How to Compare the Current Row with Next and Previous Row in Postgresql

How to compare the current row with next and previous row in PostgreSQL?

This is my solution using WINDOW functions. I used the lag and lead functions. Both returns a value from a column from a row in offset from the current row. lag goes back and lead goes next in the offset.

SELECT tokcat.text
FROM (
    SELECT text, category, chartype, lag(category,1) OVER w as previousCategory, lead(category,1) OVER w as nextCategory
    FROM token t, textBlockHasToken tb
    WHERE tb.tokenId = t.id
    WINDOW w AS (
        PARTITION BY textBlockId, sentence
        ORDER BY textBlockId, sentence, position
    )
) tokcat
WHERE 'NAME' = ANY(previousCategory)
AND 'NAME' = ANY(nextCategory)
AND 'NAME' <> ANY(category)

---

Simplified version:

SELECT text
FROM (
    SELECT text
          ,category 
          ,lag(category) OVER w as previous_cat
          ,lead(category) OVER w as next_cat
    FROM   token t
    JOIN   textblockhastoken tb ON tb.tokenid = t.id
    WINDOW w AS (PARTITION BY textblockid, sentence ORDER BY position)
    ) tokcat
WHERE  category <> 'NAME'
AND    previous_cat = 'NAME'
AND    next_cat = 'NAME';

Major points

• = ANY() is not needed, the window function returns a single value
• some redundant fields in the subquery
• no need to order by columns, that you PARTITION BY - the ORDER BY applies within partitions
• Don't use mixed case identifiers without quoting, it only leads to confusion. (Better yet: don't use mixed case identifiers in PostgreSQL ever)

Postgres Query Based on Previous and Next Rows

I am not sure if I understood your problem correctly. But getting values from other rows this can be done by window functions (https://www.postgresql.org/docs/current/static/tutorial-window.html):

demo: db<>fiddle

SELECT
    id,
    lag("to") OVER (ORDER BY id) as prev_to,
    "from",
    "to",
    lead("from") OVER (ORDER BY id) as next_from
FROM bustimes;

The lag function moves the value of the previous row into the current one. The lead function does the same with the next row. So you are able to calculate a difference between last arrival and current departure or something like that.

Result:

id   prev_to   from    to      next_from
18             33000   33300   33300
19   33300     33300   33600   33900
20   33600     33900   34200   34200
21   34200     34200   34800   36000
22   34800     36000   36300        

---

Please notice that "from" and "to" are reserved words in PostgreSQL. It would be better to chose other names.

Comparing record to previous record in postgresql

SELECT client, 
       rate,
       startdate, 
       enddate, 
       lag(rate) over client_window as pre_rate,
       lag(startdate) over client_window as pre_startdate,
       lag(enddate) over client_window as pre_enddate
FROM the_table
WINDOW client_window as (partition by client order by startdate)
ORDER BY client, stardate;

This assumes that enddate is always greater than startdate from the same row and that no enddate is greater than the following startdate

Compare Current Row with Previous row in SQL Server

You can use the following query:

SELECT RaidNo, OutComeID, RN,
       CASE 
          WHEN OutComeID <> 16 THEN 0
          ELSE ROW_NUMBER() OVER (PARTITION BY OutComeID, grp ORDER BY RN) 
       END AS Result
FROM (
   SELECT RaidNo, OutComeID, RN,
          RN - ROW_NUMBER() OVER (PARTITION BY OutComeID ORDER BY RN) AS grp
   FROM mytable) AS t
ORDER BY RN

Field grp identifies slices (also called islands) of consecutive records having the same OutComeID value. The outer query uses grp in order to enumerate each record that belongs to a '16' slice. The records that belong to the other slices are assigned value 0.

Demo here

In PostgreSQL, how to select the previous row value to compute the current row value?

This will do it. It's not especially efficient though, but for a one off it should do.

select x.id, sum(coalesce(y.value1,y.id,0))
from sample x
left outer join sample y on x.id >= y.id
group by x.id
order by x.id

Basically, it goes through every record in the table, and then sums up the IDs of all the records up to and including that record. There is a special logic to handle the initial value you've got on id=1.

SQL Fiddle at http://sqlfiddle.com/#!9/b8f25/1

Another option is to use a window function.

select x.id, sum(coalesce(x.value1,x.id,0)) over (order by x.id)
from sample x

SQL Fiddle at http://sqlfiddle.com/#!17/b8f25/2

Compare a row with ALL the previous rows in SQL

There may be a different solution using analytic functions instead of a scalar sub query, however, I haven't quite figured it out yet. Meanwhile here's this solution:

with t1 as (
  select yd.*
       , ROW_NUMBER() OVER (PARTITION BY trip ORDER BY GamingDate, DateRedeemed) rn
    from YourData yd
)
select t1.*
     , (select sum(amount) from t1 t2
         where t2.trip = t1.trip
           and t2.rn <= t1.rn
           and t2.gamingdate <= t1.gamingdate
           and t1.gamingdate < t2.dateredeemed) Running
  from t1
 order by trip, rn;

DateRedeemed        | GamingDate          | Trip | Amount | rn | Running
:------------------ | :------------------ | ---: | -----: | :- | ------:
13/07/2017 00:00:00 | 03/07/2017 00:00:00 | 8001 |    100 | 1  |     100
17/07/2017 00:00:00 | 03/07/2017 00:00:00 | 8001 |    150 | 2  |     250
18/07/2017 00:00:00 | 04/07/2017 00:00:00 | 8001 |    125 | 3  |     375
27/07/2017 00:00:00 | 16/07/2017 00:00:00 | 8001 |    250 | 4  |     525
28/07/2017 00:00:00 | 16/07/2017 00:00:00 | 8001 |    310 | 5  |     835
27/07/2017 00:00:00 | 17/07/2017 00:00:00 | 8001 |    125 | 6  |     810
31/07/2017 00:00:00 | 18/07/2017 00:00:00 | 8001 |     75 | 7  |     760
28/07/2017 00:00:00 | 27/07/2017 00:00:00 | 8001 |     80 | 8  |     465
31/07/2017 00:00:00 | 28/07/2017 00:00:00 | 8001 |    100 | 9  |     175

I do realize that the last two rows do not match the sample result, however, I think that's due to an error in the sample result rather than an error in my code. If you can explain why the sample result is correct for those two records I can try to rework my code.

[EDIT]

Here's an alternate version that avoids the inefficient scalar subquery in favor of using an analytic function:

With Actions as (
  select *, GamingDate ActionDate, 1 DBCR from Table1
  union all
  select *, DateRedeemed ActionDate, -1 DBCR from Table1
), Analytics as (
  select *
       , sum(Amount*DBCR) over (partition by trip
                                    order by ActionDate, dbcr
                                           , GamingDate, DateRedeemed) Running
    from Actions 
)
select DateRedeemed, GamingDate, Trip, Amount, Running
  from Analytics
 where dbcr = 1

---

Related Topics