Hi everyoneI guess this should be a simple question for the gurusI have a Data in a column which is to be places in 2 columns instead ofone. How do i go about doing it in MS SQL server? Could someone pleasehelp me. I could do it in access with an update query but things are alittle different in SQL server so I am a little lost.Eg.NameJohn?Doeto be split intoName LastNameJohn DoeThanks in advance.Prit
I like to push 1 column into 2 different columns just to show it on the screen. So no import in another table ore something like that. I have a table like this: Select Name from Cars; Result: Col1 BMWMercedesFordAudi But i like to make a query so it is displayed like this: Col1 Col2 BMW FordMercedes Audi So i can bound a table directly to that column!Is this possible with SQL, and how can i build it.Thanks.
Hi all, I have a requirement like this , I have Address Column.It is containing data like Mr. K KK Tank Guntur Jal Bhavan, Univercity Road, Rajkot 9843563469 I have to split this into 3 more columns like(Address1,name,phoneno)-- Means i have 4 columns including Address Column.(Address,Address1,name,phoneno)
Example: Address:Rajkot Address1:Univercity Road Name:Mr. K KK Tank Guntur Jal Bhavan PhoneNO:9843563469
How can i acheive this one with out data lose in Address Column. Thanks in advance.
Hi This is probably a very basic question for most people in this group. How do i split the data in a column in to 2 columns? This can be done in access with an update query but in MS SQL server I am not sure. Here is an example of what i want to acheive
I have a table with a string value, where all values are seperated by a space/blank. I now want to use SQL to split all the values and insert them into a different table, which then later will result in deleting the old table, as soon as I got all values out from it.
Old Table:
Code: ID, StringValue
New Table:
Code: ID, Value1, Value2 Do note: Value1 is INT, Value2 is of nvarchar, hence Value2 can contain spaces... I just need to split on the FIRST space, then convert index[0] to int, and store index[1] as it is.
I can split on all spaces and just Select them all and add them like so: SELECT t.val1 + ' ' + t.val2... If I cant find the first space that is... I mean, first 2-10 characters in the string can be integer, but does not have to be.Shall probably do it in code instead of SQL?Now I want to run a query that selects the StringValue from OldTable, splits the string by ' ' (a blank) and then inserts them into New Table.
Code: SELECT CASE CHARINDEX(' ', OldTable.stringvalue, 1) WHEN 0 THEN OldTable.stringvalue ELSE SUBSTRING(OldTable.stringvalue, 1, CHARINDEX(' ', OldTable.stringvalue, 1) - 1) END AS FirstWord FROM OldTable
Found an example using strange things like CHARINDEX..But issue still remains, because the first word is of integer, or it does not have to be...If it isn't, there is not "first value", and the whole string shall be passed into "value2".How to detect if the very first character is of integer type?
Code: @declare firstDigit int IF ISNUMERIC(SUBSTRING(@postal,2,1) AS int) = 1 set @firstDigit = CAST(SUBSTRING(@postal,2,1) AS int) ELSE set @firstDigit = -1
Our front end saves all IP addresses used by a customer as a comma separated string, we need to analyse these to check for blocked IPs which are all stored in another table.
A LIKE statement comparing each string with the 100 or so excluded IPs will be very expensive so I'm thinking it would be less so to split out the comma separated values into tables.
The problem we have is that we never know how many IPs could be stored against a customer, so I'm guessing a function would be the way forward but this is the point I get stuck.
I can remove the 1st IP address into a new column and produce the new list ready for the next removal, also as part of this we would need to create new columns on the fly depending on how many IPs are in the column.
