KDB/Q分区表筛选date列及所有_test后缀列的方法问询

Got it, let's fix this for you. The issue with your original delete all except approach is that kdb+ doesn't support wildcard patterns like *_test directly in the except clause—you need to explicitly generate the list of columns you want to keep first.

Step-by-Step Solution #

First, let's get the list of columns that end with _test using cols (to fetch all column names) and like (to filter by the suffix pattern):

test_cols: cols[t] where cols[t] like "*_test"

For your sample table, this will return ,a_testb_test.

Then, select the date column along with these filtered columns in one query:

select date, test_cols from t where date=2019.01.01

One-Liner Version #

If you prefer a more concise approach, combine the two steps into a single line:

select date, (cols[t] where cols[t] like "*_test") from t where date=2019.01.01

Alternative Using Indexing #

You can also use kdb+'s indexing operator ! to extract the desired columns directly:

t[`date, (cols[t] where cols[t] like "*_test")] where t.date=2019.01.01

Verification with Your Sample Table #

Let's test this with your example table:

t:([]date:3#2019.01.01; a:1 2 3; a_test:2 3 4; b_test:3 4 5; c: 6 7 8);

Running any of the above queries will give you the expected output:

date       a_test b_test
------------------------
2019.01.01 2      3
2019.01.01 3      4
2019.01.01 4      5

This method scales perfectly even if you have 100+ _test columns—no need to manually list each one!

内容的提问来源于stack exchange，提问作者Utsav