SQL Server 2012 :: Recursive Concatenation Of Parent Elements
Jul 28, 2015
I have a hierarchical structure for mapping products to categories, categories go 3 levels deep (depth is defined in articlegroups.catlevel, 0 being the main category and traversing down to lower category level 2). Also, a product may be in more than 1 category(!).
product details are stored in `[products]`
articlegroups are defined in `[articlegroups]`
and the mapping of the products to the articlegroups are defined in `[products_category_mapping]`
Now, I want to retrieve index the full category path for each item, so with the data provided below, I'd expect these 2 rows as a result:
Now I can get the separate fields via a statement like this:
SELECT ga.slug_nl as slug_nl_0
FROM articlegroups ga
INNER JOIN products_category_mapping pcm ON pcm.articlegroup_id=ga.id
INNER JOIN products gp on gp.id=pcm.artikelid
WHERE gp.id=2481446
Hi All, I have a requirement here to import data from XML file to SQL Database. The XML schema contains of various elements and one of the element is recursive ie. Parameter node contains parameter node within it and it can have n number of iterations. I have given the sample schema below:
But all the nodes contain the data which has to be imported to a single table dbo.Parameters. I cannot use Union ALL since i dont know how many iterations I will have in the file. Is there any way to do this operation in Data Flow Task using XML Source? Can anyone help me on this?
I am using the following code to get the next immediate parent of a customer in the hierarchy and it works fine. However, it works only for one customer at a time (i made this a scalar function and I pass the customer id as shown below).
How can I modify this so that i can get each customer and its next immediate parent (all of them in one shot in a single data set)?
WITH my_cte (CustomerID, ParentID, Level) AS ( SELECT CustomerID, ParentID, 0 AS Level FROM [dbo].MyTable WHERE CustomerID = @CustomerID
UNION ALL SELECT CustomerID, ParentID, Level + 1 FROM [dbo].MyTable e INNER JOIN my_cte AS cte ON e.CustomerID = cte.ParentID )
select CustomerID from my_cte WHERE Level <> 0 AND Level = (SELECT MIN(Level) FROM my_cte WHERE Level <> 0)
I need to write recursive query to find child of a parent until the last leaf. Below is my code.
;WITH Parent AS( SELECT [ParentID],Value FROM[DynamicColsValues_TP1] WHEREValue IS null UNION ALL SELECT t1.[ParentID],T1.Value, FROM DynamicColsValues_TP1 t1 INNER JOIN Parent t2 ON t1.[ParentID]=t2.[ParentID] ) SELECT * FROM Parent option (maxrecursion 0)
When I execute this code. It is returning me millions of rows. Whereas i have only 20 rows in a table max 40 rows it should return.
I have 8 fields - I have requirement to concatenate using '+' operator with semicolon delimiter but issues is in the
Output I get semicolons for the fields that are empty below is my code :
------------- case when [SLII Request Type] ='Job Posting' and [SmartLaborII Request Status] like 'Pending Approval (Level 4%' and [New Extension or Replacement Audit Flag] like 'FLAG%' then 'Reject – New, Extension, Replacement invalid entry' --'it is jp' else '' end as [ES Fully approved data 1], case
CREATE TABLE [dbo].[TransactionComponents]( [pkTransactionComponent] [int] IDENTITY(1,1) NOT NULL, [pkTransactionID] [int] NOT NULL, [ComponentID] [int] NULL ) ON [PRIMARY]
With the following data:-
INSERT [dbo].[TransactionComponents]([pkTransactionID], [ComponentID]) SELECT 1,5 UNION SELECT 1,6 UNION SELECT 1,7 UNION SELECT 1,8
[Code] ....
pkTransactionID and ComponentID both link to the same column on another table this enables a many to many relationship, what I need to figure out is a complete tree of relationships from one of the ID's in it. I think I need to write a recursive CTE to achieve this but I am not entirely sure how to write it. Below is my attempt:-
DECLARE @ID INT SET @ID = 1; WITH cteTxHeirachy (TxID, RelTxID, TxLevel)
[Code] ...
This returns:-
15 16 17 18 19 110
But the following are missing:-
10 2 2 11 2 12
3 and 4 should not be returned. I figured if I added the code that is commented out in the CTE that should give me everything but I think I get caught in an infinite loop.
in few words i need subtotal only for who have children.I tried with rollup but i wasn't able to have it similar to the aspected.I put it the level just to let clearer the dependencies.
