This is a patch to fix special char (accents...) bugs (bad charset, encoding) for PostgreSQL in PgSQL.

gistfile1.sql

DO $$
DECLARE
    tables CURSOR FOR (SELECT TABLE_NAME as "name", COLUMN_NAME as "column" FROM INFORMATION_SCHEMA.COLUMNS WHERE (DATA_TYPE LIKE '%char%' OR DATA_TYPE = 'text') AND TABLE_SCHEMA = 'public');
BEGIN
    FOR table_record IN tables LOOP
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''é'',''é'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''ä'',''ä'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''ö'',''ö'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''Ã¥'',''å'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''è'',''è'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''Ç'',''ç'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''é'',''é'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''ů'',''ů'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''ü'',''ü'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''ß'',''ß'')';
		EXECUTE 'UPDATE public.' || table_record.name ||' SET ' || table_record.column || ' = REPLACE(' || table_record.column || ', ''Ã…'',''Å'')';
    END LOOP;
END$$;