bar

bar(X, interval, [closed='left'])

X is an integral/temporal scalar or vector.

interval is an integral/DURATION type scalar greater than 0 or a vector of the same length as X.

When interval is of type DURATION, the following time units are supported (case-sensitive): w, d, H, m, s, ms, us, ns.

Note: As time units y and M are not supported in interval, to group X by year or month, convert the data format of X with function month or year. Specify the interval as an integer for calculation. You can refer to Example 2.

closed (optional) is a string which can take 'left' (default) or 'right', indicating whether an element of X that is divisible by interval is the left boundary (the first element of the group) or the right boundary (the last element of the group) of a group.

• closed = 'left': X-(X % interval), indicating that the value with a remainder of 0 is specified as the left boundary of a group.

• closed = 'right': iif((X % interval) == 0, X, X + (interval-(X % interval))), indicating that the value with a remainder of 0 is specified as the right boundary of a group.

bar can group X based on the length specified by interval. Return a vector with the same length as X.

bar(100,3);               // 100-(100%3)=100-1=99
// output
99

bar(0..15, 3)
// output
[0,0,0,3,3,3,6,6,6,9,9,9,12,12,12,15]

x=[7,4,5,8,9,3,3,5,2,6,12,1,0,-5,32]
bar(x, 5)
// output
[5,0,5,5,5,0,0,5,0,5,10,0,0,-5,30]

t=table(2021.01.01T01:00:00..2021.01.01T01:00:29 as time, rand(1.0, 30) as x)
select max(x) from t group by bar(time,5s)

In the following example, to group data by every 3 months, convert X with the month function, and specify duration as an integer in bar.

t=table(take(2018.01.01T01:00:00+1..10,10) join take(2018.02.01T02:00:00+1..10,10) join take(2018.03.01T08:00:00+1..10,10) join take(2018.04.01T08:00:00+1..10,10) join take(2018.05.01T08:00:00+1..10, 10) as time, rand(1.0, 50) as x)
select max(x) from t group by bar(month(time), 3);

The following example groups data by week and calculates the maximum values for each week. Depending on parameter closed, the results are different.

t=table(2022.01.01 + 1..20  as time, rand(100, 20) as x)

select max(x) from t group by bar(time, 7d);

print  select max(x) from t group by bar(time, 7d, closed='right');

When calculating 1-minute OHLC bars, the data type needs to be converted to LONG if n needs to be converted to NANOTIMESTAMP, otherwise an integer overflow will occur.

n = 1000000
nano = (09:30:00.000000000 + rand(long(6.5*60*60*1000000000), n)).sort!()
price = 100+cumsum(rand(0.02, n)-0.01)
volume = rand(1000, n)
symbol = rand(`600519`000001`600000`601766, n)
tradeNano = table(symbol, nano, price, volume).sortBy!(`symbol`nano)
undef(`nano`price`volume`symbol)
barMinutes = 7
itv = barMinutes*60*long(1000000000)

OHLC_nano=select first(price) as open, max(price) as high, min(price) as low, last(price) as close, sum(volume) as volume from tradeNano group by symbol, bar(nano, itv) as barStart

Related function: dailyAlignedBar