Hi Everyone,I've been given the painstaking project of splitting a single column into multiple columns and rows. I have a solution set up in which I will be posting further down the post but I want to see if there is a much more efficient solution to this.sample data:create table tbl_list(pk_int_itmid int(5) Primary Key,vchar_desk vchar(300));create table tbl_test1(fk_int_itmid int(5) references tbl_list(pk_int_itmid),vchar_itm varchar(60));insert into tbl_list values(1, 'this item');insert into tbl_list values(2, 'that item');insert into tbl_list values(3, 'those items');insert into tbl_test1 values(1, 'A, B - C, D, E - F, G, H - I');insert into tbl_test1 values(2, 'J, K - L, M, N - O');insert into tbl_test1 values(3, 'P, Q - R');into this table:create table tbl_output(fk_int_itmid int(5) references tbl_list(pk_int_itmid),vchar_itmA varchar(60),vchar_itmB varchar(60),vchar_itmC varchar(60));Output in comma delimited form:'1', 'A', 'B', 'C''1', 'D', 'E', 'F''1', 'G', 'H', 'I''2', 'J', 'K', 'L''2', 'M', 'N', 'O''3', 'P', 'Q', 'R'my current solution:create view vw_itm_a as select fk_int_itmid, substring(vchar_itm, 0, charindex('-',vchar_itm)) as vchar_itmA,substring(vchar_itm, charindex('-',vchar_itm)+1 , charindex(',',vchar_itm)-charindex('-',vchar_itm)) as vchar_itmB,substring(vchar_itm, charindex(',',vchar_itm)+1) as vchar_itmCfrom tbl_test1where charindex(',',vchar_itm) >1Gocreate view vw_itm_b as select fk_int_itmid, substring(vchar_itm, 0, charindex('-',vchar_itm)) as vchar_itmA,substring(vchar_itm, charindex('-',vchar_itm)+1 , charindex(',',vchar_itm)-charindex('-',vchar_itm)) as vchar_itmB,substring(vchar_itm, charindex(',',vchar_itm)+1) as vchar_itmCfrom vw_itm_awhere charindex(',',vchar_itmC) >1;Gocreate view vw_itm_c as select fk_int_itmid, substring(vchar_itmC, 0, charindex('-',vchar_itmC)) as vchar_itmA,substring(vchar_itmC, charindex('-',vchar_itmC)+1 , charindex(',',vchar_itmC)-charindex('-',vchar_itmC)) as vchar_itmB,substring(vchar_itmC, charindex(',',vchar_itmC)+1) as vchar_itmCfrom vw_itm_bwhere charindex(',',vchar_itmC) >1;Go;create view vw_itm_d asselect fk_int_itmid, vchar_itmA, vchar_itmB,substring(substring(vchar_itm, charindex(',',vchar_itm)+1), 0, charindex(',',vchar_itm)) as vchar_itmCfrom vw_itm_a ia union vw_itm_b ib on ia.fk_int_itmid = ib.fk_int_itmidGo;create view vw_itm_e asselect fk_int_itmid, vchar_itmA, vchar_itmB,substring(substring(vchar_itm, charindex(',',vchar_itm)+1), 0, charindex(',',vchar_itm)) as vchar_itmCfrom vw_itm_c ia union vw_itm_b ib on ia.fk_int_itmid = ib.fk_int_itmidGo;create view vw_itm asselect fk_int_itmid, vchar_itmA, vchar_itmC, vchar_itmC from vw_itm_awhere fk_int_itmid not in (select fk_int_itmid from vw_itm_b)unionselect fk_int_itmid, vchar_itmA, vchar_itmC, vchar_itmC from vw_itm_dunionselect fk_int_itmid, vchar_itmA, vchar_itmC, vchar_itmC from vw_itm_bwhere fk_int_itmid not in (select fk_int_itmid from vw_itm_c)unionselect fk_int_itmid, vchar_itmA, vchar_itmC, vchar_itmC from vw_itm_eunionselect fk_int_itmid, vchar_itmA, vchar_itmC, vchar_itmC from vw_itm_cGo;select fk_int_itmid, vchar_itmA, vchar_itmC, vchar_itmC into tbl_outputfrom vw_itmIs there a much more efficient manner of handling this column splitting?ThanksDC
This query was working well because I used to only be interested in one counter that was returned in the column, which was 'Free Megabytes'...I now have additional data that shows up as 'Total Disk Space'...Ideally, the query would return the total disk space next to the free megabytes on the same row for the same disk drive. Here is a couple rows of sample output:
AverageValueInstanceNameObjectNameCounterName 44549 C: LogicalDiskFree Megabytes 44548 C: LogicalDiskFree Megabytes 69452 C: LogicalDiskTotal Disk Space 69452 C: LogicalDiskTotal Disk Space
This is the ideal format, the average value column goes away:
InstanceNameObjectNameFree MegabytesTotal Disk SpaceC: LogicalDisk44549 69452
I have a table with millions of rows, 1 particular attribute "FromPerson" contains a string of email addresses, names, etc... for example (formatting done by this window... not exists in database...completely cleaned to semi-colon delimited)
tomh@gmail.com; Snyder, John N.; jsnyder@yahoo.com; Miller, Jim; millerj@gmail.com; Tenbrow, Jack; Katie Winslow (can be x number of names, emails...)
