问题
I have two tables
table 1 name : depot_tracking
-------------------------------------------------
depot_tracking_id | cont_details_id | date |
-------------------------------------------------
1 | 4 | 05-02-2015 |
2 | 1 | 25-03-2015 |
3 | 3 | 05-04-2015 |
4 | 2 | 15-02-2015 |
5 | 3 | 05-05-2015 |
6 | 2 | 23-03-2015 |
7 | 1 | 15-05-2015 |
--------------------------------------------------
second table name : cont_details
-----------------------------------
cont_no | cont_details_id |
-----------------------------------
aa | 1 |
bb | 2 |
cc | 3 |
dd | 4 |
ee | 5 |
ff | 6 |
gg | 7 |
--------------------------------------
now i want to merge this two tables and filter repeated rows and one more most important thing is i have to filter the rows which has OLDER dates. This is what exactly i am expecting output
---------------------------------------------------------
depot_tracking_id | cont_details_id | date | cont_no
----------------------------------------------------------
1 | 4 | 05-02-2015 | dd
5 | 3 | 05-05-2015 | cc
6 | 2 | 23-03-2015 | bb
7 | 1 | 15-05-2015 | aa
---------------------------------------------------------
please some one help me to filter repeated rows ( filter the rows which has old date rows) thanks
ORDER BY (str_to_date(date,'%d-%m-%Y')) DESC LIMIT 1
by using this code i can change date column (DD-MM-YYYY) to mysql format date
but dont dont know how to filter repeated rows based on dates
回答1:
Because all of your field values can be reliably determined by using aggregate functions, you only need a simple aggregation query.
select max(d.depot_tracking_id), c.cont_details_id, max(d.date), c.cont_no
from depot_tracking d
inner join cont_details c
on c.cont_details_id = d.cont_details_id
group by c.cont_details_id
In order for this to work, your date fields have to be an actual date type, not a string type. or if they ARE a string type, they need to be 'yyyy-mm-dd' format, not 'dd-mm-yyyy'.
demo here
update
Since it seems you are stuck using the broken date format - this query will work:
select max(d.depot_tracking_id), c.cont_details_id, max(str_to_date(d.date,'%d-%m-%Y')), c.cont_no
from depot_tracking d
inner join cont_details c
on c.cont_details_id = d.cont_details_id
group by c.cont_details_id
demo
回答2:
http://sqlfiddle.com/#!9/113ed/1
SELECT
d.depot_tracking_id,
c.cont_details_id,
d.`date`,
c.cont_no
FROM depot_tracking d
LEFT JOIN depot_tracking d1
ON d.cont_details_id = d1.cont_details_id
AND d.`date` < d1.`date`
LEFT JOIN cont_details c
ON c.cont_details_id = d.cont_details_id
WHERE d1.`date` IS NULL
回答3:
First and foremost, MySQL does not store dates in the 'dd-mm-yyyy' format but in 'yyyy-mm-dd'. If you are using a date type, the following solution will work. If it's not, you'll have to do some string to date formatting. I highly recommend you change to store dates the proper way though.
Start by selecting the newest date for each cont_details_id in the depot_tracking table like this:
SELECT cont_details_id, MAX(date) AS latest
FROM depot_tracking
GROUP BY cont_details_id;
Once you have that, you can join it back to your original tables using date/cont_details_id to get the rows you want:
SELECT d.depot_tracking_id, d.cont_details_id, d.date, c.cont_no
FROM depot_tracking d
JOIN(
SELECT cont_details_id, MAX(date) AS latestDate
FROM depot_tracking
GROUP BY cont_details_id) tmp ON tmp.cont_details_id = d.cont_details_id AND tmp.latestDate = d.date
JOIN cont_details c ON c.cont_details_id = tmp.cont_details_id;
Here is an SQL Fiddle example that shows both queries separately so you can see how they come together.
来源:https://stackoverflow.com/questions/30105515/how-to-filter-repeated-rows-based-on-date-in-mysql