]) decor ([%0->$$18; %0->$$24])",
while the rest between the two is a nested subplan inside the logical group-by
operator.
{noformat}
group by ([$$30 := %0->$$19]) decor ([%0->$$18; %0->$$24]) {
aggregate [$$17] <- [function-call:
asterix:listify, Args:[function-call: asterix:open-record-constructor,
Args:[AString: {linenum2}, %0->$$25, AString: {count2}, %0->$$23]]]
-- AGGREGATE |LOCAL|
select (function-call: algebricks:not,
Args:[function-call: algebricks:is-missing, Args:[%0->$$21]])
-- STREAM_SELECT |LOCAL|
nested tuple source
-- NESTED_TUPLE_SOURCE |LOCAL|
}
-- PRE_CLUSTERED_GROUP_BY[$$19] |PARTITIONED|
{noformat}
> Preclustered Group by is displayed incorrectly in the web UI
> ------------------------------------------------------------
>
> Key: ASTERIXDB-1520
> URL: https://issues.apache.org/jira/browse/ASTERIXDB-1520
> Project: Apache AsterixDB
> Issue Type: Bug
> Components: Web Interface
> Reporter: Abdullah Alamoudi
> Assignee: Yingyi Bu
> Priority: Critical
>
> To reproduce, uncheck execute query, check show optimized plan and run the
> following:
> drop dataverse test if exists;
> create dataverse test;
> use dataverse test;
> create type AuxiliaryType as open {
> id:int32,
> num:int32
> }
> create type LineType as open {
> }
> create dataset Book(LineType) with meta(AuxiliaryType)
> primary key meta().id;
> create index NumIndex1 on Book(meta().num);
> create index NumIndex2 on Book(linenum:int32) enforced;
> create index NumIndex3 on Book(count1:int32) enforced;
> create index NumIndex4 on Book(count2:int32) enforced;
> for $t1 in dataset Book
> where $t1.linenum < 10
> order by $t1.linenum
> return {
> "linenum1": $t1.linenum,
> "count1":$t1.count1,
> "t2info": for $t2 in dataset Book
> where $t1.count1 /* +indexnl */= $t2.count2
> order by $t2.linenum
> return {"linenum2": $t2.linenum,
> "count2":$t2.count2}
> };
> -- PRE_CLUSTERED_GROUP_BY[$$19] |PARTITIONED|
> {
> -- AGGREGATE |LOCAL|
> -- STREAM_SELECT |LOCAL|
> -- NESTED_TUPLE_SOURCE |LOCAL|
> }
> will show incorrectly as:
> {
> -- AGGREGATE |LOCAL|
> -- STREAM_SELECT |LOCAL|
> -- NESTED_TUPLE_SOURCE |LOCAL|
> }
> -- PRE_CLUSTERED_GROUP_BY[$$19] |PARTITIONED|
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
