Untitled query #26249

SQL

USE enwiki_p;
SELECT
    SUBSTRING(el_index FROM 1 FOR 60) AS index_prefix,
    COUNT(*)
FROM externallinks
WHERE
	-- Only links from main namespace articles.
	el_from_namespace = 0
	-- Only http, https, implicit https (//) and ftp. el_index is already lowercase.
	AND (el_index LIKE 'http://%' OR el_index LIKE 'https://%' OR el_index LIKE '//%' OR el_index LIKE 'ftp://%')
	-- Get only fairly well-formed URLs and exclude IPv4.
	AND SUBSTRING(el_to FROM 1 FOR 40) REGEXP '(?i)^(?:https?:|ftp:)?//(?:[-_0-9A-Za-z]+\\.)+[A-Za-z]+(?:[/:?].*)?$'
GROUP BY index_prefix
LIMIT 100;
SQL

Checking query status...
