Say I issue:
select date_field from table1;
date_field is like '25.11.2009'
I will try to change the positions of date fields with the month and vice versa. (of course for days > 12 some manipulations)
TO_DATE( MOD(SUBSTR(TO_CHAR(a.A_DATE, 'DD.MM.YYYY'), 4, 2), 12) || '.' ||
SUBSTR(TO_CHAR(a.A_DATE, 'DD.MM.YYYY'), 1, 2) ||
SUBSTR(TO_CHAR(a.A_DATE, 'DD.MM.YYYY'), 6, 4),
'DD.MM.YYYY')
THE THING IS THAT THE VALUE RETURNED FROM MOD() function is a number, i.e. for 01.07.2009 --> I get 1 for date, not '01' as expected. Later on I cannot get the date.
Is there a shortcut solution to my problem?
I suspect you need to seriously reconsider what you are trying to do.
I think what you started with is that you want to simply change the formatting of the date, e.g. change '25.11.2009' to '11.25.2009'.
If date_field is actually a DATE type, there is no inherent formatting stored in the field. It is a numeric value representing a specific date and time. It is formatted into text when you SELECT it in SQLPlus or some other tool, but that formatting is not stored in the table.
If you want to view the date in a particular format, you use the TO_CHAR function to force it. You can also set a default format model for a single session, a single client, or the whole database using NLS_DATE_FORMAT.
I used:
CASE MOD(SUBSTR(TO_CHAR(a.birthday, 'DD.MM.YYYY'), 1, 2), 12)
WHEN 1 THEN '01'
WHEN 2 THEN '02'
WHEN 3 THEN '03'
WHEN 4 THEN '04'
WHEN 5 THEN '05'
WHEN 6 THEN '06'
WHEN 7 THEN '07'
WHEN 8 THEN '08'
WHEN 9 THEN '09'
WHEN 10 THEN '10'
WHEN 11 THEN '11'
WHEN 12 THEN '12'
END
not very elegant but works :)
yli> "I am somehow anonymising the date, actually Hire Date of employees, to be used in a test environment."
So here was your actual requirement. If you want to change some dates, you don't need to use the date/string conversion functions (TO_CHAR/TO_DATE) at all. Oracle supports arithmetic operations on date values directly.
If you want to randomize your dates why not just use something like:
select date_field + dbms_random.value * 100 from table1;
Does this work?
TO_DATE(TO_CHAR(a.A_DATE, 'DD/MM/YYYY'), 'MM/DD/YYYY')
Have you tried TO_CHAR(<date>, '<format>')?
Related
I want to use (between, and, or) conditions in a check constraint at the same time. For example:
Alter table called_no add constraint called_no_chck
check ("call_time" between '01:00:00 AM' and '12:59:59 AM' or
between '01:00:00 PM' and '12:59:59 PM')
You certainly may use multiple conditions in a single check constraint.
Your attempt fails because you wrote an invalid condition. The same condition would fail if it was in the where clause of a query, or in any other place where you need a condition.
The correct way to check that a date-time is between A and B or between C and D looks like this:
...check (call_time between A and B OR call_time between C and D)
^^^^^^^^^
You understand the shorthand, where you don't repeat the column name (call_time) before the second between. I understand it too. A computer does not - they are not human. OR must appear between two conditions. A condition such as between must be applied to something, and you can't leave that "something" out and expect the computer to understand it's the same "something" you already used in another condition.
There are other mistakes in your condition though. 12:59:59 AM is about one hour after midnight; it's not about one hour after noon, as you seem to believe. Nothing is ever between 1 AM and 12:59:59 AM (of the same day), because 1 AM is after 12:59:59 AM.
It's also not clear what the data type of your column is. If it's date (as it almost surely should be), you can't just compare it to strings (as you have in your condition). And, you can't simply compare a date-time to something that's just time-of-day.
