Detect overlapping date ranges from the same table

I have a table with the following data

PKey  Start       End         Type
====  =====       ===         ====
01    01/01/2010  14/01/2010  S
02    15/01/2010  31/01/2010  S
03    05/01/2010  06/01/2010  A

And want to get the following results

PKey  Start       End         Type
====  =====       ===         ====
01    01/01/2010  14/01/2010  S
03    05/01/2010  06/01/2010  A

A开发者_运维技巧ny ideas on where to start? A lot of the reading I've done suggests I need to create entries and for each day and join on matching days, is this the only way?

---

If you already have entries for each day that should work, but if you don't the overhead is significant, and if that query is used often, if will affect performance.

If the data is in this format, you can detect overlaps using simple date arithmetic, because an overlap is simply one interval starting after a given interval, but before the given is finished, something like

select dr1.* from date_ranges dr1
inner join date_ranges dr2
on dr2.start > dr1.start -- start after dr1 is started
  and dr2.start < dr1.end -- start before dr1 is finished

If you need special handling for interval that are wholly within another interval, or you need to merge intervals, i.e.

PKey  Start       End         Type
====  =====       ===         ====
01    01/01/2010  20/01/2010  S
02    15/01/2010  31/01/2010  S

yielding

Start       End         Type
=====       ===         ====
01/01/2010  31/01/2010  S

you will need more complex calculation.

In my experience with this kind of problems, once you get how to do the calculation by hand, it's easy to transfer it into SQL :)

---

When I needed to compare two time spans in SQL for overlap, here are the four scenarios I could think of:

1. Span1 start is between Span2 start and Span2 end
2. Span1 end is between Span2 start and Span2 end
3. Span1 start and end are both between Span2 start and Span2 end
4. Span2 start and end are both between Span1 start and Span1 end

Here is the OR statement I created to capture these scenarios (in my case Oracle SQL):

and (
    s1.start between s2.start and s2.end
    OR
    s1.end between s2.start and s2.end
    OR
    s2.start between s1.start and s1.end
)

---

Perhaps:

SELECT A.PKey, A.Start, A.End, A.Type
FROM calendar AS A, calendar AS B
WHERE (p.pkey<>a.pkey
AND b.start>=a.start
AND b.end<=a.end)
OR (b.pkey<>a.pkey
AND b.start<=a.start
AND b.end>=a.end)

---

If you are using PostgreSQL, simply use the built-in overlap operator

SELECT (DATE '2021-01-01', DATE '2021-04-09') 
OVERLAPS (DATE '2021-01-20', DATE '2021-02-10');

---

select A.*
from MyTable A
inner join MyTable B
on (B.start <= A.end)
and (B.end >= A.start)

or something like that (assuming dates are not nullable and equal dates count as an overlap).

---

In MySQL you basically need:

SELECT COUNT(*) FROM date_ranges AS A, date_ranges AS B WHERE A.id <> B.id AND A.id > B.id AND A.end_at > B.start_at AND B.end_at > A.start_at

> in the second and the third statement can be replaced with >= to follow includes matching.

This topic is related to the "Allen's Interval Algebra" and there are some more reading on this can be found by those links:

• http://www.ics.uci.edu/~alspaugh/cls/shr/allen.html
• http://salman-w.blogspot.com.es/2012/06/sql-query-overlapping-date-ranges.html

---

We've all needed this kind of overlapping predicate in our queries for quite some time and I think I've found a really simple solution here.

In my application, as an example, I have policies that have the same Policy Number but maybe the Policy Description changes from one fiscal year to the next. When a user is entering a new record (same Policy Number, different Policy Description), I needed a way to tell if that policy already exists for the specified time range. If the new Policy Effective/Expiration dates overlap with whatever is already in the database, I needed to error out and tell the user why their input was not correct.

To do this, I went with the following predicate statement:

AND @_expiration >= EffectiveDate AND ExpirationDate >= @_effective

Hopefully someone else finds this as useful as I have.

---

I had to do a very similar thing for to stop duplicate holiday being entered into a table. it was in access and written to a temptable on input so had to query it in VBA SQL:

 stCommandText = "SELECT " _
                    & "* " _
                    & "FROM " _
                    & "TableName a, " _
                    & "TableName b " _
                    & "WHERE " _
                    & "a.ID = b.ID " _
                    & "AND a.Startdate >= b.Startdate AND a.StartDate <= b.EndDate " _
                    & "AND a.AutoNo <> b.AutoNo "

---

BTW - If you don't have a unique id , against your dates you can do this is oracle..FYI

with date_ranges
as
(
SELECT 
     rownum as pkey,
    date_ranges.*
FROM  date_ranges
) 
select 
dr1.* 
from 
date_ranges dr1 , date_ranges dr2
where  dr1.pkey > dr2.pkey
AND dr1.end_dt >= dr2.start_dt 
AND dr2.end_dt >= dr1.start_dt

---

Sql='SELECT task_id, task_start_date, task_due_date FROM (wba_task) WHERE (task_start_date <="2016-07-13" AND task_due_date >="2016-07-25") OR (task_due_date BETWEEN "2016-07-13" and "2016-07-25")';

Codeigniter Query is below.

$fromdaysDate="2016-07-13";//changed date
$todaysDate="2016-07-25";//changed date
$this->db->select('task_id,task_start_date, task_due_date'); 
$this->db->where('task_start_date <="'.date('Y-m-d', strtotime($fromdaysDate)).'"');
$this->db->where('task_due_date >="'.date('Y-m-d', strtotime($todaysDate)).'"');    
$this->db->or_where('task_due_date BETWEEN "'. date('Y-m-d', strtotime($fromdaysDate)). '" and "'. date('Y-m-d', strtotime($todaysDate)).'"');   
$alltask=$this->db->get('wba_task')->result_array();
echo $this->db->last_query();

get all overlap data form database....

---

To solve if the date is overlapping or not:

Table Creation:

create  table testing (
   id int,
    s_date date,
    e_date date
    
);

Loading Data:

INSERT INTO testing ( id,s_date,e_date)
VALUES ('1','1/1/2020','1/31/2020'),
('2',   '1/16/2020',    '1/26/2020'),
('3',   '1/28/2020',    '2/6/2020'),
('4',   '2/16/2020',    '2/26/2020');

Query:

select id, case when sum(Overlap) > 0 then 'True' else 'False' end as overlap  

from (

select a.*, b.id as ids,  b.e_date, b.s_date
,case when a.s_date < b.e_date then 1 else 0 end as Overlap

from testing a cross join testing b  where a.id <> b.id and a.e_date > b.s_date
) group by 1