Numeric column names in PostgreSQL

In the Udemy PostgreSQL Bootcamp : Go From Beginner to Advanced, 60+hours course, there was a lesson where we wrote a function to output some SQL to enable dynamic pivot tables:
pass the table_name, vertical_header_column, horizontal_header_column, value_column, and data_type and any table can be “pivoted”. Pretty nifty.

But… it was hideously difficult to transcribe the code from video source.

And, the horizontal headers were year values, so the instructor prepended an underscore (_) to the column names since databases do not (normally) like a column named 2020.

The underscore prepended to field names is not necessary

One can create a table where column names start with / are completely numeric characters:

\echo NOTE: field names CAN start with numbers, like I tried to do above:

DROP TABLE IF EXISTS crosstabs.column_name_numeric;

CREATE TABLE crosstabs.column_name_numeric (
  "123" integer
  );
\x on
\d crosstabs.column_name_numeric

       Table "crosstabs.column_name_numeric"
 Column |  Type   | Collation | Nullable | Default
--------+---------+-----------+----------+---------
 123    | integer |           |          |

This way, with a lot of work at trying to figure out the nested quotes in the dynsql1 statement, we can get the following headers that are purely numeric:

 location  | 2012 | 2013 | 2014 | 2015 | 2016 | 2017
-----------+------+------+------+------+------+------
 Dubai     |  112 |  116 |  111 |   80 |  110 |  102
 France    |  114 |  111 |  113 |   85 |  120 |  110
 Germany   |  153 |  120 |  136 |  128 |  138 |  113
 London    |  116 |  138 |  137 |  101 |  117 |  110
 Malaysia  |   53 |   77 |   78 |   50 |   52 |   43
 Qatar     |   86 |  116 |  111 |   83 |  102 |   89
 Singapore |  124 |  132 |  142 |  111 |  106 |  110
 Sydney    |   98 |  120 |  110 |   93 |  122 |  115
 Tokyo     |   47 |   63 |   68 |   53 |   54 |   43
(9 rows)

So, yes, it is possible to have numeric column names in PostgreSQL.