thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-24 08:32:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

better tests

Browse Source
This commit is contained in:
sundy-li 2019-05-20 12:31:52 +08:00
parent 6843254ce4
commit 66c60c2985
5 changed files with 55 additions and 53 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
dbms/src/AggregateFunctions/AggregateFunctionSequenceMatch.h
View File

@ -194,10 +194,10 @@ private:
struct PatternAction final
{
PatternActionType type;
std::uint32_t extra;
std::uint64_t extra;
PatternAction() = default;
PatternAction(const PatternActionType type, const std::uint32_t extra = 0) : type{type}, extra{extra} {}
PatternAction(const PatternActionType type, const std::uint64_t extra = 0) : type{type}, extra{extra} {}
};
static constexpr size_t bytes_on_stack = 64;
@ -432,7 +432,7 @@ protected:
}
else if (action_it->type == PatternActionType::TimeLessOrEqual)
{
if (events_it->first <= static_cast<UInt64>(base_it->first + action_it->extra))
if (events_it->first <= base_it->first + action_it->extra)
{
/// condition satisfied, move onto next action
back_stack.emplace(action_it, events_it, base_it);
@ -444,7 +444,7 @@ protected:
}
else if (action_it->type == PatternActionType::TimeLess)
{
if (events_it->first < static_cast<UInt64>(base_it->first < action_it->extra))
if (events_it->first < base_it->first + action_it->extra)
{
back_stack.emplace(action_it, events_it, base_it);
base_it = events_it;
@ -455,7 +455,7 @@ protected:
}
else if (action_it->type == PatternActionType::TimeGreaterOrEqual)
{
if (events_it->first >= static_cast<UInt64>(base_it->first + action_it->extra))
if (events_it->first >= base_it->first + action_it->extra)
{
back_stack.emplace(action_it, events_it, base_it);
base_it = events_it;
@ -466,7 +466,7 @@ protected:
}
else if (action_it->type == PatternActionType::TimeGreater)
{
if (events_it->first > static_cast<UInt64>(base_it->first + action_it->extra))
if (events_it->first > base_it->first + action_it->extra)
{
back_stack.emplace(action_it, events_it, base_it);
base_it = events_it;

84
dbms/tests/queries/0_stateless/00222_sequence_aggregate_function_family.sql
View File

@ -4,48 +4,48 @@ create table sequence_test (time UInt32, data UInt8) engine=Memory;
insert into sequence_test values (0,0),(1,0),(2,0),(3,0),(4,1),(5,2),(6,0),(7,0),(8,0),(9,0),(10,1),(11,1);
select 1 = sequenceMatch('')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('.')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('.*')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?4)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?1)(?1)(?1)(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)(?1)(?1)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t>10)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?1)(?t>11)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t<11)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t<3)(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t<=2)(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?1)(?t<2)(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?2)(?t>=7)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?2)(?t>7)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?2)(?3)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('.')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('.*')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?4)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?1)(?1)(?1)(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?1)(?1)(?1)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t>10)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?1)(?t>11)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t<11)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t<3)(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?1)(?t<=2)(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?1)(?t<2)(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?2)(?t>=7)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceMatch('(?2)(?t>7)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceMatch('(?2)(?3)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select count() = sequenceCount('')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select count() = sequenceCount('.')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select count() = sequenceCount('.*')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 8 = sequenceCount('(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 3 = sequenceCount('(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?4)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 4 = sequenceCount('(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?1)(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?1)(?1)(?1)(?1)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?1)(?1)(?1)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?1)(?t>10)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?1)(?t>11)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?t<11)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?1)(?t<3)(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?1)(?t<=2)(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?1)(?t<2)(?3)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?2)(?t>=7)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?2)(?t>7)(?2)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?2)(?3)(?1)')(toDateTime(time), data = 0, data = 1, data = 2, data = 3) from sequence_test;
select count() = sequenceCount('')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select count() = sequenceCount('.')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select count() = sequenceCount('.*')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 8 = sequenceCount('(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 3 = sequenceCount('(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?4)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 4 = sequenceCount('(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?1)(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?1)(?1)(?1)(?1)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?1)(?1)(?1)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?1)(?t>10)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?1)(?t>11)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 2 = sequenceCount('(?1)(?t<11)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?1)(?t<3)(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?1)(?t<=2)(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?1)(?t<2)(?3)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?2)(?t>=7)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 0 = sequenceCount('(?2)(?t>7)(?2)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
select 1 = sequenceCount('(?2)(?3)(?1)')(time, data = 0, data = 1, data = 2, data = 3) from sequence_test;
drop table sequence_test;

1
dbms/tests/queries/0_stateless/00823_sequence_match_dfa.reference
View File

@ -1,2 +1,3 @@
ABC
ABBC
CD

7
dbms/tests/queries/0_stateless/00823_sequence_match_dfa.sql
View File

@ -5,12 +5,13 @@ DROP TABLE IF EXISTS sequence;
CREATE TABLE sequence
(
userID UInt64,
eventType Enum8('A' = 1, 'B' = 2, 'C' = 3),
eventType Enum8('A' = 1, 'B' = 2, 'C' = 3, 'D' = 4),
EventTime UInt64
)
ENGINE = Memory;
INSERT INTO sequence SELECT 1, number = 0 ? 'A' : (number < 1000000 ? 'B' : 'C'), number FROM numbers(1000001);
INSERT INTO sequence SELECT 1, 'D', 1e14;
SELECT 'ABC'
FROM sequence
@ -27,9 +28,9 @@ FROM sequence
GROUP BY userID
HAVING sequenceMatch('(?1).*(?2).*(?3).*(?4)')(EventTime, eventType = 'A', eventType = 'B', eventType = 'B',eventType = 'C');
SELECT 'CBA'
SELECT 'CD'
FROM sequence
GROUP BY userID
HAVING sequenceMatch('(?1).*(?2).*(?3)')(EventTime, eventType = 'C', eventType = 'B', eventType = 'A');
HAVING sequenceMatch('(?1)(?t>=10000000000000)(?2)')(EventTime, eventType = 'C', eventType = 'D');
DROP TABLE sequence;

4
docs/en/query_language/agg_functions/parametric_functions.md
View File

@ -8,7 +8,7 @@ Pattern matching for event chains.
`pattern` is a string containing a pattern to match. The pattern is similar to a regular expression.
`time` is the time of the event with the DateTime type.
`time` is the time of the event, type support: `Date`,`DateTime`, and other unsigned integer types.
`cond1`, `cond2` ... is from one to 32 arguments of type UInt8 that indicate whether a certain condition was met for the event.
@ -59,7 +59,7 @@ windowFunnel(window)(timestamp, cond1, cond2, cond3, ...)
**Parameters:**
- `window` — Length of the sliding window in seconds.
- `timestamp` — Name of the column containing the timestamp. Data type: [DateTime](../../data_types/datetime.md) or [UInt32](../../data_types/int_uint.md).
- `timestamp` — Name of the column containing the timestamp. Data type support: `Date`,`DateTime`, and other unsigned integer types(Note that though timestamp support `UInt64` type, there is a limitation it's value can't overflow maximum of Int64, which is 2^63 - 1).
- `cond1`, `cond2`... — Conditions or data describing the chain of events. Data type: `UInt8`. Values can be 0 or 1.
**Algorithm**