All these mistakes, though, are out of the scope of your question as you asked it (which was: Can you use "compound" conditions in a constraint; the answer, again, is YES).
Do not use a string column.
If you just want a time without a date, use an INTERVAL DAY(0) TO SECOND(0) column.
If you want a date-time column with accuracy down to whole seconds (and no time-zone) then use a DATE column (which always has the components year, month, day, hour, minute and seconds).
With either of those two options, you do not need a CHECK constraint as it will not accept invalid times.
If you want to format the value using the 12-hour clock then you can use a virtual column.
CREATE TABLE table_name (
time INTERVAL DAY(0) TO SECOND(0),
datetime DATE,
formatted_time VARCHAR2(11)
GENERATED ALWAYS AS (TO_CHAR(DATE '1970-01-01' + time, 'HH12:MI:SS AM')),
formatted_datetime VARCHAR2(11)
GENERATED ALWAYS AS (TO_CHAR(datetime, 'HH12:MI:SS AM'))
);
INSERT INTO table_name (time, datetime)
SELECT INTERVAL '00:12:34' HOUR TO SECOND, SYSDATE FROM DUAL UNION ALL
SELECT INTERVAL '12:34:56' HOUR TO SECOND, SYSDATE + 0.5 FROM DUAL;
Then:
SELECT * FROM table_name;
Outputs:
TIME
DATETIME
FORMATTED_TIME
FORMATTED_DATETIME
+0 00:12:34
2021-12-18 15:37:34
12:12:34 AM
03:35:47 PM
+0 12:34:56
2021-12-19 03:37:34
12:34:56 PM
03:35:47 AM
If you are storing it in a string (don't) then, yes, you can use multiple conditions in a single constraint:
CREATE TABLE table_name (
time VARCHAR2(11)
CHECK (
SUBSTR(time, 1, 2) BETWEEN '01' AND '12'
AND SUBSTR(time, 4, 2) BETWEEN '00' AND '59'
AND SUBSTR(time, 7, 2) BETWEEN '00' AND '59'
AND SUBSTR(time, 10, 2) IN ('AM', 'PM')
AND time LIKE '__:__:__ __'
)
);
or multiple constraints:
CREATE TABLE table_name (
time VARCHAR2(11)
CONSTRAINT invalid_hours CHECK (SUBSTR(time, 1, 2) BETWEEN '01' AND '12')
CONSTRAINT invalid_minutes CHECK (SUBSTR(time, 4, 2) BETWEEN '00' AND '59')
CONSTRAINT invalid_seconds CHECK (SUBSTR(time, 7, 2) BETWEEN '00' AND '59')
CONSTRAINT invalid_meridian CHECK (SUBSTR(time, 10, 2) IN ('AM', 'PM'))
CONSTRAINT invalid_format CHECK (time LIKE '__:__:__ __')
);
But its much easier to not implement all those constraints and just use either an INTERVAL or a DATE when you won't be allowed to enter invalid data.
db<>fiddle here
Prior to posting this question, I have looked at case when, but it doesn't solve my problem.
The scenario is, based on whether it is a Monday in the week, I need to fetch some data from a table. I can write it in PL/SQL block, but I want to achieve this in plain SQL.
Any suggestions please?
Assuming the tables have the same structure this would work:
select * from table1
where to_char(sysdate, 'DY') = 'MON'
union all
select * from table2
where to_char(sysdate, 'DY') != 'MON'
Using day literals like this is dependent on your NLS settings. So adjust the solution if you need to.
to_char(sysdate-3, 'DAY') will return MONDAY you can trim it and using decode return different value than null:
select decode(TRIM(to_char(sysdate-3, 'DAY')),'MONDAY', 1, null) from dual
EDIT:
In where you can
where decode(to_char(sysdate-3, 'fmDAY'), 'MONDAY', 1, null) =1
*sysdate-3 was choose because it's Thursday today
I need to convert a string to a date field. The field stores 30 characters. Dates, when present, are formatted as 'yyyymmdd' (20170202). In all cases, dates have 22 spaces after. I need to format this field as a date field like this: dd-mm-yyyy.
