I have a postgres column double[]: {100, 101, 102}.
I want to divide each element by 10, so the result should be {10.0, 10.1, 10.2}.
I've only found solutions with for-statements, but how to realize this with a simple query? (I need to update through liquibase)
The alternative is to write a Java-migration, but I would prefer a simple query...
Thanks in advance!
UPDATE:
A second problem that arose is:
When doing this through a Liquibase java-migration script, you get a liquibase.database.jvm.JdbcConnection (through liquibase.change.custom.CustomTaskChange), which of course does not support postgres-arrays =/.
How to handle arrays this way? (I use liquibase-core 3.5.5)
You need to unnest, divide, then aggregate back.
update the_table
set the_array = array(select t.val / 10
from unnest(the_table.the_array) as t(val));
If you need to preserve the original order in the array use with ordinality
update the_table
set the_array = array(select t.val / 10
from unnest(the_table.the_array) with ordinality as t(val,idx)
order by t.idx);
To run this in Liquibase you need to use a <sql> change
Online example: https://rextester.com/IJGA96691
Related
So, I have Hibernate 5.3.1 in a project which connects to different enginees (MySql, Oracle, PostgreSQL and MS SQL), so I can't use native queries.
Let's say I have 3 records in a table, which all of them have the same datetime, but I need to group them only by date (not time). For example, 2019-12-04;
I execute this query:
SELECT
CONCAT(year(tx.date_), month(tx.date_), day(tx.date_)),
iss.code,
COUNT(tx.id)
FROM
tx_ tx
JOIN
issuer_ iss
ON
tx.id_issuer = iss.id
GROUP BY
CONCAT(year(tx.date_), month(tx.date_), day(tx.date_)), iss.code
But, when I test it connected to SQL SERVER 2017, instead of return 20191204, it's returning 2035. In Oracle and MySQL is working fine.
Anyone has any idea why is this happen? I've tried different ways, like use + instead of CONCAT but the result is the same.
I've also tried to extract them for separate (without concat), and they have been returning correct. The problem is, I need to group them by the complete date.
And just for the record, the field is declared as datetime2 in DDBB
How about simply adding them, instead of using CONCAT.
(year(tx.date_)*10000 + month(tx.date_)*100 + day(tx.date_)*1) AS datenum
Thus, try this:
SELECT
CAST((year(tx.date_)*10000 + month(tx.date_)*100 + day(tx.date_)*1) AS string) AS datenum,
iss.code
FROM tx_ tx
JOIN issuer_ iss
ON tx.id_issuer = iss.id
GROUP BY year(tx.date_), month(tx.date_), day(tx.date_), iss.code
Thanks for the hint Gert Arnold gave me. I just didn't realize that the query was adding like if they were numbers in MSSQL.
Finally, I manage to make it work in the 4 RDBMS casting to string first
SELECT
CONCAT(CAST(year(tx.date_) AS string), CAST(month(tx.date_) AS string), CAST(day(tx.date_) AS string)),
iss.code
FROM
tx_ tx
JOIN
issuer_ iss
ON
tx.id_issuer = iss.id
GROUP BY
CONCAT(year(tx.date_), month(tx.date_), day(tx.date_)), iss.code
I tried also casting to TEXT, but it throws exception in MySQL
Why use concat() to begin with?
Assuming Hibernate takes care of converting the non-standard year(), month() and day() functions, then the following should work on any DBMS
SELECT year(tx.date_), month(tx.date_), day(tx.date_), iss.code
FROM tx_ tx
JOIN issuer_ iss ON tx.id_issuer = iss.id
GROUP BY year(tx.date_), month(tx.date_), day(tx.date_), iss.code
I'm trying to build a query in MS Access and having an issue figuring out the best way to build it. What I'm trying to do is is make sure all the Higher retails match within a set of matching pack numbers. For example:
PackNum Prefix Retail
6451618 DF 37.99
6451618 SK 37.99
6451618 VJ 34.99
6451618 SG 37.99
One of the group is off and I want the query to show it.
I was attempting to use something like this to have check but I'm not getting the results I'm looking for
IIf([dbo_PIC704Current]![PackNum]=[dbo_PIC704Current]![PackNum]
And [dbo_PIC704Current]![Ret2]<>[dbo_PIC704Current]![Ret2],True,False)
Any help or push in the right direction would be greatly appreciated!
