SQL

How do I convert an integer to a string in SQL Server?

If you want to reformat data, use the function convert as shown:

select convert(varchar, cast(18 as integer)) as 'a number'

SQL Server: Why do I get "multi-part identifier could not be bound" ?

Let's say you've developed a stored procedure that assigns a count value to a variable which looks like this:

SELECT @HasDefaultShipTo = count(*)
FROM StakeHolderLoc_Assoc
WHERE assoc.StakeHldrID = @StkhldrID

When you execute to compile you will see the error:

Msg 4104, Level 16, State 1, Procedure MetroDevETL_1, Line 80
The multi-part identifier "assoc.StakeHldrID" could not be bound.

The reason this error appears is because you have forgotten to associate/bind the table to "assoc". The corrected SQL will look like this:

SELECT @HasDefaultShipTo = count(*)
FROM StakeHolderLoc_Assoc assoc
WHERE assoc.StakeHldrID = @StkhldrID

A 'Get it Done Now!' Approach for transforming SQL Server Data to XML Data

First, for all of the MS SQL Server database purists out there that will complain vehemently against this method of extraction/formatting because of the special characters (SEE: http://www.webmonkey.com/reference/Special_Characters) that may (will likely) surface and need to be escaped, or that there's a better way to do this such as utilizing the XML Schema Collection/XML Explicit features available with SQL Server, etc. -- I recognize your concern, however, you'll have to admit, this is still (low-tech/mid-90's as it is) one of the fastest and most easily understood ways of simply selecting data out of existing tables and into an XML file with commonly available tools!

This method is intended for a one-time only conversion process. Anything else will require further examination of XML Explicit/XML Schema collection conversion processing.

..For that one-time conversion....here's what to do....

Let's say this is the XSD schema you have for storing Sales Agent data in XML:

<?xml version="1.0" encoding="utf-16"?>
<xsd:schema id="NewDataSet" xmlns="" xmlns:xsd="http://www.w3.org/2001/XMLSchema" xmlns:msdata="urn:schemas-microsoft-com:xml-msdata">
<xsd:element name="SalesRep">
    <xsd:complexType>
      <xsd:sequence>
        <xsd:element name="First" type="xsd:string" minOccurs="0" />
        <xsd:element name="Last" type="xsd:string" minOccurs="0" />
        <xsd:element name="Email" type="xsd:string" minOccurs="0" />
        <xsd:element name="Mobile" type="xsd:string" minOccurs="0" />
        <xsd:element name="Phone" type="xsd:string" minOccurs="0" />
        <xsd:element name="Fax" type="xsd:string" minOccurs="0" />
        <xsd:element name="Address1" type="xsd:string" minOccurs="0" />
        <xsd:element name="Address2" type="xsd:string" minOccurs="0" />
        <xsd:element name="City" type="xsd:string" minOccurs="0" />
        <xsd:element name="State" type="xsd:string" minOccurs="0" />
        <xsd:element name="Country" type="xsd:string" minOccurs="0" />
        <xsd:element name="CountryName" type="xsd:string" minOccurs="0" />
        <xsd:element name="Postal" type="xsd:string" minOccurs="0" />
        <xsd:element name="Territory" type="xsd:string" minOccurs="0" />
        <xsd:element name="TerritoryState" type="xsd:string" minOccurs="0" />
      </xsd:sequence>
    </xsd:complexType>
</xsd:element>
<xsd:element name="NewDataSet" msdata:IsDataSet="true" msdata:UseCurrentLocale="true">
    <xsd:complexType>
      <xsd:choice minOccurs="0" maxOccurs="unbounded">
        <xsd:element ref="SalesRep" />
      </xsd:choice>
    </xsd:complexType>
</xsd:element>
</xsd:schema>

.... take a close look at the elements, then frame the elements for each row in a SELECT statement...

