I have a simple problem which gave em a headache
I need to sort integers in a Database table TDBGrid ( its ABS database from component ace ) with the following order
0
1
11
111
121
2
21
211
22
221
and so on
which means every number starting with 1 should be under 1
1
11
111
5
55
can anyone help me?
thanks
This should work to get stuff in the right order:
Convert the original number to a string;
Right-pad with zeroes until you have a string of 3 characters wide;
(optional) Convert back to integer.
Then sorting should always work the way you want. Probably it's best to let the database do that for you. In MySQL you'd do something like this:
select RPAD(orderid,3,'0') as TheOrder
from MyTable
order by 1
I just ran this in SQL Server Management Studio - note I mixed up the rows in the input so they were not in sorted order:
create table #temp( ID Char(3));
insert into #temp (ID)
select '111' union
select '221';
select '0' union
select '21' union
select '1' union
select '11' union
select '211' union
select '121' union
select '2' union
select '22' union
select * from #temp order by ID;
I got the following output:
ID
----
0
1
11
111
121
2
21
211
22
221
(10 row(s) affected)
If you're getting different results, you're doing something wrong. However, it's hard to say what because you didn't post anything about how you're retrieving the data from the database.
Edit: Some clarification by the poster indicates that the display is in a TDBGrid attached to a table using Component Ace ABS Database. If that indeed is the case, then the answer is to create an index on the indicated column, and then set the table's IndexName property to use that index.
select cast(INT_FIELD as varchar(9)) as I
from TABxxx
order by 1
Related
I really been working to solve this problem for quite a while, I have 2 tables with the same structure as follows:
registerationNumber int, CompanyName nvarchar, areaName nvarchar,phoneNumber int , email nvarchar, projectStatus nvarchar
All columns with data type nvarchar contains Arabic text except the email column,
tableA contains 675 rows and tableB contains 397 rows all exists in tableA
What I am trying to do is select the non matching rows from tableA,
they should be 675 - 398 = 277 rows
everytime I run the where clause I get all tables returned
The join clause I am writing is like this:
select a.registerationNumber
from tableA a left outer join tableB b
on a.registerationNumber = b.registerationNumber
but I am not getting any results, I tried all types of joins but I am getting the same results.
I created a sample database and inserted English data in the tables and it worked fine with the following clause:
select * from tblAllProjects a right join tblhalfProjects h
on a.registerationNumber = h.registerationNumber
which means that I am writing the correct the correct syntax,
I know that I should use the following syntax on selecting Arabic text:
Select * from tableA where comanyName like N'arabic_text'
Anyone knows what seems to be the problem ?
I meant to say that you should do:
select a.registerationNumber
from tableA a left outer join tableB b
on a.registerationNumber = b.registerationNumber
where b.registrationNumber is NULL
This should select all registrationNumbers from tableA, with no match in tableB.
another way to write this (but probably slower) is:
select a.registerationNumber
from tableA a
where a.registerationNumber not in (
select b.registerationNumber
from tableB )
You should/can use this second approach if there are only "a few" records returned from the sub-query.
Background
I have 2 data tables.
For each row in tableA, I want to find the rows in tableB with the closest dates and join those values onto the row from tableA.
Example tables:
tableA:
p_id
category
l_date
1
catA
2005-01-05
1
catB
2005-06-10
2
catC
2000-01-10
tableB:
p_id
e_id
e_date
1
22
2005-01-01
1
23
2005-01-06
1
24
2005-01-06
1
28
2005-01-10
2
29
2010-08-10
desired result:
p_id
category
l_date
e_id
e_date
1
catA
2005-01-05
23
2005-01-06
1
catA
2005-01-05
24
2005-01-06
1
catB
2005-06-10
28
2005-01-10
2
catC
2000-01-10
29
2010-08-10
Tried
This query does not work, but I think this is the direction I should be going.
select a.p_id, a.category, a.l_date, c.e_id, c.e_date from tableA a
left join lateral
(
select top 1 p_id, e_id, e_date from tableB b
where a.pid = b.pid
order by abs(datediff(days, a.l_date, b.e_date))
) c on True;
TableA and tableB are massive, 17m and 150m respective rows.
Does this sound like the correct approach?
Using redshift cluster, running postgres 8.x
Correlated subquery approaches or a full cross join approach will all perform the task of comparing every row in one table with every row in the other (in one manner or another). Comparing (joining) all these rows when the tables get large get prohibitive. In these cases different approaches are needed.
Brute forcing won't be fast (if it even completes) so we need to be a bit more efficient in going about this. I tell clients to think about how they would do this query (by hand) if I gave them stacks of index cards. A person values their time so they don't go about this by making all possible combinations, they would come up with a more efficient way that they can complete quickly and get back to their lives. In cases like the one you are describing you need to find the more efficient approach. I'd be happy to talk to you more about building these types of queries.
Taking your data (and sprucing it up a bit for some more interesting cases) I created an example of how you can do this. (Yes, you could cross join the small tables and do this with simpler SQL but that won't scale.)
