Error: a PRIMARY KEY constraint cannot be defined on a column that allows NULLs, so the discussion is irrelevant to primary keys.

Should be: "On the other hand, a UNIQUE constraint, sorting, and grouping treat NULLs as equal:
* You cannot insert into a table two rows with a NULL in a column that has a UNIQUE constraint defined on it."

Should be: http://msdn2.microsoft.com/en-us/library/ms345130.aspx

The duration column reported in traces in SQL Server 2000 was expressed in millisecond units, but in SQL Server 2005 it is expressed in microseconds. Therefore, in order to calculate the duration in seconds, the value should be divided by 1000000 instead of by 1000. This is relevant to the queries in pages 101, 102 and 103. In the calculation of the result columns total_s and running_total_s the expression is currently: / 1000, and should be: / 1000000. Consequentially, the values in Table 3-13, columns total_s and running_total_s should reflect this change. Note that this does not have any impact on the percentage calculations discussed in that section.

2007-10-14
T. Wong

2006-05-27
Itzik Ben-Gan

2006-11-27
Bill Kan

2006-05-27
Lilach Ben-Gan

2007-04-29
Benjamin Nevarez

2006-05-27
Itzik Ben-Gan

2006-06-01
Stephen Hunt

2006-06-27
Itzik Ben-Gan

Error: the word "covering" was added in these tables/figures to the access method's name by mistake.

Should be: "Unordered Nonclustered Index Scan + Lookups"

2007-07-17
zhao lidong

2006-05-27
Itzik Ben-Gan

2006-05-27
Lilach Ben-Gan

2008-10-24
Rudolf

2006-05-27
Lilach Ben-Gan

2007-07-17
zhao lidong

Currently: " ..., and a third time with LIKE N'D%: "
Error: missing closing quote
Should be: " ..., and a third time with LIKE N'D%': "

Currently: "The predicate b NOT IN (a, b, NULL) therefore returns NOT UNKNOWN..."
Should be: "The predicate c NOT IN (a, b, NULL) therefore returns NOT UNKNOWN..."

Currently: "To make the NOT IN query logically equivalent to the EXISTS query..."
Should be: "To make the NOT IN query logically equivalent to the NOT EXISTS query..."

Page 206, 2nd paragraph, 2nd to last sentence

Currently:
"… the filter generates an empty set and the expression MIN(keycol) + 1 yields a NULL."

Should be:
"… the filter generates an empty set and the expression MIN(keycol + 1) yields a NULL."

Currently: "Changing the compatibility mode of a database to an earlier version will prevent you from using the new language elements (for example, ranking functions, recursive queries, and so on)."
Should be: "Changing the compatibility mode of a database to an earlier version will prevent you from using some of the new language elements (for example, PIVOT, UNPIVOT, and so on)."

Currently: "Note that the outer rows represent the points before the gaps because the next highest key value is missing."
Should be: "Note that the outer rows represent the points before the gaps because the next key value is missing."

Should be: "CHECKSUM returns an integer between -2147483648 and 2147483647."

Error: typo in text references to table name.

Should be: #CustsStage

Error: typo in text references to stored procedure name.

Should be: #usp_AsyncSeq

Should be:
UPDATE dbo.Orders
  SET (ShipCountry, ShipRegion, ShipCity) =
    (SELECT Country, Region, City
     FROM dbo.Customers AS C
     WHERE C.CustomerID = dbo.Orders.CustomerID)
WHERE CustomerID IN
  (SELECT CustomerID FROM dbo.Customers WHERE Country = 'USA');

2007-10-14
Matthew Bryde

2008-05-21
Dan Vilnoiu

2008-05-21
Dan Vilnoiu

2006-10-02
Feelanet

2007-03-04
Zhao Lidong

Currently: "In SQL Server 7.0, the BIT value was converted to an INT because that's how filters worked."
Should be: "In SQL Server 7.0, the INT value was converted to a BIT because that's how filters worked."

2006-11-27
Herbert Albert

Currently: “Note that as mentioned earlier in the chapter, the XML datatype can be adequate to store the state of the objects using XML serialization in a database as well.”

