I'm trying to return a SUBSTR between two delimiters. The Problem is, that the second delimiter is non-unique in the whole string.
Example:
testurl.de/some:text/&[uniquedelimiter]=texttoextract:sometext:someothertext
So I want to extract the string between "&[uniquedelimiter]=" and the next ":" after the unique delimiter. The occurrence of ":" in the string is variable.
Right now I'm using a double SUBSTRING_INDEX function, but I get errors if the string contains more than one ":".
SELECT SUBSTRING_INDEX(SUBSTRING_INDEX(list_of_urls, '&[uniquedelimiter]=',-1),':')
FROM url_table;
I couldn't find a comparable solution on any Community.
Thank you so much for your help, Lars
This seems to be working, maybe you can sort it, not sure:
SELECT
SUBSTRING(
url,
(POSITION('&[uniquedelimiter]=' IN url) + LENGTH('&[uniquedelimiter]=')),
LOCATE(':', url, (POSITION('&[uniquedelimiter]=' IN url) + LENGTH('&[uniquedelimiter]='))) -
(POSITION('&[uniquedelimiter]=' IN url) + LENGTH('&[uniquedelimiter]='))
)
from bla
My strategy was like this:
SET @url = (SELECT url from bla limit 1);
SET @pos_uniq_deli_start = POSITION('&[uniquedelimiter]=' IN @url);
SET @uniq_deli_length = LENGTH('&[uniquedelimiter]=');
SET @pos_uniq_deli_end = @pos_uniq_deli_start + @uniq_deli_length;
-- LOCATE(char, field, offset)
SET @pos_colon_after_deli = LOCATE(':', @url, @pos_uniq_deli_end);
SET @delta_unqi_del_colon = @pos_colon_after_deli - @pos_uniq_deli_end;
-- SUBSTRING(field, start_pos, length)
SET @result = SUBSTRING(@url,@pos_uniq_deli_end, @delta_unqi_del_colon);
SELECT @result;