Let's say I have a scalar functions that I'd like it's input to be the output from the previous row, with a recursive CTE I can do the following:
;with rCTE(iterations,computed) as ( select 0 [iterations], 1e+0 [computed] union all select iterations+1,dbo.f(computed) from where rCTE where iterations < 30 ) select * from rCTE
Thus for each iteration, is the nTh fold of the function f applied to 1. [e.g 5 is f(f(f(f(f(1)))))]
However, for some illogical reason this relatively simple function did lots of read and write in tempdb. Can I reform this query somehow to just use lag instead? I know for a fact I only want to get let's say 30 iterations. It'd be very nice to be able to enjoy a window spool which will spawn a worktable with minimal IO.
I know I can put 30 rows into a table variable and do a quirky update across it, but Is there a nice way to do this without doing some sort of hack.
CREATE TABLE #tree ( Childid varchar(20), Parentid varchar(20) ) INSERT INTO #tree (Childid,ParentId) SELECT '123' , null UNION ALL SELECT '456' , '123' UNION ALL SELECT '789' , '456' UNION ALL SELECT '870' , '456' UNION ALL SELECT '985' , '870';
I have made an SSRS report using the recursive parent functionality to show a hierarchical tree of values. The problem I have is that some children have more than one parent, but because (in order to use the recursive parent nicely) I need to group the results by Id, I only see distinct entries. This means that I only see each child once, even if it "should" appear in multiple locations in the report (under each of its parents).
I have a master table and i need to import the rows into the parent and child table.
Master table name is Flatfile_Inventory Parent Table name is INVENTORY Child Tables name are INVENTORY_AMOUNT,INVENTORY_DETAILS,INVENTORY_VEHICLE, Error details will be goes to LOG_INVENTORY_ERROR
I have 4 duplicate rows in the Flatfile_Inventory which i have already inserted in the Parent and child table.
Again when i run the query using stored procedure,its tells that all the 4 rows are duplicate and will move to the Log_Inventory_Error.
I need is if i have the duplicate rows in the flatfile_Inventory when i start inserting into the parent and child table the already inserted row have the unique ID i must identify it and delete that row in the both parent and chlid table.And latest row must get inserted into the Parent and child table from Flatfile_Inventory.
We have equipment table which stores Equipment_ID,Code,Parent_Id etc..for each Equipment_ID there is a Parent_Id. The PK is Equipment_ID Now i want to select the Code for the Parent_Id which also sits in the same table. All the Parent_Id's also are Equipment_ID's.
I've 2 tables QuestionAnswers and ConditionalQuestions and fetching data from them using CTE join and I'm seeing repetitive rows (not duplicate) like, If you have multiple answers for 1 question, the output is like
where london where paris where toronto
why us why japan why indonesia
I want to eliminate the repetitive question and group them as parent child items.
with cte as ( select cq.ConditionalQuestionID from ConditionalQuestions cq inner join QuestionAnswers qa on cq.QuestionID=qa.QuestionID where cq.QuestionID=5 and qa.IsConditional='Y') select distinct q.Question, a.Answer from QuestionAnswers qa inner join Answers a on a.AnswerID = qa.AnswerID inner join Questions q on q.QuestionID = qa.QuestionID inner join cte c on c.ConditionalQuestionID = qa.QuestionID;
I am having problem to apply updates into this function below. I triedusing cursor for updates, etc. but no success. Sql server keeps tellingme that I cannot execute insert or update from inside a function and itgives me an option that I could write an extended stored procedure, butI don't have a clue of how to do it. To quickly fix the problem theonly solution left in my case is to convert this recursive functioninto one recursive stored procedure. However, I am facing one problem.How to convert the select command in this piece of code below into an"execute" by passing parameters and calling the sp recursively again.### piece of code ############SELECT @subtotal = dbo.Mkt_GetChildren(uid, @subtotal,@DateStart, @DateEnd)FROM categories WHERE ParentID = @uid######### my function ###########CREATE FUNCTION Mkt_GetChildren(@uid int, @subtotal decimal ,@DateStart datetime, @DateEnd datetime)RETURNS decimalASBEGINIF EXISTS (SELECTuidFROMcategories WHEREParentID = @uid)BEGINDECLARE my_cursor CURSOR FORSELECT uid, classid5 FROM categories WHERE parentid = @uiddeclare @getclassid5 varchar(50), @getuid bigint, @calculate decimalOPEN my_cursorFETCH NEXT FROM my_cursor INTO @getuid, @getclassid5WHILE @@FETCH_STATUS = 0BEGINFETCH NEXT FROM my_cursor INTO @getuid, @getclassid5select @calculate = dbo.Mkt_CalculateTotal(@getclassid5, @DateStart,@DateEnd)SET @subtotal = CONVERT (decimal (19,4),(@subtotal + @calculate))ENDCLOSE my_cursorDEALLOCATE my_cursorSELECT @subtotal = dbo.Mkt_GetChildren(uid, @subtotal,@DateStart, @DateEnd)FROM categories WHERE ParentID = @uidENDRETURN @subtotalENDGORod
I'm playing with CTE and just want to expand my skills and ask how you would build this tree structure to fill that [Tree] column for table like in sample below:
/* CREATE TABLE #T1 (child_id INT, parent_id INT, tree VARCHAR(MAX))
I have set up an FTP connection that tests successfully. I can log on to the FTP site with the same credentials and see my root folder, and within that, two more folders. In my FTP task, I want to receive files to my local machine. The problem is that the only remote path available is not at the root level, and the only thing I can see are files from one of the child folders, but not the child folder I want.