as you can see... some of the delimited values match up with an email address, others have just names, or emails... I have successfully split this into 1 row per value per say, but how do i get them to match up the values that do. I think the splitting is really not helping the cause... what i want is this...
tomg@hotmail.com tomg@hotmail.com Snyder, John N. jsnyder@yahoo.com Miller, Jim millerj@gmail.com Tenbrow, Jack Tenbrow, Jack Katie Winslow Katie Winslow
there could/can be other attributes, like datetime stamp, domain(if any)... etc... but I think I can add that later...
I have to split a column using comma a delimiter into multiple columns. I am able to do it if i know how many column will be present in the final output. But in daily run, the columns may vary randomly.
how to split columns without hardcoding how many columns it ll come.
This is the code am using
Code: WITH Split_Names (Fil_id,Name, xmlname) AS ( SELECT Fil_ID,
I have a table that has multiple postal codes in one of the columns. Those have to be split up one per line and stored in another table. The zip codes are comma seperated. Is there a function that can do this...?
Example data in ZipCodeTable. (Name and ZipCode are 2 columns in a table)
I am trying to return as a dataset of this that shows the ID as column 1 then each value in the 2nd column as individual columns if I use SELECT LEFT('2ndColumn,PATINDEX(',',2ndColum)-1) I can return the first value as a column but then can't return any further values individually after the first column, I am just learning the new functionality in SSIS so not sure whether this would be my answer as apposed to T-SQL, if anyone has any advice on this it would be greatly appreciated?
all the columns are separated with a "|" but the amount of columns are not fixed, so in lines 1 & 2 they are 4 columns and in line 3 there is 7 columns
Hi i want to create a table as follows :if exists (select * from dbo.sysobjects where id =object_id(N'[Indexes]') and OBJECTPROPERTY(id, N'IsUserTable') = 1)drop table [Indexes]GOCreate table Indexes(indexname Varchar(100), index_DescriptionVarchar(100), index_keys Varchar(100))GOINSERT INTO Indexes EXEC sp_helpindex 'SDM_Fact_Order_Detail'GOThis will give me a table (northwind)IX_Auto_SDM_Fact_FK_Shipped_Date nonclustered located onSAMIS_SDM_Index FK_Shipped_DateIX_Auto_SDM_Fact_Order_Detail_FK_Insert_Dateclustered located onSAMIS_SDM_Data1FK_Insert_Date, FK_Insert_TimeAs you see sp_helpindex will give me a comma seperated field. I wantto split the third column FK_Insert_Date, FK_Insert_Time into a extrarowLike this :IX_Auto_SDM_Fact_FK_Shipped_Date FK_Shipped_DateIX_Auto_SDM_Fact_Order_Detail_FK_Insert_Date FK_Insert_DateIX_Auto_SDM_Fact_Order_Detail_FK_Insert_Date FK_Insert_TimeCan anyone help me with this?ThanxHennie
Currently I have a column with multiple postcodes in one value which are split with the “/” character along with the corresponding location data. What I need to do is split these postcode values into separate rows while keeping their corresponding location data.