Data setup:
create table tableA (p_id int, category varchar(64), l_date date);
insert into tableA values
(1,'catA','2005-01-05'),
(1,'catB','2005-06-10'),
(2,'catC','2000-01-10');
create table tableB (p_id int, e_id int, e_date date);
insert into tableB values
(1,22,'2005-01-01'),
(1,23,'2005-01-06'),
(1,24,'2005-06-01'),
(1,28,'2005-06-15'),
(2,29,'2010-08-10');
The query looks like:
with combined as
(
select
*,
coalesce(max(l_date) OVER (partition by p_id order by
dt rows between unbounded preceding and 1 preceding), '1970-01-01'::date) cb,
coalesce(min(l_date) OVER (partition by p_id order by
dt desc rows between unbounded preceding and 1 preceding), '2100-01-01'::date) ca
from
(
select
p_id,
category,
l_date,
NULL as e_id,
NULL as e_date,
l_date dt
from
tableA
union all
select
p_id,
NULL as category,
NULL as l_date,
e_id,
e_date,
e_date dt
from
tableB
) c
)
,
closest as
(
select
p_id,
e_id,
e_date,
cb,
ca,
case
when
coalesce(e_date - cb, 0) > (ca - e_date)
then ca
else cb
end closest
from
combined
where
e_date is not NULL
)
select
c.p_id,
a.category,
a.l_date,
c.e_id,
c.e_date
from
closest c
left join tableA a
on c.closest = a.l_date and c.p_id = a.p_id
order by
c.p_id,
c.e_id ;
While this can look like a lot it isn't that complex. First CTE finds the closest l_date earlier than e_date (cb) and the closest l_date later than e_date (ca). It does this on on UNIONed set of data to allow for windowing. The second CTE just determines which is closer, ca or cb, and produces this as "closest". It also strips out all the tableB information that was added by the UNION (no longer needed). Lastly this "closest" date provides the join on information needed to build the final result.
Now this query doesn't account of many possible real-world data issues that can happen so take this as a starting point. I'm also making some assumptions about your data based on the test data (like no 2 rows in tableA will have the same l_date and P_id). So use this as a starting point.
And a last word on performance - while window functions are not cheap and will do more work as your data tables increase in size, they are orders of magnitude more performant than cross-joining massive tables. What you are looking to do is complex so will take some time but this is the fastest way I have found perform these complex operations that would normally be a massive loop problem.
I am using rails with postgresql to populate a dataTable and was wondering if there is a way to change the default behaviour of NULLS being the higher value (coming after large numbers when sorted) to become equivalent to lower than 0 in a sort. From what I understand this is a built in postgresql behaviour, so I think I will have to use the sql call to achieve this. And I need to apply this to all columns so it works with the DataTables sort ASC/DESC functionality.
Example some functionality similar to:
def get_raw_records
Analytics::Database.where(id: 7).order('give nulls < 0 value here for all columns?')
end
NULLS FIRST/ LAST does not give this functionality I need something like coalesce that ideally does not return a sorted instance but changes the default behaviour of nulls placed after large values when for it is sorted client side
you can use
order by coalesce(col_name,-1)
close to Sabari suggested:
db=# with c(v) as (values('1'),('a'),(null),(null),('b'))
select * from c
order by coalesce(v,'-infinity') asc;
v
---
1
a
b
(5 rows)
db=# with c(v) as (values('1'),('a'),(null),(null),('b'))
select * from c
order by coalesce(v,'-infinity') desc;
v
---
b
a
1
(5 rows)
instead of fixed integer below zero I use -inf here. which works fine with text (not because text understands infinity, but rather because - goes before [a-z] or [0-9]), but would not work with normal integer. Of course you can cast it to float:
db=# with c(v) as (values(1::float),(3),(null),(null),(-9))
select * from c
order by coalesce(v,'-infinity') desc;
v
----
3
1
-9
(5 rows)
db=# with c(v) as (values(1::float),(3),(null),(null),(-9))
select * from c
order by coalesce(v,'-infinity') asc;
v
----
-9
1
3
(5 rows)
with is dangerous (well, not in sorting though) itself (and ugly). which leads to the answer - I don't see good solution here from the top of my head... You should better make separate boundaries for different data types.
In Postgres you can specify in the order by clause how you want NULL sorted:
select 1 as col UNION select 2 UNION SELECT NULL ORDER BY col ASC NULLS FIRST;
You can either specify NULLS FIRST or NULLS LAST
See https://www.postgresql.org/docs/current/static/queries-order.html
For hive outer join, if a joining key does not exist in one table,hive will put NULL. Is that possible to use another value for this? For example:
Table1:
user_id, name, age
1 Bob 23
2 Jim 43
Table2:
user_id, txn_amt, date
1 20.00 2013-12-10
1 10.00 2014-07-01
If I do a LEFT OUTER JOIN on user_id:
INSERT INTO TABLE user_txn
SELECT
Table1.user_id,
Table1.name,
Table2.txn_amt,
Table2.date
FROM
Table2
LEFT OUTER JOIN
Table1
ON
Table1.user_id = Table2.user_id;
I want the output be like this:
user_id, name, tnx_amt, date
1 Bob 20.00 2013-12-10
1 Bob 10.00 2014-07-01
2 Jim 0.00 2099-12-31
Note the txn_amt and date columns for Jim. Is there any way in hive to define default values like this?