Is it possible in the remote path to change folders? The remote path just shows a "/", which I thought would be the root level, but it is somehow linked to a child folder.I've tried various combinations to get to the folder I want, /root folder/child folder, but that gives the error that the folder does not exist.
If I open the FTP Task Editor, click on File Transfer, and click the ellipses for the Remote Path, the Browse For File box opens, with a "/" in the Location section, and a list of files in the child folder that I do not want to be in. If I click the "Up Directory" button, I get the message "Already at top level directory".
how I can get the files from one particular folder on the FTP site?
Hi,I am trying to concatenate 2 text fields to update a sql db column byusing:UPDATE MessagesSET private_messages = private_messages || @newmessageBut the stored procedure I am using doesn't accept ||, I've also triedCONCAT() and CAT() but so far this has failed. Is there another way Ican do this?Thanks in advance,Kevin
Is there a limit to how much text you can concatenate (other than datatype limits, of course)? For example, if you write:SELECT 'This is some text from: ' + Convert(char(15), SomeFieldName01)+ 'some more text' + Convert(char(25), SomeFieldName02) + 'yet moretext.' As BigNoteFROM TableIs there a limit on how much you can concatenate into that big notefield, meaning is there a limit other than the size of the data type?
I have a need to create a table in a sql server database from C# code. The kicker is that the user must be able to specify the table and field names via the UI. I can do a bit of sanity checking but as long as they enter something reasonable I need to accept it. Normaly I always ADO parameters to sanitise any user parameters but they can't be applied to table and field names, only values. As far as I'm aware that leaves me needing to concatenate strings and that's something I usually avoid like the plague due to risk of SQL injection.
My actual question : Assuming string concatenation is my only way forward, how can I sanitise the values that would go into the table name and fieldname bits of a CREATE TABLE statement to ensure that injection can't occur? I've been pondering it and I think I just need to check for semi-colons. Without a semi-colon I don't think a user could inject an extra statement could they?
Request ID Parent ID Account Name Addresss 1452 1254789 Wendy's Atlanta Georgia 1453 1254789 Wendy's Norcross Georgia 1456 1254789 Waffle House Atlanta Georgia
I have the below query which in turn goes in the where clause
Code:
SELECT CASE WHEN EXISTS (SELECT child.id FROM InspectionType parent, InspectionType child WHERE parent.id = child.parentId AND parent.code = 'INSPECTION_TYPE_SAFETY_LIFE' AND parent.zoneId = 10129) THEN (SELECT child.id FROM InspectionType parent, InspectionType child WHERE parent.id = child.parentId AND parent.code ='INSPECTION_TYPE_SAFETY_LIFE' AND parent.zoneId = 10129) ELSE (SELECT id FROM InspectionType WHERE code = 'INSPECTION_TYPE_SAFETY_LIFE' AND zoneId = 10129) END
I am getting below error
Subquery returned more than 1 value. This is not permitted when the subquery follows =, !=, <, <= , >, >= or when the subquery is used as an expression.
I tried to solve this problem with If and else but it is not going well with in-clause
Code: select * from InspectionType where id in (select IF EXISTS (SELECT child.id FROM InspectionType parent, InspectionType child WHERE parent.id = child.parentId AND parent.code = 'INSPECTION_TYPE_SAFETY_LIFE' AND parent.zoneId = 10129) BEGIN (SELECT child.id FROM InspectionType parent, InspectionType child WHERE parent.id = child.parentId AND parent.code ='INSPECTION_TYPE_SAFETY_LIFE' AND parent.zoneId = 10129) END ELSE (SELECT id FROM InspectionType WHERE code = 'INSPECTION_TYPE_SAFETY_LIFE' AND zoneId = 10129) )
I am using select * from InspectionType where id in() for illustation only. The sub-query results will go another query
Code: SELECT child.id FROM InspectionType parent, InspectionType child WHERE parent.id = child.parentId AND parent.code ='INSPECTION_TYPE_SAFETY_LIFE' AND parent.zoneId = 10129
This returns more than one row.I know why is the error message but need to achieve this functionality.