This needs to be repeated for each row
SELECT IP_List , LEFT(IP_List, CHARINDEX(',', IP_List) - 1) AS IP_1 , REPLACE(IP_List, LEFT(IP_List, CHARINDEX(',', IP_List) +0), '') AS NewIPList1 FROM IpExclusionTest
I have a string that contains series of parameters with separators.i need to split the parameters and its values as rows and columns.e.g string = "Param1 =3;param2=4,param4=testval;param6=11;..etc" here the paramerter can be anything and in any number not fixed parameters. Currently am using the below function and getting the parameters by each in select statement as mentioned below.
select [dbo].[rvlf_fn_GetParamValueWithIndex]('Param1=3;param2=4,param4=testval;param6=11;','param1=',';') as param1, [dbo].[rvlf_fn_GetParamValueWithIndex]('Param1=3;param2=4,param4=testval;param6=11;','param2=',';') as param2 CREATE FUNCTION [dbo].[rvlf_fn_GetParamValueWithIndex] ( @CustomProp varchar(max),
Create table DBInfo (Path varchar (500)) Insert into DBInfo values('/Data Sources') Insert into DBInfo values('/Data Sources/SALES') Insert into DBInfo values('/PRODUCTION') Insert into DBInfo values('/PRODUCTION/SERVICE') Insert into DBInfo values('/PRODUCTION/SERVICE/MAINTENANCE') Insert into DBInfo values('/PRODUCTION/SERVICE/LOGISTICS')
My Expected Output
Column1,Column2,Column3 Data SourcesNullNull Data SourcesSalesNull PRODUCTIONNullNull PRODUCTIONSERVICENull PRODUCTIONSERVICEMAINTENANCE PRODUCTIONSERVICELOGISTICS
I have a table of users. Every user has an e-mail address and (hashed) password. Some of those users work for a company, and some of them do not. Of those who do not work for a company, some are salespeople who sell to one or more companies. Some users are simply administrators who don't work for a specific company. So here's what my users table looks like right now: "UserID, Email, Password, CompanyID (Nullable), IsAdmin" And here's my companies table: "CompanyID, CompanyName, SalespersonID"
Of course, I could separate it out and make a Users table, an Employees table, and a Salespeople table. The way the relationship works out, though, I could use the same ID number for all three tables, and that indicates to me that perhaps they all belong in the same table. It seems silly, after all to have a Salespeople table whose only field is "UserID."
Two factors of the first design concern me: First is the fact that a salesperson could also have a company. I guess I could write a check constraint to prevent this, but doesn't having the companyID in the Users table violate a normalization rule? Maybe? The second is the fact that the Companies table relies upon Users, which in turn relies upon Companies. In OOP, this usually isn't a good thing, but I'm not sure whether it's cause for concern in a relational database.
Anyway, I really don't know what I should be doing with this design. Any suggestions?
I've a table that has salescode(124!080) and salesamount(125.65!19.25) and I need to split the columns. Salesman(124) has commission(125.65). Here is the DDL:
USE tempdb; GO DECLARE @TEST_DATA TABLE ( DT_ID INT IDENTITY(1,1) NOT NULL PRIMARY KEY CLUSTERED , InvNoVARCHAR(10) NOT NULL , SalesCode NCHAR(80) NOT NULL , Amount NCHAR(80) NOT NULL
I am trying to join two tables and looks like the data is messed up. I want to split the rows into columns as there is more than one value in the row. But somehow I don't see a pattern in here to split the rows.
This how the data is
Create Table #Sample (Numbers Varchar(MAX)) Insert INTO #Sample Values('1000') Insert INTO #Sample Values ('1024 AND 1025') Insert INTO #Sample Values ('109 ,110,111') Insert INTO #Sample Values ('Old # 1033 replaced with new Invoice # 1544') Insert INTO #Sample Values ('1355 Cancelled and Invoice 1922 added') Select * from #Sample
This is what is expected...
Create Table #Result (Numbers Varchar(MAX)) Insert INTO #Result Values('1000') Insert INTO #Result Values ('1024') Insert INTO #Result Values ('1025') Insert INTO #Result Values ('109') Insert INTO #Result Values ('110')
[Code] ....
How I can implement this ? I believe if there are any numbers I need to split into two columns .