(Here's how you would write that statement)

SELECT
'<SalesRep>' +
'<First>' + first + '</First>' +
'<Last>' + last + '</Last>' +
'<Email>' + email + '</Email>' +
case
    when Mobile is null then '<Mobile></Mobile>'
    when Mobile = '' then '<Mobile></Mobile>'
    else '<Mobile>' + PhoneCode + '-' + Mobile + '</Mobile>'
   end
+
case
    when Phone is null then '<Phone></Phone>'
    when Phone = '' then '<Phone></Phone>'
    else '<Phone>' + PhoneCode + '-' + Phone + '</Phone>'
   end
+
case
    when Fax is null then '<Fax></Fax>'
    when Fax = '' then '<Fax></Fax>'
    else '<Fax>' + PhoneCode + '-' + Fax + '</Fax>'
   end
+
'<Address1>' + Address1 + '</Address1>' +
   case
    when Address2 is null then '<Address2></Address2>'
    when Address2 = '' then '<Address2></Address2>'
    else '<Address2>' + Address2 + '</Address2>'
   end
+
   case
    when City is null then '<City></City>'
    when City = '' then '<City></City>'
    else '<City>' + City + '</City>'
   end
+
   case
    when State is null then '<State></State>'
    when State = '' then '<State></State>'
    else '<State>' + State + '</State>'
   end
+
'<Country>' + Country + '</Country>' +
'<CountryName>' + CountryName + '</CountryName>' +
'<Postal>' + Postal + '</Postal>' +
'<Territory>' + Territory + '</Territory>' +
   case
    when TerritoryState is null then '<TerritoryState></TerritoryState>'
    when TerritoryState = '' then '<TerritoryState></TerritoryState>'
    else '<TerritoryState>' + TerritoryState + '</TerritoryState>'
   end
+
'</SalesRep>'
FROM Metro_SalesForce

This statement yields rows that will look like this:

<SalesRep><First>Jack</First><Last>Sprat</Last><Email>Jack.Sprat@metro-design-dev.com</Email><Mobile>1-678-999-9999</Mobile><Phone>1-770-999-8888</Phone><Fax></Fax><Address1>1313 Mockingbird Lane</Address1><Address2></Address2><City>Conyers</City><State>GA</State><Country>USA</Country><CountryName>United States</CountryName><Postal>30013</Postal><Territory>Georgia</Territory><TerritoryState>GA</TerritoryState></SalesRep>

....Once you've executed the statement in SQL Server here's what's next....

1. Right click on the results grid to save the file in .csv format.
2. Open the file in Microsoft EXCEL. (Do NOT double click on the file to open Microsoft EXCEL. Instead, open MS EXCEL first--then follow the dialog box with the format prompting to assure that you are using comma delimited, rather than fixed-length or tab formatting.)
3. In EXCEL, do a Find and replace all -- searching for any special characters such as '&', etc.
4. Save the file.
5. Copy the file -- but use the file type .xml instead of .csv
6. Wrap the XML you have with the appropriate utf header AND node to manage the recursion of the rows (e.g. <SalesReps> . In this example that would be:

<?xml version="1.0" encoding="utf-16"?>
<SalesReps>

7. End the management node at the tail end of the file. In this case that would be:

</SalesReps>

8. Save the file-- which would look like this:

<?xml version="1.0" encoding="utf-16"?>
<SalesReps>
<SalesRep><First>Jack</First><Last>Sprat</Last><Email>Jack.Sprat@metro-design-dev.com</Email><Mobile>1-678-999-9999</Mobile><Phone>1-770-999-8888</Phone><Fax></Fax><Address1>1313 Mockingbird Lane</Address1><Address2></Address2><City>Conyers</City><State>GA</State><Country>USA</Country><CountryName>United States</CountryName><Postal>30013</Postal><Territory>Georgia</Territory><TerritoryState>GA</TerritoryState></SalesRep>
</SalesReps>

9. Double click on the file to open it. If the file comes up in your default browser or IDE bench you're DONE! If not, check your XML header/trailer first. If it still won't come up check for other possible escape characters using EXCEL.

