>>I am trying to build a dynamic UPDATE statement in a SQL Server Stored Procedure.
>>
>>I pass to the SQL Stored procedure an XML string (@XmlDocWo) that has field name/value pairs. Then I extract a value for a field from the XML as following:
>>
>>

>>SELECT @Fld1Value = [Fld FldName].value('.','char(10)') FROM @XmlDocWo.nodes('/TABLE_NAME/Fld') Fld([Fld FldName]) 
>>	 WHERE [Fld FldName].value('@name','varchar(20)') = 'FLD1NAME'
>>
>>SELECT @ Fld2Value = [Fld FldName].value('.','char(10)') FROM @XmlDocWo.nodes('/TABLE_NAME/Fld') Fld([Fld FldName]) 
>>	 WHERE [Fld FldName].value('@name','varchar(20)') = 'FLD2NAME'
>>

>>
>>Then the Update command would be as follows:
>>
>>

>>UPDATE MyTable SET FLD1NAME = @Fld1Value, FLD2NAME = @Fld2Value where PK_FLD = @PkFldValue
>>

>>
>>The above expression would have more than 2 fields, of course. And I am looking for the way to eliminate one or some fields if they are not included in XML. For example, if XML string does not have value for FLD2NAME the UPDATE expression would be:
>>
>>

>>UPDATE MyTable SET FLD1NAME = @Fld1Value where PK_FLD = @PkFldValue
>>

>>
>>How would you suggest I go about it?
>
>By default all declared variables will be NULL. Assuming you will not be having NULL values in your XML file you can do
>
>update myTable SET Fld1Name = COALESCE(@Fld1Value, Fld1Name), etc.

I didn't know that I can skip declaring a variable. I thought that I always have to do
declare @MyVarName INT before using it.

I have never used COALESCE() either. The approach you describe should work. I will test it.
Thank you.