AustrianToast/HDB
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

add sql

Browse Source
This commit is contained in:
AustrianToast
2024-11-01 16:09:16 +01:00
parent 1e7f878a64
commit b2429246e4
10 changed files with 234 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
sql/convert_artist_name.sql Normal file
View File

@ -0,0 +1,29 @@
CREATE OR REPLACE FUNCTION convert_artist_name(artist_name text) RETURNS text AS $$
BEGIN
artist_name = trim(artist_name);
-- partially deal with url encoding
artist_name = regexp_replace(artist_name, '%(20|21|22|23|24|26|27|28|29|2A|2C|2E|2F|3B|3C|3E|3F|5B|5C|5D|5E|60|7B|7C|7D|7E)|artist', '', 'g');
artist_name = regexp_replace(artist_name, '%25', '%', 'g');
artist_name = regexp_replace(artist_name, '%2B', '+', 'g');
artist_name = regexp_replace(artist_name, '%2D', '-', 'g');
artist_name = regexp_replace(artist_name, '%2E', '.', 'g');
artist_name = regexp_replace(artist_name, '%2F', '/', 'g');
artist_name = regexp_replace(artist_name, '%3A', ':', 'g');
artist_name = regexp_replace(artist_name, '%3D', '=', 'g');
artist_name = regexp_replace(artist_name, '%40', '@', 'g');
artist_name = regexp_replace(artist_name, '%5F', '_', 'g');
-- check if any other url encoding still exists
IF regexp_count(artist_name, '%[0-9A-Fa-f]{2}') <> 0 THEN
RAISE EXCEPTION 'Artist % contains unconvertable percent-encoded Unicode', artist_name;
END IF;
artist_name = rtrim(artist_name, '-_');
-- lower has to be last, because it can and will fuck everything else up
artist_name = lower(artist_name);
RETURN artist_name;
END;
$$ LANGUAGE plpgsql;