Postgres 15

By using db<>fiddle, you agree to license everything you submit by Creative Commons CC0.

CREATE TABLE tbl (id int, doc_text text);

INSERT INTO tbl
VALUES (1, 'The fat rat ate a cat.')

UNION ALL
SELECT 2, 'big but repetitive: ' || repeat('bla ', 30)

UNION ALL
SELECT 3, 'big and not repetitive: ' || string_agg('word' || g, ' ')
FROM   generate_series (1, 30) g;

CREATE TABLE



INSERT 0 3


SELECT id
     , length(to_tsvector('simple', doc_text))
     , left(doc_text, 50) || CASE WHEN length(doc_text) > 50 THEN ' ...' ELSE '' END
FROM   tbl;


id
length
?column?


1
6
The fat rat ate a cat.

2
4
big but repetitive: bla bla bla bla bla bla bla bl ...

3
34
big and not repetitive: word1 word2 word3 word4 wo ...



SELECT 3


-- debug:
-- this will report last row for the demo
DO
$do$
DECLARE
   r record;
BEGIN
   FOR r IN
      SELECT id, doc_text FROM tbl
   LOOP
      PERFORM to_tsvector('simple', r.doc_text);
   END LOOP;
   
   -- My small sample values don't hit the limit, so force it:
    PERFORM to_tsvector('simple', string_agg('longwordnr' || g, ' ')) FROM generate_series (1, 100000) g;  
EXCEPTION
   WHEN program_limit_exceeded THEN
      RAISE EXCEPTION 'Row with this "id" exceeds ts_vector length: %', r.id;
END
$do$;

ERROR:  Row with this "id" exceeds ts_vector length: 3
CONTEXT:  PL/pgSQL function inline_code_block line 15 at RAISE

id	length	?column?
1	6	The fat rat ate a cat.
2	4	big but repetitive: bla bla bla bla bla bla bla bl ...
3	34	big and not repetitive: word1 word2 word3 word4 wo ...