-Deke
SELECT dbo_pic704current.packnum
, Min(dbo_pic704current.Ret2) AS LowRet2
, Max(dbo_pic704current.Ret2) AS HighRet2
, dbo_CatalogInfo.MediaId
, dbo_pic704current.DiscountReasonCode
, dbo_CatalogInfo.Brand
FROM dbo_pic704current INNER JOIN dbo_CatalogInfo ON (dbo_pic704current.year =
dbo_CatalogInfo.mailyear) AND (dbo_pic704current.catid = dbo_CatalogInfo.catalog)
WHERE (((dbo_CatalogInfo.MediaId)='CAT Catalog'))
GROUP BY dbo_pic704current.packnum, dbo_CatalogInfo.MediaId,
dbo_pic704current.DiscountReasonCode, dbo_CatalogInfo.Brand, dbo_pic704current.Year
HAVING (((Min(dbo_pic704current.Ret2))<>Max([Ret2])))
ORDER BY dbo_pic704current.packnum;
Your existing solution using aggregation is likely to yield better performance, but to offer an alternative, here is an example using a correlated subquery:
select
pc.packnum, pc.ret2, ci.mediaid, pc.discountreasoncode, ci.brand
from
dbo_pic704current pc inner join dbo_cataloginfo ci on
pc.year = ci.mailyear and pc.catid = ci.catalog
where
ci.mediaid = 'cat catalog' and exists
(select 1 from dbo_pic704current t where t.packnum = pc.packnum and t.ret2 > pc.ret2)
order by
pc.packnum
I have a pair of databases (one mssql and one oracle), ran by different teams. Some data are now being synchronized regularily by a stored procedure in the mssql table. This stored procedure is calling a very large
MERGE [mssqltable].[Mytable] as s
USING THEORACLETABLE.BLA as t
ON t.[R_ID] = s.[R_ID]
WHEN MATCHED THEN UPDATE SET [Field1] = s.[Field1], ..., [Brokenfield] = s.[BrokenField]
WHEN NOT MATCHED BY TARGET THEN
... another big statement
Field Brokenfield was a numeric one until today, and could take value NULL, 0, 1, .., 24
Now, the oracle team introduced a breaking change today for some reason, changed the type of the column to string and now has values NULL, "", "ALFA", "BRAVO"... in the column. Of course, the sync got broken.
What is the easiest way to fix the sync here? I (Mysql team lead, frontend expert but not so in databases) would usually apply one of our database expert guys here, but all of them are now ill, and the fix must go online today....
I thought of a stored procedure like CONVERT_BROKENFIELD_INT_TO_STRING or so, based on some switch-case, which could be called in that merge statement, but not sure how to do that.
Edit/Clarification:
What I need is a way to make a chunk of SQL code (stored procedure), taking an input of "ALFA" and returning 1, "BRAVO" -> 2, etc. and which can be reused, to avoid writing huge ifs in more then one place.
If you can not simplify the logic for correct values the way #RichardHansell desribed, you can create a crosswalk table for BrokenField to the correct values. Then you can use a common table expression or subquery with a left join to that crosswalk to use in the merge.
create table dbo.BrokenField_Crosswalk (
BrokenField varchar(32) not null primary key
, CorrectedValue int
);
insert into dbo.BrokenField_Crosswalk (BrokenField,CorrectedValue) values
('ALFA', 1)
, ('ALPHA', 1)
, ('BRAVO', 2)
...
go
And your code for the merge would look something like this:
;with cte as (
select o.R_ID
, o.Field1
, BrokenField = cast(isnull(c.CorrectedValue,o.BrokenField) as int)
....
from oracle_table.bla as o
left join dbo.BrokenField_Crosswalk as c
)
merge into [mssqltable].[Mytable] t
using cte as s
on t.[R_ID] = s.[R_ID]
when matched
then update set
[Field1] = s.[Field1]
, ...
, [Brokenfield] = s.[BrokenField]
when not matched by target
then
If they are using names with a letter at the start that goes in a sequence:
A = 1
B = 2
C = 3
etc.