I've tried several formulas:
TO_CHAR(PERSACTION.NEW_VALUE_02, 'dd-mm-yyyy') ,TO_CHAR(PERSACTION.NEW_VALUE_02, 'yyyymmdd'), trim(TO_CHAR(PERSACTION.NEW_VALUE_02, 'yyyymmdd')) with error message: invalid number format model. Your expertise is welcome and appreciated.
to_char(to_date( rtrim(new_value_02), 'yyyymmdd'), 'dd-mm-yyyy')
Should do the trick. rtrim removes spaces on right side of string. Then I convert it to date using the date format specified, and then convert it to a string again in the desired format.
Did tried to convert to date format and then to char again?
TO_CHAR(TO_DATE(PERSACTION.NEW_VALUE_02,'yyyymmdd'),'dd-mm-yyyy')
Please, please, please do not store DATEs and CHARACTER datatypes. This will only lead to issues that can be avoided when using the DATE datatype.
If you want to change the string 20170202 to another string and not actually a date (which would have no intrinsic formatted text representation), you could optionally use a regular expression to transform it, instead of converting to a date and back:
select regexp_replace('20170202 ', '^(\d{4})(\d{2})(\d{2}) +$', '\3-\2-\1')
from dual;
REGEXP_REPLACE(
---------------
02-02-2017
Or you could use substr instead of regexp_substr, which may perform better even if you have to call it three times; using a CTE just to avoid repeating the value:
with t(str) as (
select '20170202 ' from dual
)
select substr(str, 7, 2) ||'-'|| substr(str, 5, 2) ||'-'|| substr(str, 1, 4)
from t;
SUBSTR(STR
----------
02-02-2017
If you do convert to a date and back you would uncover any values which cannot be converted, as they will cause an exception to be thrown. That would imply you have bad data; which would have been avoided by using the right data type in the first place, of course. These will convert any old rubbish, with varying results depending on how far the strings stray from the pattern you expect - but including strings like '20170231' which represent an invalid date. And null value or strings of just spaces will be converted to odd things with the substr version, but you could filter those out.
You can see the kind of variation you would get with some sample data that doesn't match your expectations:
with t(str) as (
select '20170202 ' from dual
union all select '20170231 ' from dual
union all select '2017020c ' from dual
union all select '2017020 ' from dual
union all select '201702021 ' from dual
union all select ' ' from dual
union all select null from dual
)
select str,
regexp_replace(str, '^(\d{4})(\d{2})(\d{2}) +$', '\3-\2-\1') as reg,
substr(str, 7, 2) ||'-'|| substr(str, 5, 2) ||'-'|| substr(str, 1, 4) as sub
from t;
STR REG SUB
------------- ------------- -------------
20170202 02-02-2017 02-02-2017
20170231 31-02-2017 31-02-2017
2017020c 2017020c 0c-02-2017
2017020 2017020 0 -02-2017
201702021 201702021 02-02-2017
- -
--
With the anchors and whitespace expectation, the regular expression doesn't modify anything that doesn't consist entirely of 8 numeric characters. But it can still form invalid 'dates'.
I run two queries and they are returning different results. Please help me find what am I missing.
select UPPER(TO_CHAR(Hire_date,'MONTH'))
from employees
returns a list of months(containing "2" entries for 'MARCH')
SELECT COUNT(EMPLOYEEID) "March Joinees"
FROM Employees
WHERE UPPER(TO_CHAR(Hire_date,'MONTH')) = 'MARCH';
returns 0 as count
to_char(<date>, 'Month') produces a string of length equal to the greatest length of any month name (in the current session's language) - it does so by padding with spaces. This is why the second query produces no results; if you wrap the upper.... within trim(...) it will work, but it would be better not to use to_char(..., 'Month') for this kind of query to begin with. I am sure you came to the same conclusion but you just want to know what is going on...