Error: this is mentioned later in the chapter; not earlier.

Should be: “Note that the XML datatype can be adequate to store the state of the objects using XML serialization in a database as well.”

2006-07-09
Tom Moreau

2006-07-06
Tom Moreau

2006-07-13
Stephen Hunt

Currently: "You can see that the second complex number (2, 3i) is sorted before the first one (2, 3i)"
Should be: "You can see that the second complex number (1, 7i) is sorted before the first one (2, 3i)"

2007-03-04
Zhao Lidong

Currently: "C# .NET-based ComplexNumberCS UDA"
Should be: "C# .NET-based ComplexNumberCS_SUM UDA"

2006-09-17
Tom Moreau, Feelanet

#Region "arithmetic operations"

    ' Addition
    Public Function AddCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)
        If Me.isNullValue Or c.isNullValue Then
            Return New ComplexNumberVB(True)
        End If
        ' addition
        Return New ComplexNumberVB(Me.Real + c.Real, _
            Me.Imaginary + c.Imaginary)
    End Function

    ' Subtraction
    Public Function SubCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)
        If Me.isNullValue Or c.isNullValue Then
            Return New ComplexNumberVB(True)
        End If
        ' addition
        Return New ComplexNumberVB(Me.Real - c.Real, _
            Me.Imaginary - c.Imaginary)
    End Function

    ' Multiplication
    Public Function MulCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)
        If Me.isNullValue Or c.isNullValue Then
            Return New ComplexNumberVB(True)
        End If
        ' addition
        Return New ComplexNumberVB(Me.Real * c.Real - _
          Me.Imaginary * c.Imaginary, _
            Me.Imaginary * c.Real + Me.Real * c.Imaginary)
    End Function

    ' Division
    Public Function DivCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)

          If Me.isNullValue Or c.isNullValue Then
              Return New ComplexNumberVB(True)
          End If
          ' addition
          Return New ComplexNumberVB(_
              (Me.Real * c.Real + Me.Imaginary * c.Imaginary) _
                / (c.Real * c.Real + c.Imaginary * c.Imaginary), _
              (Me.Imaginary * c.Real - Me.Real * c.Imaginary) _
                / (c.Real * c.Real + c.Imaginary * c.Imaginary) _
          )
    End Function

#End Region

#Region "arithmetic operations"

    ' Addition
    Public Function AddCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)
        If Me.isNullValue Or c.isNullValue Then
            Return New ComplexNumberVB(True)
        End If
        ' addition
        Return New ComplexNumberVB(Me.Real + c.Real, _
            Me.Imaginary + c.Imaginary)
    End Function

    ' Subtraction
    Public Function SubCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)
        If Me.isNullValue Or c.isNullValue Then
            Return New ComplexNumberVB(True)
        End If
        ' subtraction
        Return New ComplexNumberVB(Me.Real - c.Real, _
            Me.Imaginary - c.Imaginary)
    End Function

    ' Multiplication
    Public Function MulCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)
        If Me.isNullValue Or c.isNullValue Then
            Return New ComplexNumberVB(True)
        End If
        ' multiplication
        Return New ComplexNumberVB(Me.Real * c.Real - _
          Me.Imaginary * c.Imaginary, _
            Me.Imaginary * c.Real + Me.Real * c.Imaginary)
    End Function

    ' Division
    Public Function DivCN(ByVal c As ComplexNumberVB) As ComplexNumberVB
        'Null(checking)
        If Me.isNullValue Or c.isNullValue Then
            Return New ComplexNumberVB(True)
        End If
        'Zero checking
        If c.Real = 0 And c.Imaginary = 0 Then
            Throw New ArgumentException()
        End If
        ' division
        Return New ComplexNumberVB( _
            (Me.Real * c.Real + Me.Imaginary * c.Imaginary) _
              / (c.Real * c.Real + c.Imaginary * c.Imaginary), _
            (Me.Imaginary * c.Real - Me.Real * c.Imaginary) _
              / (c.Real * c.Real + c.Imaginary * c.Imaginary) _
          )
    End Function

#End Region

2006-07-06
Tom Moreau

Currently: "...and copy the .vsd files..."
Should be: "...and copy the .vdx files..."