So I have been trying to get mySQL query to work for a large database that I have. I have (lets say) two tables Table_One and Table_Two. Table_One has three columns: Type, Animal and TestID and Table_Two has 2 columns Test_Name and Test_ID. Example with values is below:
In Table_One all types come under one column and the values of all Types (Mammal, Fish, Bird, Reptile) come under another column (Animals). Table_One and Two can be linked by Test_ID
I am trying to create a table such as shown below:
This should be my final table. The approach I am currently using is to make multiple instances of Table_One and using joins to form this final table. So the column Bird, Reptile, Mammal and Fish all come from a different copy of Table_one.
For e.g
Select Test_Name AS 'Test_Name', Table_Bird.Animal AS 'Birds', Table_Mammal.Animal AS 'Mammal', Table_Reptile.Animal AS 'Reptile, Table_Fish.Animal AS 'Fish' From Table_One
[Code] .....
The problem with this query is it only works when all entries for Birds, Mammals, Reptiles and Fish have some value. If one field is empty as for Test_Two or Test_Three, it doesn't return that record. I used Or instead of And in the WHERE clause but that didn't work as well.
Deciding whether or not to use a CTE or this simple faster approach utilizing system tables, hijacking them.
SELECT s.ORDER_NUMBER, s.PRODUCT_ID, 1 AS QTY, s.VALUE/s.QTY AS VALUE FROM @SPLITROW s INNER JOIN master.dbo.spt_values t ON t.type='P' AND t.number BETWEEN 1 AND s.QTY
Just wanted to know if its okay to use system tables in a production environment and if there are any pit falls of using them ?
I'm using SQL Server 2012 and I need to run a query against my database that will output the difference between 2 dates (namely, DateOfArrival and DateOfDeparture) into the correct month column in the output.
Both DateOfArrival and DateOfDeparture are in the same table (let's say GuestStay). I will also need some other fields from this table and do some joins on some other tables but I will simplify things so as to solve my main problem here. Let's say the fields needed from the GuestStay table looks like below:
I need my query to output in the following format:
I am trying to get an address field into 2 colums. I need the number value in one column and street name in another column.
The data is stored: 876 blue ct 9987 red dr 23 windyknoll
This is what I haveelect substring(Address,0,charindex('',Address)) as number ,substring(Address, (charindex('',Address)+1) ,len(Address)) as address from contact
i have labels for data stored in one cell eg: item1; item22; item231; and i want to convert it in following output (probably using substring and charindex)
Hoping someone here can help. Perhaps I'm missing something obvious, but I'm surprised not to see a data flow task in SSIS for splitting *columns* to different destinations. I see the Conditional Split task can be used to route a *row* one way or another, but what about columns of a single row?
As a simple and somewhat contrived example, let's say I have a row with twelve fields and I'm importing the row into a normalized data structure. There are three target tables with a 1-to-1 relationship (that is, logically they are one table, but physically they are three tables, with one of them considered the "primary" table), and the twelve input fields can be mapped to four columns in each of the three tables.
How do I "split" the columns? The best way I can see is to Multicast the row to three different OLE-DB Destinations, each of which inserts to one of the three target tables, only grabbing the four fields needed from the input row.
Or should I feed the row through three successive OLE-DB Command tasks, each one inserting into the appropriate table? This would offer the advantage, theoretically, of allowing me to grab the identity-based surrogate primary key from the first of the three inserts in order to enable the two subsequent inserts.
I would appreciate any help on this project. I have created an Access database that contains one vehicle. I have also included all options on that vehicle, which are in one column. Therefore the main criteria for the vehicle is listed each time for each different option. On my report I am grouping by the vin and placing the main criteria in the group header area of the report. The options are going into the detail section. How do I get the options to print in two columns within the detail section? I am unable to find any help on this subject, so I am asking you for help.
I've a table similar to the one below, with a SKU, Category and Cost, and need using a simple select command, split the cost in two columns one for each category (1,2), I used a self-join, and it works, but it doesn't show values not equal in both categories
Declare @Tmp_SKUCatValue Table( SKU char(7) ,Cetegory Int ,Unit_cost Decimal ); INSERT INTO @Tmp_SKUCatValue (SKU, Cetegory,Unit_cost) Values ('sku-001',1,120)