latestIndexedTable

Syntax

latestIndexedTable(keyColumns, timeColumn, [X1], [X2], .....)

or

latestIndexedTable(keyColumns, timeColumn, capacity:size, colNames, colTypes)

or

latestIndexedTable(keyColumns, timeColumn, table)

Arguments

Compared to the indexedTable function, latestIndexedTable includes an additional parameter called timeColumn to specify the temporal dimension of the data.

timeColumn is a string that specifies the name of the time column which can be either of integral or temporal data type.

Details

Create an indexed table, which is a special type of in-memory table with primary key. The primary key can be one column or multiple columns. Compared to the indexedTable, latestIndexedTable adds a time column to determine whether to update records.

When a new record is appended to the indexed table, if its timestamp is smaller than that of the existing row which has the same primary key, it does not overwrite the existing row. latestIndexedTable deduplicates records with the same primary key based on the time column, which affects its writing performance (relatively slow compared with indexedTable).

Examples

Example 1. Create an indexed table.

Scenario 1:

sym=`A`B`C`D`E
id=5 4 3 2 1
val=52 64 25 48 71
timeCol = 2022.12.07T00:00:00.001+0..4
t=latestIndexedTable(`sym`id,`timeCol,sym,id,timeCol,val)
t;

Scenario 2:

t=latestIndexedTable(`sym`id,`timeCol, 1:0,`sym`id`timeCol`val,[SYMBOL,INT,TIMESTAMP, INT])
insert into t values(`A`B`C`D`E,5 4 3 2 1,2022.12.07T00:00:00.001+0..4,52 64 25 48 71);

Scenario 3:

tmp=table(sym, id, timeCol, val)
t=latestIndexedTable(`sym`id, `timeCol, tmp);

Example 2. Update an indexed table.

If the new row has the same primary key value as an existing row, whether to update the record is determined by the time column.

insert into t values(`A`A`E,5 5 1, 2022.12.07T00:00:00.001 2022.12.07T00:00:00.007 2022.12.07T00:00:00.003, 44 66 28);
t;