2007-03-06
Zhao Lidong

Currently: "The method returns a scalar .value"
Should be: "The method returns a scalar value"

2006-07-13
Stephen Hunt

Currently: "After creating the Primary XML index, you can create up to three secondary XML Indexes:"
Should be: "After creating the Primary XML index, you can create three other types of secondary XML Indexes:"

2006-07-06
Tom Moreau

Currently: references in code and text to element/word "Hobbie"
Should be: "Hobby"

2006-10-25
Lorenzo Benaglia

Currently: "To explain other aspects of the XML data type, I'll need to insert a contact who speaks two foreign languages. Therefore, you will change the OtherAttributes XML column from schema-validated back to well-formed." 

Error: To obtain the same result, it's not required to remove the schema binding with the XML column, so you can validate the XML fragments against the XSD Schema.

Should be: "To explain other aspects of the XML data type, I'll need to insert a contact who speaks two foreign languages. Therefore, you will change the schema associated to the OtherAttributes XML column using the minOccurs and maxOccurs XML Representation Attributes."

You can find more details on the subject in SQL Server MVP Lorenzo Benaglia’s blog entry (http://blogs.dotnethell.it/lorenzo/Post_7895.aspx).

2007-03-06
Zhao Lidong

Currently: "...2) use a temporary table..."
Should be: "...2) using a temporary table..."

2006-07-13
Tom Moreau

Currently: "though proc2's code creates a table called #T2"
Should be: "though proc2's code creates a table called #T1"

2006-07-13
Tom Moreau

Currently: #T42
Should be: #T

2006-08-16
Feelanet

Currently: (n - 1) % 100000 + 1
Should be: (n - 1) % 10000 + 1

Note: this change will impact the cost estimates of the operators in Figure 2-3 and the statistics I/O measures that you will get when running the code against #T1. But the rest remains the same, and the points made in the discussion are just as valid.

2006-09-10
Itzik Ben-Gan

Currently: "Cursor code for custom aggregate"
Should be: "Cursor code for running aggregations"

2006-09-10
Itzik Ben-Gan

Currently: "Max Concurrent Sessions Benchmark"
Should be: "Running Aggregations Benchmark"

2007-10-14
Itzik Ben-Gan

This is not a correction rather a suggestion for additional reading (strongly recommended). SQL Server MVP Erland Sommarskog wrote a great paper about the subject here.

2006-08-04
Feelanet

Currently: "Use the sp_droplinkedsrvlogin stored procedure to map local logins to a security account on the remote server."
Should be: "Use the sp_addlinkedsrvlogin stored procedure to map local logins to a security account on the remote server."

2007-02-15
Zhao Lidong

Currently: comments saying "-- If only one row was inserted, don't use a cursor" and "-- If only multiple rows were inserted, use a cursor"
Should be: "-- If only one row was affected, don't use a cursor" and "-- If only multiple rows were affected, use a cursor"

2006-08-16
Feelanet

Currently: "Table 4-6 Contents of Computations Table After Inserts"
Should be: "Table 4-6 Contents of Computations Table After Updates"

2006-11-13
Morten Munch-Andersen

Currently:
"...
OR UPPER(@cols) LIKE UPPER(N'%xp_%')
OR UPPER(@cols) LIKE UPPER(N'%sp_%')
..."

Error: in the lines of code referring to the patterns N'%sp_%' and N'%xp_%', _ (underscored) is treated as a wildcard instead of as a character. The fix is to use N'%sp[_]%' and N'%xp[_]%'.

Should be:
"...
OR UPPER(@cols) LIKE UPPER(N'%xp[_]%')
OR UPPER(@cols) LIKE UPPER(N'%sp[_]%')
..."

2006-07-23
Tom Moreau

Currently: "I will discussed..."
Should be: "I will discuss..."