Then you could do something like this:
MERGE [mssqltable].[Mytable] as s
USING THEORACLETABLE.BLA as t
ON t.[R_ID], 1)) - ASCII('A') + 1 = s.[R_ID]
WHEN MATCHED THEN UPDATE SET [Field1] = s.[Field1], ..., [Brokenfield] = s.[BrokenField]
WHEN NOT MATCHED BY TARGET THEN
... another big statement
Edit: but actually I re-read your question and you are talking about [Brokenfield] being the problem column, so my solution wouldn't work.
I don't really understand now, as it seems as though the MERGE statement is updating the oracle table with numbers, so surely you need the mapping to work the other way, i.e. 1 -> ALFA, 2 -> BETA, etc.?
I am trying to remove entries from an array field that are found in a query.
tablename.listfield::integer[] has a full list
I am trying to remove a list of values from that field, which are gathered within the update query.
the ARRAY_REMOVE method only accepts single values, and the intarray module which has int[] - int[] doesn't seem to be an option.
the ARRAY[] && ARRAY[] can boolean return if there is overlap, but that doesn't help me
basically what I need is a real working version of this concept, which I know does not work.
UDPATE tablename SET listfield = ARRAY_REMOVE( listfield, ( select id from othertable ) )
is it possible to get this done with maybe a tricky CTE setup or something?
thanks!
I'm not sure why you say intarray doesn't seem to be an option, because it works just fine:
... SET listfield = listfield - ( SELECT array_agg(id) FROM othertable )
But if you want to do this without installing the extension, you can UNNEST the array and use the EXCEPT construct:
... SET listfield = ARRAY(SELECT UNNEST(listfield) EXCEPT SELECT id FROM othertable)
I have an xml column called OrderXML in an Orders table...
there is an XML XPath like this in the table...
/Order/InternalInformation/InternalOrderBreakout/InternalOrderHeader/InternalOrderDetails/InternalOrderDetail
There InternalOrderDetails contains many InternalOrderDetail nodes like this...
<InternalOrderDetails>
<InternalOrderDetail>
<Item_Number>FBL11REFBK</Item_Number>
<CountOfNumber>10</CountOfNumber>
<PriceLevel>FREE</PriceLevel>
</InternalOrderDetail>
<InternalOrderDetail>
<Item_Number>FCL13COTRGUID</Item_Number>
<CountOfNumber>2</CountOfNumber>
<PriceLevel>NONFREE</PriceLevel>
</InternalOrderDetail>
</InternalOrderDetails>
My end goal is to modify the XML in the OrderXML column IF the Item_Number of the node contains COTRGUID (like '%COTRGUID') AND the PriceLevel=NONFREE. If that condition is met I want to change the PriceLevel column to equal FREE.
I am having trouble with both creating the xpath expression that finds the correct nodes (using OrderXML.value or OrderXML.exist functions) and updating the XML using the OrderXML.modify function).
I have tried the following for the where clause:
WHERE OrderXML.value('(/Order/InternalInformation/InternalOrderBreakout/InternalOrderHeader/InternalOrderDetails/InternalOrderDetail/Item_Number/node())[1]','nvarchar(64)') like '%13COTRGUID'
That does work, but it seems to me that I need to ALSO include my second condition (PriceLevel=NONFREE) in the same where clause and I cannot figure out how to do it. Perhaps I can put in an AND for the second condition like this...
AND OrderXML.value('(/Order/InternalInformation/InternalOrderBreakout/InternalOrderHeader/InternalOrderDetails/InternalOrderDetail/PriceLevel/node())[1]','nvarchar(64)') = 'NONFREE'
but I am afraid it will end up operating like an OR since it is an XML query.
Once I get the WHERE clause right I will update the column using a SET like this:
UPDATE Orders SET orderXml.modify('replace value of (/Order/InternalInformation/InternalOrderBreakout/InternalOrderHeader/InternalOrderDetails/InternalOrderDetail/PriceLevel[1]/text())[1] with "NONFREE"')
However, I ran this statement on some test data and none of the XML columns where updated (even though it said zz rows effected).
I have been at this for several hours to no avail. Help is appreciated. Thanks.
if you don't have more than one node with your condition in each row of Orders table, you can use this:
update orders set
data.modify('
replace value of
(
/Order/InternalInformation/InternalOrderBreakout/
InternalOrderHeader/InternalOrderDetails/
InternalOrderDetail[
Item_Number[contains(., "COTRGUID")] and
PriceLevel="NONFREE"
]/PriceLevel/text()
)[1]
with "FREE"
');
sql fiddle demo
If you could have more than one node in one row, there're a several possible solutions, none of each is really elegant, sadly.