You can use COALESCE for this, instead of solely Table2.txn_amt
COALESCE(Table2.txn_amt, 0.0)
What this does is returns the first value that is not null. So, if txn_amt is null, it'll go to the second value in the list. 0.0 is never null, so it'll pick that. If txn_amt has a value in it, it'll return that value.
I'm using the following database:
CREATE TABLE datas (d_id INTEGER PRIMARY KEY, name_id numeric, countdata numeric);
INSERT INTO datas VALUES(1,1,20); //(NULL,1,20);
INSERT INTO datas VALUES(2,1,47); //(NULL,1,47);
INSERT INTO datas VALUES(3,2,36); //(NULL,2,36);
INSERT INTO datas VALUES(4,2,58); //(NULL,2,58);
INSERT INTO datas VALUES(5,2,87); //(NULL,2,87);
CREATE TABLE names (n_id INTEGER PRIMARY KEY, name text);
INSERT INTO names VALUES(1,'nameA'); //(NULL,'nameA');
INSERT INTO names VALUES(2,'nameB'); //(NULL,'nameB');
What I would like to do, is to select all values (rows) of names - to which all columns of datas will be appended, for the row where datas.countdata is maximum for n_id (and of course, where name_id = n_id).
I can somewhat get there with the following query:
sqlite> .header ON
sqlite> SELECT * FROM names AS n1
LEFT OUTER JOIN (
SELECT d_id, name_id, countdata FROM datas AS d1
WHERE d1.countdata IN (
SELECT MAX(countdata) FROM datas
WHERE name_id=1
)
) AS p1 ON n_id=name_id;
n1.n_id|n1.name|p1.d_id|p1.name_id|p1.countdata
1|nameA|2|1|47
2|nameB|||
... however - obviously - it only works for a single row (the one explicitly set by name_id=1).
The problem is, the SQL query fails whenever I try to somehow reference the "current" n_id:
sqlite> SELECT * FROM names AS n1
LEFT OUTER JOIN (
SELECT d_id, name_id, countdata FROM datas AS d1
WHERE d1.countdata IN (
SELECT MAX(countdata) FROM datas
WHERE name_id=n1.n_id
)
) AS p1 ON n_id=name_id;
SQL error: no such column: n1.n_id
Is there any way of achieving what I want in Sqlite2??
Thanks in advance,
Cheers!
Oh, well - that wasn't trivial at all, but here is a solution:
sqlite> SELECT * FROM names AS n1
LEFT OUTER JOIN (
SELECT d1.*
FROM datas AS d1, (
SELECT max(countdata) as countdata,name_id
FROM datas
GROUP BY name_id
) AS ttemp
WHERE d1.name_id = ttemp.name_id AND d1.countdata = ttemp.countdata
) AS p1 ON n1.n_id=p1.name_id;
n1.n n1.name p1.d_id p1.name_id p1.countdata
---- ------------ ---------- ---------- -----------------------------------
1 nameA 2 1 47
2 nameB 5 2 87
Well, hope this ends up helping someone, :)
Cheers!
Notes: note that just calling max(countdata) screws up competely d_id:
sqlite> select d_id,name_id,max(countdata) as countdata from datas group by name_id;
d_id name_id countdata
---- ------------ ----------
3 2 87
1 1 47
so to get correct corresponding d_id, we must do max() on datas separately - and then perform sort of an intersect with the full datas (except that intersect in sqlite requires that there are equal number of columns in both datasets, which is not the case here - and even if we made it that way, as seen above d_id will be wrong, so intersect will not work).
One way to do that is in using a sort of a temporary table, and then utilize a multiple table SELECT query so as to set conditions between full datas and the subset returned via max(countdata), as shown below:
sqlite> CREATE TABLE ttemp AS SELECT max(countdata) as countdata,name_id FROM datas GROUP BY name_id;
sqlite> SELECT d1.*, ttemp.* FROM datas AS d1, ttemp WHERE d1.name_id = ttemp.name_id AND d1.countdata = ttemp.countdata;
d1.d d1.name_id d1.countda ttemp.coun ttemp.name_id
---- ------------ ---------- ---------- -----------------------------------
2 1 47 47 1
5 2 87 87 2
sqlite> DROP TABLE ttemp;
or, we can rewrite the above so a SELECT subquery (sub-select?) is used, like this:
sqlite> SELECT d1.* FROM datas AS d1, (SELECT max(countdata) as countdata,name_id FROM datas GROUP BY name_id) AS ttemp WHERE d1.name_id = ttemp.name_id AND d1.countdata = ttemp.countdata;
d1.d d1.name_id d1.countda
---- ------------ ----------
2 1 47
5 2 87