2006-06-26
Benjamin Nevarez

2006-08-16
Feelanet

2007-07-17
Matthew Wise, zhao lidong

2007-10-14
Erland Sommarskog

Currently: "The stored procedure and the underlying objects belong to the same schema."

Should be: "The stored procedure and the underlying objects belong to the same owner."

2006-11-13
Morten Munch-Andersen

Currently:
"...
OR UPPER(@cols) LIKE UPPER(N'%xp_%')
OR UPPER(@cols) LIKE UPPER(N'%sp_%')
..."

Error: in the lines of code referring to the patterns N'%sp_%' and N'%xp_%', _ (underscored) is treated as a wildcard instead of as a character. The fix is to use N'%sp[_]%' and N'%xp[_]%'.

Should be:
"...
OR UPPER(@cols) LIKE UPPER(N'%xp[_]%')
OR UPPER(@cols) LIKE UPPER(N'%sp[_]%')
..."

2007-01-21
Chris Bohling

SELECT COLUMN_NAME AS updated_column
FROM INFORMATION_SCHEMA.COLUMNS AS C JOIN @UpdCols AS U
  ON C.ORDINAL_POSITION = U.ordinal_position
WHERE TABLE_SCHEMA = 'dbo'
  AND TABLE_NAME = 'T1'
ORDER BY C.ORDINAL_POSITION;

Error: according to Books Online (http://msdn2.microsoft.com/en-us/library/ms186329.aspx): "In SQL Server 2005, the ORDINAL_POSITION column of the INFORMATION_SCHEMA.COLUMNS view is not compatible with the bit pattern of columns returned by COLUMNS_UPDATED. To obtain a bit pattern compatible with COLUMNS_UPDATED, reference the ColumnID property of the COLUMNPROPERTY system function when you query the INFORMATION_SCHEMA.COLUMNS view."

Should be:

SELECT COLUMN_NAME AS updated_column
FROM INFORMATION_SCHEMA.COLUMNS AS C JOIN @UpdCols AS U
  ON COLUMNPROPERTY(OBJECT_ID(QUOTENAME(TABLE_SCHEMA)
                              + '.' + QUOTENAME(TABLE_NAME)),
                    COLUMN_NAME, 'ColumnID') = U.ordinal_position
WHERE TABLE_SCHEMA = 'dbo'
  AND TABLE_NAME = 'T1'
ORDER BY U.ORDINAL_POSITION;

2007-03-02
Zhao Lidong

2006-07-26
Philipp Ohnemus

2006-08-06
Tom Moreau

Currently the .query method is used to extract an element from the XML value containing the DDL event info. Specifically with DDL triggers this works since currently there are no cases with multiple elements with the same name at the same level. But in general, XML allows multiple elements with the same name at the same level (in which case the .query method would return all of them). In general, when you're after a single value, it is more appropriate to use the .value method where you can specify an index (ordinal).
So in all of the code samples that are supposed to extract a scalar value instead of using the .query method, use the following form of the .value method:
xml_value.value('(<path_to_element>)[1]', <sql_type>)
For example, here's the expression you need to use in the code samples in the chapter to extract the object name from the variable @eventdata:
@eventdata.value('(/EVENT_INSTANCE/ObjectName)[1]', 'sysname')
You can find the revised source code for DDL Triggers in Chapter 8 here.

2007-03-02
Zhao Lidong

2007-03-02
Zhao Lidong

2007-03-02
Zhao Lidong

Page 357, Listing 8-14, first line comment in second block of code

2006-06-13
Benjamin Nevarez

2006-06-13
Benjamin Nevarez

2006-09-22
Feelanet

2007-02-15
Zhao Lidong

2007-03-02
Peter Bourlet

2007-03-02
Zhao Lidong

2007-03-02
Zhao Lidong

2007-03-02
Zhao Lidong

2007-03-02
Zhao Lidong

Page 446, third paragraph, first sentence

2007-03-02
Zhao Lidong

Page 452, last code line comment in page

2007-02-15
Zhao Lidong

2007-02-15
Zhao Lidong