For example PostCode Latitude Longitude 66000/66100 42.696595 2.899370 20251/20270 42.196471 9.404951
Would become PostCode Latitude Longitude 66000 42.696595 2.899370 66100 42.696595 2.899370 20251 42.196471 9.404951 20270 42.196471 9.404951
I have a very interesting problem in T-SQL coding for which I can't figure out the solution. Actually there is a Line_1_Address column in our data warehouse address table which is being populated from various sources. Some sources have already concatenated house number + street address fields in the Line_1_Address column whereas one source has separated columns for both data fields.
Now I'm trying to extract data from this data warehouse table and I need to split the house number from street address and load it into separate columns in my destination table. In case there is no data for house number then I should load it as NULL.
The issue is that data in this Line_1_Address column is very inconsistent so I don't know which functions to use. Here is some sample data for your consideration:
Line_1_Address 101 E Commerce ST 120 E Commerce ST 2 Po Box 301 W. Bel Air Ave West Main Street, PO Box 1388
I have a description field in a table which also stores unit of measure in the same column but with some space between them, I need to split these into two different columns.
Currently I have a column with multiple postcodes in one value which are split with the “/” character along with the corresponding location data. What I need to do is split these postcode values into separate rows while keeping their corresponding location data.
I had a query regarding splitting MDF files. As one of our databases has an mdf sized 75GB, I was planning to split it into 2 or 3 files and put them across different RAID disks. But I could not find a thorough source for information as to how to go about splitting an existing MDF file into smaller files. Can anybody help?
I have an SSIS package that moves data from SQL Server to an legacy Access database. In SQL Server, there is a date/time column that I need to split into a separate date column and time column in the access database. Initially I just created a derrived column for the time and set the expression equal to the source date/time column from SQL Server. Unfortunately, that just makes the date column and time column the same having the full date/time.
What expression can I use during a derrived column transformation to assign just the date to a derrived column and just the time to another derrived column?
Hi,Suppose I have a table containing monthly sales figures from my shopbranches:Branch Month Sales-----------------------London Jan 5000London Feb 4500London Mar 5200Cardiff Jan 2900Cardiff Feb 4100Cardiff Mar 3500The question I am trying to ask is this: in which month did each branchachieve its highest sales? So I want a result set something like this:Branch Month----------------London MarCardiff FebI can do a "SELECT Branch, MAX(Sales) FROM MonthlySales GROUP BY Branch" totell me what the highest monthly sales figure was, but I just can't figureout how to write a query to tell me which month corresponded to MAX(Sales).Ideas anyone?Cheers,....Andy
-- The 3rd query uses an incorrect column name in a sub-query and succeeds but rows are incorrectly qualified. This is very DANGEROUS!!! -- The issue exists is in 2008 R2, 2012 and 2014 and is "By Design"
set nocount on go if object_id('tempdb.dbo.#t1') IS NOT NULL drop table #t1 if object_id('tempdb.dbo
[code]....
This succeeds when the invalid column name is a valid column name in the outer query. So in this situation the sub-query would fail when run by itself but succeed with an incorrectly applied filter when run as a sub-query. The danger here is that if a SQL Server user runs DML in a production database with such a sub-query which then the results are likely not the expected results with potentially unintended actions applied against the data. how many SQL Server users have had incorrectly applied DML or incorrect query results and don't even know it....?
do i need to nest a query in RS if i want a calculated column to be compared against a multi value variable? It looks like coding WHERE calcd name in (@variable) violates SQL syntax. My select looked like
SELECT ... ,CASE enddate WHEN null then 1 else 0 END calcd name FROM... WHERE ... and calcd name in (@variable)
I have a sql2005 db. I'm trying to perform a query that takes a list of values in a parm, breaks them up and uses them in an IN clause as follows: <code> ALTER PROCEDURE [dbo].[SelectPartialOrderListByDate] @StoreID varchar(255)ASBEGIN SET NOCOUNT ON; SELECT DISTINCT O.OrderID, M.InitialSalesCode, M.DateOccurred, M.ErrorCode FROM dbo.OrderInfo O WITH (NOLOCK) INNER JOIN dbo.Message M WITH (NOLOCK) ON O.OrderID = M.OrderID WHERE O.StoreID IN (@StoreID)END </code> The values coming into the procedure are obviously in a varchar string format. Notice the boldfacing above. The values within this IN clause should be something like this after the conversion: IN (87,108). How can I accomplish this? I've seen examples of creating a UDF and returning a table, but am not sure how to apply that to my situation via a code sample. Could someone help me out?
