Split string by number of character in Oracle SQL - sql

Question

I am looking to split string at 70 character and I found below code from this link which is slightly different than I am looking. Below code is splitting character at the max of 70 using the last space as break but what I am looking is without the last space as break. I tried to make changes from this code but honestly I am still learning the regexp_substr function and unable to figure out where the last space is considered. It will be very helpful if someone guide me through this. Thanks
Code:
with val as (
select 'I am also looking for similar solution but slightly different where the split occurs at 70 character and not considering last space.' str
from dual
), words as (
select regexp_substr(str, '[^ ]+', 1, level) w
from val
connect by regexp_substr(str, '[^ ]+', 1, level) is not null
), grps as (
select * from words
match_recognize (
measures
match_number () grp,
count ( text.* ) word#
all rows per match
pattern ( init text* )
define
text as sum ( length ( w ) ) + count ( text.* ) <= 70
)
)
select listagg ( w, ' ' )
within group ( order by word# ) split_strs
from grps
group by grp
order by grp;
Result from above code:
What I am looking to achieve is,

Answer 1

You need a Hierarchical query containing SUBSTR() function such as
SELECT level-1 AS "Row_Number",
SUBSTR(str,70*(level-1)+1,70*level)
FROM val
CONNECT BY level <= CEIL(LENGTH(str)/70)
Demo

Answer 2

For exactly 70 chars you can use SUBSTR instead of regexp_substr.
e.g. first 70 chars = SUBSTR(val,1,70)

Answer 3

This one done the trick, thanks those responded before I get to this.
with val as (
select 'I am also looking for similar solution but slightly different where the split occurs at 70 character and not considering last space.' str
from dual
), words as (
select regexp_substr(str, '(.{1})', 1, level) w
from val
connect by regexp_substr(str, '(.{1})', 1, level) is not null
), grps as (
select * from words
match_recognize (
measures
match_number () grp,
count ( text.* ) word#
all rows per match
pattern ( init text* )
define
text as sum ( length ( w ) ) <= 70
)
)
select listagg ( w, '' )
within group ( order by word# ) split_strs
from grps
group by grp
order by grp
;