That's all there is to it! (Really)

SQL Update Statement - using SubSelect

This is a simple, but helpful example of changing a boolean or bit flag in a table to 'True' when the condition in the subSelect is met. In this context, the SQL statement finds all the cases where a sales agent's email exists in the Metro_SalesForce table and exists in the related table Metro_SalesForceMult. The Metro_SalesForceMult table contains the USA state assignments only for sales agents representing more than a single state.

UPDATE Metro_SalesForce
SET multipleStates = 'True'
WHERE Metro_SalesForce.email in
(
     Select x.email
       From Metro_SalesForce x, Metro_SalesForceMult y
        Where x.email = y.email
)

How do I select a value from a table into a variable in SQL Server?

Here's an example...

DECLARE @LocID

SELECT @LocID = LocationID
FROM Locations
WHERE LocationName = 'WAREHOUSE X'

Does SQL server have an encryption function?

Yes. The functions are PWENCRYPT and PWCOMPARE. This basically allows the ability to encrypt a value on an insert/update and offer a comparison of the value on a select. There isn't a decryption function available.

Here are a few T-SQL statements to illustrate how pwencrypt/pwcompare work:

create table #MetroTest
(
UserLogIn varchar(10),
UserPass nvarchar(256)
)

insert #MetroTest
(UserLogIn,UserPass) values ( 'MaryMary', PWDENCRYPT('QuiteContrary'))

select UserLogIn, password = 'QuiteContrary', PWDCOMPARE('QuiteContrary', UserPass, 0) from #MetroTest
select UserLogIn, password = 'QuietContrary', PWDCOMPARE('QuietContrary', UserPass, 0) from #MetroTest

It is possible to write user defined functions which will encrypt/decrypt and compare. An excellent example of this is Peter Larsson's RC4 hash:

http://www.sqlteam.com/forums/topic.asp?TOPIC_ID=76258

Can I insert multiple rows with a single insert statement?

Yes! .....(Now, we're not talking about bulk inserts--which is a completely different task... we're just talking about a small number of rows that need to get inserted into a relatively static table)..

Typically folks do something like this to quickly add rows into a reference table:

INSERT INTO ReasonType
(DisplayName,Description)
VALUES
('Delay','Reason for a project delay')
GO
INSERT into ReasonType
(DisplayName,Description)
VALUES
('Cancellation','Reaon for a project cancellation')
GO

...but it is possible to do this instead...

INSERT INTO ReasonType (DisplayName, Description)
    SELECT 'Delay' ,'Reason for project delay'
    UNION ALL
SELECT 'Cancellation' ,'Reason for project cancellation'
    UNION ALL

How do I change the owner of my table in SQL Server?

Go to the ms sql server database instance that has the table object and open a new query window. Execute the system stored proc called sp_changeobjectowner. The 2 parameters it needs to run are: (1) the current owner+object name and (2) the target owner. Take a look at the example below:

exec sp_changeobjectowner 'METRO\paula.EmployeeContact', 'dbo'

When the stored proc is completed, refresh the instance. The table will then be owned by dbo. In this example, the table would display as dbo.EmployeeContact

Why does "use db" with a hyphen throw "Incorrect syntax near '-'" ?

I think this SQL Server x.x oddity goes back to the old Sybase 4 days!

If you want to use a database name that's been dubbed something like 'paula-objects' and type out:

use paula-objects

then execute the request, you'll get the error message: Incorrect syntax near '-'

just type this instead:

use [paula-objects]

when you execute the request, the syntax error will clear up!

Does SQL Server have an "ALL_TAB_COLUMNS" feature?

Yes!

For those Oracle X.Xers finding themselves working in the SQL Server 2005/2008 world these days, you'll be happy to know all you need to do is:

1. Open up a new execute query window.

2. Type the following:

use FavoriteDbInstance

SELECT
table_name,column_name
FROM
information_schema.columns
WHERE
column_name LIKE '%whateverYouWant%'

That's it!:-) Your table names with the column name you are searching for should show up!