The MLID in the header table will be generated automatically.All the Parents will be stored in the HEADER and their childs in the DETAIL.When a child is added to a parent,the Parent's MLID will be stored in the MLID field in the DETAIL table with the newly added child.That child will come to the PARENT table when a child is added to that.The MLITEM id in the parent table can be repeated when that item undergoes a rivision.But the MLID for this will be a new one.An item in the Parent Table can have any number of childs and these childs can have any number of children(there is no limit for the level.)
Now if i select an item id '1000' (for example from the Header Table) with a concatstring (it could be without a concat string also).all its childs and their children should be printed in a report like the following
************************************************** *************************** I NEED TO CREATE THE TREE USING BOTH THE HEADER(MSTHDRML) AND THE DETAIL TABLE(MSTDTLML) ************************************************** ***************************
How can this be done.Is it necessary to use a recursive function in a stored procedure to generate this ...... i have never used Recursive function in SQL SERVER Stored Procedures.Can anyone help me on this(with Code).if not stored procedure, then what else can be done for this.
Dear fellows, Can anybody tell me how can i apply recusive/Tree query using select statement. For example I've a table structure for an Organization as follows:
emp_id emp_name supervisor_id ---------- --------------- ------------------- 101 ZAFIAN 102 BRUNNER 101 108 CALLAHAN 102 105 RUSSO 102 110 SIM 102 103 DUELL 101 and so on
1. How can I get the above records in Hirarchical format starting from top or from anywhere else in the hierarchy?
In Oracle it can be done as follows: SELECT emp_id,emp_name,supervisor_id FROM employee_tbl CONNECT BY supervisor_id = PRIOR emp_id START WITH supervisor_id is null;
Please reply me at the following address if possible: faisal@visualsoft-inc.com
INSERT INTO Details SELECT '','7419-01',0 union all SELECT '7419-01','44342-00',1 union all SELECT '7419-01','45342-00',1 union all SELECT '7419-01','46342-00',1 union all SELECT '7419-01','47342-00',1 union all SELECT '7419-01','48342-00',1 union all SELECT '7419-01','49342-00',1 union all
[code]....
From the above table data i want a search query , for example if I search data with "52342-00" I want output to be below format using CTE.
I have searched the net for atmost two days to find the solution of this problem but we not able to get the solution. I would appritiate if any one could help me in solving this issue:
I want to write a trigger for deleting the EMPLOYEE with EMPID=1 and the trigger should delete all the employees as there is cascading among them i.e EMPID 1 is the Manager of EMPID=2 and so on..
I found a solution at: http://msdn2.microsoft.com/en-us/library/aa902684(sql.80).aspx but the solution does not work when i try to implement it . It deletes the record for abc,xyz in the above table but rest are not deleted by the trigger.
Suppose there's a table named [Category], which has 2 columns: CategoryID int, ParentCategoryID int
Each category, except the top most, has a parent category, it's a 1-to-n parent-children relationship. Now I want to write a stored proc./function that accepts a CategoryID input parameter, and output all the descendent CategoryIDs (son, grand son, ...). How to do that in MSSQL 2000?
I have created a package which has to move huge volume of data in chuncks from source to target with a few sorter transformations and many lookup transformation. Once a chunck is completed the Execute SQL Task in the package updates the etldates into a control table against the chunkid. I want this run in a loop, as on completion of a chunk another should start and follow the same process until there are no more chuncks in the control table. First i execute the package manually, then i have written a trigger to call the SQL Server Job (etl) on update of the dates for the chunkids in the control table. The job when being called by the on update trigger is throwing up the following error. "The job failed. The Job was invoked by User fwt obin. The last step to run was step 1 (Package1)". Am i missing something basic here?
PS: So i tried putting the package into a for each loop container, performance suffers. A chunck which used to take 2 hours is taking more than 12 hours to complete. Then i removed sorter transformation and used db sorting in my source OLE db. Then thought of cacheing the lookups but did not know how and what to give so i gave up. While iam here, DB sorting/SSIS sorting? which is better?
