thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-23 08:02:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #48603 from ilejn/sequence_state_fix

Browse Source 
sequence state fix
This commit is contained in:
Nikolai Kochetov 2023-05-02 21:25:27 +02:00 committed by GitHub
commit 6beff3bd44
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 45 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
src/AggregateFunctions/AggregateFunctionSequenceMatch.h
View File

@ -50,7 +50,7 @@ struct AggregateFunctionSequenceMatchData final
bool sorted = true;
PODArrayWithStackMemory<TimestampEvents, 64> events_list;
/// sequenceMatch conditions met at least once in events_list
std::bitset<max_events> conditions_met;
Events conditions_met;
void add(const Timestamp timestamp, const Events & events)
{
@ -101,6 +101,11 @@ struct AggregateFunctionSequenceMatchData final
size_t size;
readBinary(size, buf);
/// If we lose these flags, functionality is broken
/// If we serialize/deserialize these flags, we have compatibility issues
/// If we set these flags to 1, we have a minor performance penalty, which seems acceptable
conditions_met.set();
events_list.clear();
events_list.reserve(size);

3
tests/queries/0_stateless/02713_sequence_match_serialization_fix.reference Normal file
View File

@ -0,0 +1,3 @@
serialized state is not used 1
serialized state is used 1
via Distributed 1

36
tests/queries/0_stateless/02713_sequence_match_serialization_fix.sql Normal file
View File

@ -0,0 +1,36 @@
DROP TABLE IF EXISTS 02713_seqt;
DROP TABLE IF EXISTS 02713_seqt_distr;
SELECT
'serialized state is not used', sequenceMatch('(?1)(?2)')(time, number_ = 1, number_ = 0) AS seq
FROM
(
SELECT
number AS time,
number % 2 AS number_
FROM numbers_mt(100)
);
CREATE TABLE 02713_seqt
ENGINE = MergeTree
ORDER BY n AS
SELECT
sequenceMatchState('(?1)(?2)')(time, number_ = 1, number_ = 0) AS seq,
1 AS n
FROM
(
SELECT
number AS time,
number % 2 AS number_
FROM numbers_mt(100)
);
SELECT 'serialized state is used', sequenceMatchMerge('(?1)(?2)')(seq) AS seq
FROM 02713_seqt;
CREATE TABLE 02713_seqt_distr ( seq AggregateFunction(sequenceMatch('(?1)(?2)'), UInt64, UInt8, UInt8) , n UInt8) ENGINE = Distributed(test_shard_localhost, currentDatabase(), '02713_seqt');
SELECT 'via Distributed', sequenceMatchMerge('(?1)(?2)')(seq) AS seq FROM 02713_seqt_distr;