Join Tables Side by Side

Source: adopted from here

Introduction

In our data analysis, it is quite common to join two kdb tables side by side. Kdb/q provides multiple functions to join two tables. These functions provides different flavors of joining to meet various needs. See more details in Joins.

Question

Given two tables with the same number of rows, join them side by side. For example,

t1:([] c1:10?0D;c2:10?til 1000);
t2:([] c3:rand each 10#3.14f;c4:10?`3);

Joining them side by side gives:

c1                   c2  c3         c4
---------------------------------------
0D23:01:17.352842688 158 1.149706   mpo
0D00:24:01.519761085 382 2.204227   ghh
0D02:30:54.237189888 770 2.28275    lfh
0D16:46:37.401892393 655 1.50694    dna
0D16:59:18.329750597 156 3.100881   meg
0D12:43:13.668088316 401 1.253591   ifk
0D11:43:06.399342119 11  2.280672   eag
0D00:06:21.593725830 183 1.534317   idf
0D07:13:23.823950886 819 2.052295   acj
0D08:29:24.138744771 849 0.08825518 oji

Answer

Two approaches are suggested:

Method 1: Use the fill operator (^). The typical use case is to replace nulls with non-null values. It can also join tables side by side.

t1^t2

Method 2: Use join operator (,). It can join atoms, lists, dictionaries and tables. When used together with each ('), it can join tables column-wise.

t1,'t2