Question

Imagine that you have a field with data like this (each line represents the data in that field for different rows):

28746: Porto Alegre
28302: Osasco
28797: Praia Grande

But sometimes there is more than one number in the field:

26755: Floriano Peixoto,26754: Floriano 25582: Bela Vista
25579: Bela Vista 25581: Bela Vista De Goiás,26867: Goiás

I know this data it’s not normalized, but that’s not the issue I want to solve.

I was wondering if it’s possible to make a query that shows only the rows that have more than one number in that field.

Any suggestions on how to achieve that?

UPDATE

From Dark_a answer I tried this:

SELECT data_c1
  FROM test_data
  WHERE INSTR(data_c1,':',1,2) <> 0;

But it seems that INSTR only accepts to arguments:

http://dev.mysql.com/doc/refman/5.5/en/string-functions.html

However if I use this other query:

SELECT data_c1
  FROM test_data
  WHERE INSTR(data_c1,',') <> 0;

It seems to work fine, because the data in fields with more than one number, it’s separeted by a comma.

Answer 1

You can try this:

SELECT data_c1
  FROM test_data
 WHERE INSTR(data_c1,':',1,2) <> 0;

If comma is used in text. if not then is better to check for comma:

SELECT data_c1
  FROM test_data
 WHERE INSTR(data_c1,',',1,2) <> 0;