You can reconstruct all xmls in table - sql fiddle demo
or you can do your updates in the loop - sql fiddle demo
This may get you off the hump.
Replace #HolderTable with the name of your table.
SELECT T2.myAlias.query('./../PriceLevel[1]').value('.' , 'varchar(64)') as MyXmlFragmentValue
FROM #HolderTable
CROSS APPLY OrderXML.nodes('/InternalOrderDetails/InternalOrderDetail/Item_Number') as T2(myAlias)
SELECT T2.myAlias.query('.') as MyXmlFragment
FROM #HolderTable
CROSS APPLY OrderXML.nodes('/InternalOrderDetails/InternalOrderDetail/Item_Number') as T2(myAlias)
EDIT:
UPDATE
#HolderTable
SET
OrderXML.modify('replace value of (/InternalOrderDetails/InternalOrderDetail/PriceLevel/text())[1] with "MyNewValue"')
WHERE
OrderXML.value('(/InternalOrderDetails/InternalOrderDetail/PriceLevel)[1]', 'varchar(64)') = 'FREE'
print ##ROWCOUNT
Your issue is the [1] in the above.
Why did I put it there?
Here is a sentence from the URL listed below.
Note that the target being updated must be, at most, one node that is explicitly specified in the path expression by adding a "[1]" at the end of the expression.
http://msdn.microsoft.com/en-us/library/ms190675.aspx
EDIT.
I think I've discovered the the root of your frustration. (No fix, just the problem).
Note below, the second query works.
So I think the [1] is some cases is saying "only ~~search~~ the first node".....and not (as you and I were hoping)...... "use the first node..after you find a match".
UPDATE
#HolderTable
SET
OrderXML.modify('replace value of (/InternalOrderDetails/InternalOrderDetail/PriceLevel/text())[1] with "MyNewValue001"')
WHERE
OrderXML.value('(/InternalOrderDetails/InternalOrderDetail/PriceLevel[text() = "NONFREE"])[1]', 'varchar(64)') = 'NONFREE'
/* and OrderXML.value('(/InternalOrderDetails/InternalOrderDetail/Item_Number)[1]', 'varchar(64)') like '%COTRGUID' */
UPDATE
#HolderTable
SET
OrderXML.modify('replace value of (/InternalOrderDetails/InternalOrderDetail/PriceLevel/text())[1] with "MyNewValue002"')
WHERE
OrderXML.value('(/InternalOrderDetails/InternalOrderDetail/PriceLevel[text() = "FREE"])[1]', 'varchar(64)') = 'FREE'
Try this :
;with InternalOrderDetail as (SELECT id,
Tbl.Col.value('Item_Number[1]', 'varchar(40)') Item_Number,
Tbl.Col.value('CountOfNumber[1]', 'int') CountOfNumber,
case
when Tbl.Col.value('Item_Number[1]', 'varchar(40)') like '%COTRGUID'
and Tbl.Col.value('PriceLevel[1]', 'varchar(40)')='NONFREE'
then 'FREE'
else
Tbl.Col.value('PriceLevel[1]', 'varchar(40)')
end
PriceLevel
FROM (select id ,orderxml from demo)
as a cross apply orderxml.nodes('//InternalOrderDetail')
as
tbl(col) ) ,
cte_data as(SELECT
ID,
'<InternalOrderDetails>'+(SELECT ITEM_NUMBER,COUNTOFNUMBER,PRICELEVEL
FROM InternalOrderDetail
where ID=Results.ID
FOR XML AUTO, ELEMENTS)+'</InternalOrderDetails>' as XML_data
FROM InternalOrderDetail Results
GROUP BY ID)
update demo set orderxml=cast(xml_data as xml)
from demo
inner join cte_data on demo.id=cte_data.id
where cast(orderxml as varchar(2000))!=xml_data;
select * from demo;
SQL Fiddle
I have handled following cases :
1. As required both where clause in question.
2. It will update all <Item_Number> like '%COTRGUID' and <PriceLevel>= NONFREE in one
node, not just the first one.
It may require minor changes for your data and tables.