I founded couple thread about splitting but not find same situation what i have. I have server with Raid10 with big storage massive, and Test server with two 30Gb HardDrives (not into a Raid or Somethink). I have situation when in real server database grown over 30Gb and now i can not restore database copy into a test server because i have one big Data File.
Can i split somehow 35Gb Data file when i restore to test server to 25Gb and 10Gb ???
I have a table that contains agrrements and contracts with dates. Now I need to calculate some things and I'd like the rows to only have one month per row.
I have rows like:
Agreement, Start, End ID001, 2004-01-01, 2004-04-30
If I could get these single rows that contains 4 months into a temptable like this:
I everyone, I have been on the admin side of IT for the past 20 years and recently started to do some scripting (VBscript) and a little SQL.
I have developed a solution to meet the needs of some federal auditors, but not really met my needs yet. What I have done is this.
I use MS Logparser to go out to 64 servers and copy the event logs into a DB on a SQL 2000 Ent. Server.
On the SQL server I have one StoredProcedure that parses out information from the security event log DB and put that info into a temp DB.
set ANSI_NULLS ON set QUOTED_IDENTIFIER ON go
ALTER PROCEDURE [dbo].[stp_SecurityAuditReport]
AS
TRUNCATE TABLE SecurityEvents_Tmp
-- Parse Bank Number & UserName INSERT INTO SecurityEvents_Tmp(DepartmentNumber, UserName, EventLog, RecordNumber, TimeGenerated, TimeWritten, EventID, EventType, EventTypeName, EventCategory, EventCategoryName, SourceName, Strings, ComputerName, SID, Message, Data) SELECT DepartmentNumber = '001', UserName = CASE WHEN Strings LIKE '[0-9][0-9][0-9]%' THEN SUBSTRING(Strings,1,charindex('|',Strings,1)-1) WHEN Strings LIKE '-|[0-9][0-9][0-9]%' THEN SUBSTRING(Strings,3,charindex('|',Strings,3)-3) WHEN Strings LIKE '-|[a-z]%' THEN SUBSTRING(Strings,3,charindex('|',Strings,3)-3) WHEN Strings LIKE 'Account Unlocked. |%' THEN SUBSTRING(Strings,21,charindex('|',Strings,21)-21) ELSE SUBSTRING(Strings,1,charindex('|',Strings,1)-1) END, Events.* FROM Events JOIN EventsToLog on Events.EventID = EventsToLog.EventID WHERE SID NOT LIKE 'S-%'
-- Update blank usernames UPDATE SecurityEvents_Tmp SET UserName = 'NO USERNAME' WHERE UserName = '' OR UserName = '-'
-- Update DepartmentNumbers with zeros UPDATE SecurityEvents_Tmp SET DepartmentNumber = CASE WHEN UserName LIKE '[0-9][0-9][0-9][a-z]%' OR UserName LIKE '[0-9][0-9][0-9]#%' OR UserName LIKE '[0-9][0-9][0-9]$%' THEN SUBSTRING(UserName,1,3) ELSE '001' END
As you can see, we use 3 didgit numeric prefixes on all Departmental employee accounts. This is later used to produce departmenntal user audit reports.
I then have this script in a DTS that exports the report to an excel spreadsheet. (All works well for this purpose!)
