使用正则表达式从SQL文件中提取表

Question

I want to be able to extract a query's source tables using a regular expression. This will help me build a script to test dependencies without having to hit the cluster.

Example:

use datalab2;
select 
   *
from hello as table1
left join (
  select 
     *
  from ${hiveconf:varable_db}.events
  where  test=1
) as table2 on 
  table2.test = table1.test
inner join datalab3.table3 as table3 on 
  table3.test = table1.test;

Would Return:

<br>
hello <br>
${hiveconf:varable_db}.events <br>
datalab3.table3 <br>

Answer 1

Solved it using the following two regex expressions

1: "(with\\s+|,\\s*)([a-zA-Z0-9_]+)\\s+as\\s*\\("

2: "(?<=[^a-zA-Z0-9\\_]FROM|[^a-zA-Z0-9\\_]JOIN)(\\s+|\\s+[^\\(])([a-zA-Z0-9\\.\\_\\{\\}\\$\\:]+)"

The first one gathers a list of all the with statement names
Example: https://regex101.com/r/sX5hZ2/25

The second one gets a list of all the tables used in the query and with statements
Example: https://regex101.com/r/sX5hZ2/26

The distinct items from the second list will be all the tables that the query uses