Here's an illustration. It may seem like the first column in the result is the string 'March'; however, the second column doesn't lie: the result in the first column is actually 'March ' (with four spaces at the end).
select to_char(date '2016-03-01', 'Month') as month_as_string,
length( to_char(date '2016-03-01', 'Month') ) as len
from dual
;
MONTH_AS_STRING LEN
--------------- ---
March 9
Then you may ask why Oracle made such a weird choice. That's a much tougher question. Anyway, this behavior is documented. http://docs.oracle.com/cd/B19306_01/server.102/b14200/sql_elements004.htm#i34924 and scroll down to MONTH in table 2.15.
As stated by mathguy already, 'MONTH' makes padding with spaces. Try
SELECT '"'||TO_CHAR(Hire_date, 'Month')||'"' FROM employees
to see the effect. Use either function TRIM or Format Model Modifiers FM
Then result of TO_CHAR(Hire_date, 'Month') depends your current session NLS_DATE_LANGUAGE value. Other session may get "März" or "Μάρτιος". Specify either date language or use Month numbers.
Actually UPPER(TO_CHAR(Hire_date,'MONTH')) is redundant. Format MONTH returns month name in upper case, you don't have to make UPPER() again.
Taking all this into account you should use one of the expressions below
WHERE TO_CHAR(Hire_date, 'fmMONTH', 'NLS_DATE_LANGUAGE = english') = 'MARCH'
WHERE UPPER(TO_CHAR(Hire_date, 'fmMonth', 'NLS_DATE_LANGUAGE = english')) = 'MARCH'
WHERE TRIM(TO_CHAR(Hire_date, 'MONTH', 'NLS_DATE_LANGUAGE = english')) = 'MARCH'
WHERE TO_CHAR(Hire_date, 'MM') = '03'
WHERE EXTRACT(MONTH FROM Hire_date) = 3
I tried the below query in oracle
select cast(TO_DATE (cal.MONTH,'MM') AS varchar2(30)) as result
FROM JOBCONTROL_USER.ods_calendar_weeks cal
WHERE cal.YEAR NOT IN (0, 9999)
it gives result in dd-mon-yy format. Now I want only mon from the result, how can I achieve this without using to_char()?
If you're avoiding Oracle functions and the month number is stored on its own as a varchar2 field, then you could brute-force it:
select case cast(month as number)
when 1 then 'Jan'
when 2 then 'Feb'
when 3 then 'Mar'
when 4 then 'Apr'
when 5 then 'May'
when 6 then 'Jun'
when 7 then 'Jul'
when 8 then 'Aug'
when 9 then 'Sep'
when 10 then 'Oct'
when 11 then 'Nov'
when 12 then 'Dec'
end as mon
from ods_calendar_weeks cal
where cal.year not in (0, 9999);
But you're having to specify the language; you don't get Oracle's conversion of the month to the NLS date language. Which might be a bonus or a problem depending on your context.
I'd be tempted to put the conversions into a look-up table instead and join to that; or to add the month name as a separate column on the table, as a virtual column in 11g.
You can try somthing like this:-
SELECT EXTRACT(MONTH FROM CAST(TO_DATE (cal.MONTH,'MM') AS varchar2(30))) as RESULT
FROM JOBCONTROL_USER.ods_calendar_weeks cal
WHERE cal.YEAR NOT IN (0, 9999)
Hope this will help you.
select to_char(cal.Month,'month')
) AS result
FROM JOBCONTROL_USER.ods_calendar_weeks cal
WHERE cal.YEAR NOT IN (0, 9999);
This will gives month. the to_char() is a function which has two arguments 1. Column name ans 2. Month. Column name is of date data type so we have to convert the date into character data type and we required only the month so the second column will describes what will be extracted from the date. Finally the result is displayed as a character datatype. This query will returns the months name if year neither 0 nor 9999.