DECLARE @TimeGenerated datetime SELECT @TimeGenerated = TimeGenerated FROM SecurityEvents_TimeGenerated
DECLARE @TimeGeneratedEnd datetime SELECT @TimeGeneratedEnd = TimeGeneratedEnd FROM SecurityEvents_TimeGenerated
SELECT DepartmentName = CASE WHEN b.DepartmentName IS NULL THEN 'All Department' ELSE b.DepartmentName END, a.EventID,d.EventDescription,a.UserName, a.TimeGenerated,c.Email1,c.Email2,c.Email3,c.Email4 FROM SecurityEvents_Tmp a LEFT JOIN DepartmentList b on a.DepartmentNumber = b.DepartmentNumber LEFT JOIN EmailToList c on b.DepartmentNumber = c.DepartmentNumber JOIN EventsToLog d on a.EventID = d.EventID WHERE b.Departmentnumber in (select Departmentnumber from Departmentlist) AND a.TimeGenerated BETWEEN @TimeGenerated AND @TimeGeneratedEnd ORDER BY b.DepartmentNumber,a.EventID,a.TimeGenerated
This combination of utils and scripts does very good for producing generic security reports for branch officers.
But now I am getting requests to justify/explain what is in these reports. The problem I have is that the information needed to delve further into the event logs is in a field called Strings. This field not only changes in length and the amount of fields within this string, but the information in this field changes depending on the type of event record it came from.
This is the Strings field from a failed logon (529) 200jenil|DOMAIN|10|User32 |Negotiate|SERVER|SERVER$|DOMAIN|(0x0,0x3E7)|6920|-|10.190.12.10|48397
And this is from Event ID 642 which was an account being created. -|381$cmiller|DOMAIN|%{S-1-5-21-3554868564-134719009-1577582102-7972}|Jmotta|DOMAIN|(0x0,0x58F635E)|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|-|%%1792|-|-
Now, my script does a good job of getting the first user name out but as in the 642 event the second users name would be useful as well. This is the person that created/modified the user account.
So what I was hoping was that I could use a function (or whatever) to automaticaly split the Strings value into it's individual components and put them into an auto-sizing temp table as something like Field1, Field2, Field3, and so on until the end of string.
I could then use a case to get the information needed.
I have a large db (sql 2005 express) that is, well, basically a mess. It is not normalized properly and contains masses amount of data (due to so much repeating data). To make a long story short - this db needs to be redesigned, but management said no, so that is not an option - so please, no one suggest that that's what I do.
My application creates reports based on this db - the problem is, the sp's are slow - and when a report needs to run several reports, it takes a long time to run. The sp's and db have been optimized as best I can (adding indexes etc.,)
I was wondering if there is a way to split the db - what I want to do is just retain, say 2 years of data in 1 db, and store the rest of the data in the other db, as 2 years worth of data is 95% of what will be queried. I did copy over 2 years worth for testing, and reports that took 30 minutes in the existing db, take less than 1 minute (sometimes even faster) in the new db - a huge improvement.
My problem is how to deal with the times that I need more than 2 years worth of data - how do I query both db's to get my application to read the data from both db's so that it seems that I am only running one db? The new db would be updated daily with new data, but not the old db - so if I had to query 10 years worth of data, I need 2 years from the first db, and then the 8 yeas from the second db.
If anyone can provide some feedback or point me in the right direction of what I should research in order to accomplish the above - I would appreciate it.
If anyone knows of a better solution - please don't be shy - speak up! :)
hiI am new to SQL.I have database of 30GB.I have just heard aboutspliting Databases which helps in Performance.So please can any oneguide me in what are the steps involved in it.Anxious to know How does it work if i split my database in twodifferent location.We are using SQl 2000.Operating System - Windows 2000 serverregardsTV
Hello,I have been placed in charge of migrating an old access based databaseover to sql server 7.0. So far, I have imported all the tables intosql server, but now I have come across the issue of needing to split astring variable. For instance, in the old database, the variable forname was such that it included both first and last names, whereas inthe new database there are seperate entities for first and last name.I know that there is a way to write a script that will separate out thetwo strings by using the "space" in between the name, but I'munfamiliar how to do this. Any suggestions? Thanks!Rick
I have a field that contains codes likefhj#asdskjjljlj#12And so on.What I want to do is create two new fields (field1 and filed2) thatsplit the original filed at '#'If a field does not contain '#' I would like its entire contents to besaved in field1.Also how do I ensure that I save these changes?Thanks fo any help in advance.Regards,Ciarn
