Why I get error when using TO_DATE() while converting from VARCHAR to Date data type - sql

Question

I am currently stuck in error which I don't understand clearly.
ORA-01843: not a valid month
01843. 00000 - "not a valid month"
*Cause:
*Action:
So far I check a couple of post here in Stackoverflow but doesn't help me so much.
I have query something like
SELECT * FROM (
SELECT uta.StartDate,uta.EndDate
FROM user_timesheets_absence uta
LEFT JOIN users u
ON u.UserID = uta.UserID
AND uta.Approved = '0'
AND
((
'2020-01-30' >= TO_DATE(uta.StartDate,'YYYY-MM-DD')
AND
'2020-02-06' <= TO_DATE(uta.EndDate,'YYYY-MM-DD')
)
OR
(
'2020-01-30' <= TO_DATE(uta.StartDate,'YYYY-MM-DD')
AND
'2020-01-30' >= TO_DATE(uta.StartDate,'YYYY-MM-DD')
))
--GROUP BY uta.UserAbsenceID
UNION
SELECT ut.DATE_ AS StartDate,
ut.DATE_ AS EndDate
FROM user_timesheets ut
INNER JOIN moments m ON
m.UserTimesheetsID = ut.UserTimesheetsID
WHERE
TO_DATE(ut.DATE_,'YYYY-MM-DD') BETWEEN '2020-01-21' AND '2020-01-30' + SYSTIMESTAMP + 1
AND ut.user_id = 1
) a
Here is the problem which StartDate and EndDate filled are VARCHAR2
I try using 'TO_TIMESTAMP' and try to change format to YYYY/MM/DD but doesn't work.
I have to be honest that I don't have any idea so far what I made wrong here.
Where I made mistake ? What is wrong here ?
UPDATE
As GMB from answer said that using validate_conversion() gives meall invalid date strings
Result
And so far from query above, when I run it I get following error
ORA-01861: literal does not match format string
01861. 00000 - "literal does not match format string"
*Cause: Literals in the input must be the same length as literals in
the format string (with the exception of leading whitespace). If the
"FX" modifier has been toggled on, the literal must match exactly,
with no extra whitespace.
*Action: Correct the format string to match the literal.

Answer 1

Some of the strings in your column are invalid dates. Starting version 12.2, it is easy to exhibit them with function validate_conversion() clause of to_date(). The following query gives you all invalid date strings:
select date_
from user_timesheets
where validate_conversion(date_ as date, 'YYYY/MM/DD') = 0
You can then fix your data.
You should consider using the date datatype to store dates, so that integrity is enforced at the time when the data is written.
In the meantime, let me suggest, however, to optimize the filtering logic: the formats of your strings make it possible to use direct filtering (without prior conversion to a date): this is more efficient, since it does not requires converting the entire column before filtering (on says that the predicate is SARGeable). Typically, you would replace something like:
to_date(uta.startdate,'yyyy/mm/dd') <= '2020-01-30'
With:
ut.startdate <= '2020/01/30'
Or if you are filtering agains the current system date:
ut.startdate <= to_char(sysdate, 'YYYY/MM/DD')

Answer 2

Not all startdate and enddate columns have values in YYYY/MM/DD format or - if they do - they are invalid.
For example: values like 2020/A8/43 or ab3f/xy/2i or 2020/15/02 and similar are invalid.
Error says that MM part of those strings isn't in range between 01 and 12.

Answer 3

First you should be storing date values using date, not strings. One possibility is that these actually are dates and you don't need to reconvert them. Without sample data, it is hard to say. But you should put your effort into fixing the data.
Second, date constants in Oracle should look like:
DATE '2020-01-30'
So:
DATE '2020-01-30' >= TO_DATE(uta.StartDate,'YYYY/MM/DD')