Truncate text by byte

From PostgreSQL wiki
Jump to navigationJump to search

Library Snippets

Truncate UTF-8 Text by byte width

Works with PostgreSQL

Any version

Written in

PL/pgSQLl

Depends on

Nothing


Several different ways to truncate a String/Text that is encoded in UTF-8 or other variable encoding method to specified byte width:



Method 1: author Laruenz Albe is an SQL implementation generating a record set containing all possible byte lengths to character length in a table then filters the result down

CREATE OR REPLACE FUNCTION get_prefix (
   string text,
   max_bytes bigint
) RETURNS text
   LANGUAGE sql STRICT AS
$$SELECT p
FROM (SELECT p.p, octet_length(p.p) AS len
      FROM generate_series(0, length($1)) AS len
         CROSS JOIN LATERAL substr($1, 1, len.len) AS p) AS q
WHERE len <= $2
ORDER BY len DESC
LIMIT 1$$;



Method 2: Author Zsheep is an PLpg/SQL implementation, iterating over the string 1 character at a time moving left to right until the string byte width is approximately the input width

Create or Replace function max_bytea_length(pchars text, bytea_length int)
RETURNS bytea
LANGUAGE plpgsql

COST 100
VOLATILE 
AS $BODY$

declare 
_i int;
_length_of_chars int;
_newchars text;
_testchars text;
begin

if octet_length(pchars::bytea) <= bytea_length then
   return pchars::bytea;
end if;
_i = least( octet_length(pchars)-4, bytea_length-4);
_length_of_chars =  char_length(pchars);
loop 
   _newchars= left(pchars, _i);
    _testchars = left(pchars, _i+1); 
  if octet_length(_testchars::bytea) > bytea_length or _i = _length_of_chars  then
     return _newchars::bytea;
  end if ;
  _i = _i+1;
end loop ;

end;
$BODY$



Method 3: Author Daniel Veritte is an SQL function that inspects the UTF-8 byte sequence looking for the high order bytea that is <= max byte width.

create function utf8_truncate(str text, len int) returns text
as $$
select case when octet_length(str) <= len then str
else (
   with bstr(s) as (select convert_to(str, 'UTF-8'))
   select
   case
   when len>=1 and (get_byte(s, len) & 192) <> 128
   then convert_from(substring(s, 1, len), 'UTF-8')
   else
     case
     when len>=2 and (get_byte(s, len-1) & 192) <> 128
     then convert_from(substring(s, 1, len-1), 'UTF-8')
     else
       case
       when len>=3 and (get_byte(s, len-2) & 192) <> 128
       then convert_from(substring(s, 1, len-2), 'UTF-8')
       else
         case
         when len>=4 and (get_byte(s, len-3) & 192) <> 128
         then convert_from(substring(s, 1, len-3), 'UTF-8')
         else ''
         end
       end
     end
   end
 from bstr)
   end;
$$ language sql strict immutable parallel safe;


Examples of usage

SELECT get_prefix('abc€', 4);

select max_bytea_length('i ♥ u function changed it to be faster', 56);

select utf8_truncate('method 3 €', 9);