Errata

You are instructed to delete the file checkpoint.xml - the file does not exist because it was deleted automatically in step 6 - which executes successfully.

Note from the Author or Editor:
FROM:

In Windows Explorer, delete the checkpoint.xml file under C:\TK463\Chapter08 \Lesson2\Starter\.

TO:

If you did not finish the previous exercise, please check if the checkpoint.xml file exists and delete it (in Windows Explorer, delete the checkpoint.xml file under C:\TK463\Chapter08\Lesson2\Starter\ ).

GJ

" ... select DeleteOutput ... - this happens automatically as the 2 other options are already selected.

Note from the Author or Editor:
Please correct the step 21 on page 314 from:

Inside the IncLoadStgSalesHeader package, add another OLE DB destination adapter. Connect the CDC Splitter transformation with the new OLE DB destination adapter and, in the Input Output Selection dialog box, select DeleteOutput in the Output drop-down list.

TO:

Inside the IncLoadStgSalesHeader package, add another OLE DB destination adapter. Connect the CDC Splitter transformation with the new OLE DB destination adapter.

FillStageTables, Customer Information flat file import fails because BirthDate type is set incorrectly. If I go to Advanced Editor, Input and Output Properties and change the type to DT_Date, the task is successful.

I saw the errata on Page 196, Step 12, but that says DT_I4 should work, and that's what the Starter and Solution of Ch. 6, Lesson 3 are set at as well as several previous lessons. DT_I4 doesn't work on my system.

thanks,

John

Note from the Author or Editor:
I was unable to reproduce the problem, even though there actually is a mismatch between the data types in the source file and the data types in the destination table.


ML


[UPDATE, March 17th 2014]:

After a conversation with Grega Jerkič, the author of Chapter 5, I have realized that the changes made in that Chapter may have been the cause of the problem in Chapter 6. I have now been able to reproduce it, and to correct it appropriately.

I will send the new sample files for Lesson 3 of Chapter 6 via email.


ML

In table 2-7 the remarks for column MonthNumberName state that it should concatenate MonthNumberOfYear and EnglishMonthName.

In the INSERT statement, the MonthNumberName column is concatenating SUBSTRING(CONVET(CHAR(8), FullDateAlternateKey,112), 5,2) (not MonthNumberOfYear as it states in table 2-7) and EnglishMonthName.

Should
SUBSTRING(CONVET(CHAR(8), FullDateAlternateKey,112), 5,2)
be replaced with
Format(MonthNumberOfYear,'00')

Note from the Author or Editor:
Nice finding. Although both the code and the instructions are correct, a different technique is used in the code. Please correct in the code as stated in the error. Replace

SUBSTRING(CONVET(CHAR(8), FullDateAlternateKey,112), 5,2)

with

FORMAT(MonthNumberOfYear,'00')

As the reader, I would also prefer to correct the code than the instructions, because the newly suggested code is more SQL Server 2012 oriented.

Dejan Sarka

When running this step, the following error is encountered:

"Error: 0xC020803C at Dim Customer, Slowly Changing Dimension [239]: If the FailOnFixedAttributeChange property is set to TRUE, the transformation will fail when a fixed attribute change is detected. To send rows to the Fixed Attribute output, set the FailOnFixedAttributeChange property to FALSE."

I have used the starter code from the Code Lab Solutions collection, ensured that the various Gender and MaritalStatus columns have the NCHAR(5) data type (covered under the ?Confirmed Errata?) and set the data access mode for the 'Insert Destination' transformation (generated by the SCD transformation) to 'Table or view - fast load'

I believe the issue lies with the Gender column (which is defined as a fixed attribute for the SCD) and the 'Set default values' transformation. This transformation creates a derived column based on Gender and returns "N/A" when a NULL value is encountered. Given the Gender column in dbo.Customers is of type NCHAR(5), the SCD transformation will be evaluating "N/A" vs "N/A " - hence the detection of a fixed attribute change.

Changing the derived column expression for Gender within the 'Set default values' transformation to: 'ISNULL(Gender) ? "N/A " : Gender' allows the data flow to complete successfully

Note from the Author or Editor:
Thank you and you are correct. There should be a NVARCHAR(5) data type and not NCHAR(5), so that you do not need to modify the SSIS. I will upload the new exercises.

Thank you and regards

Grega

A step should be added before Step 1 or the following information appended to Step 1 with guidance on how to enable MDS integration with Data Quality Services. Since this had not been completed prior to this final chapter, the Data Quality group of the Master Data tab will not be activated within Excel. Here is suggested verbiage:
Open the Master Data Services Configuration Manager console. You can find it in All Programs | Microsoft SQL Server 2012 | Master Data Services. Click "Web Configuration" in the left pane. In the Website drop-down in the right pane, select MDS. Lastly, click the "Enable integration with Data Quality Services" button in the bottom of the right pane. Click the "Exit" button.

Note from the Author or Editor:
Add additional text at the end of the intro text for the exercise 2, before the first step. Text should be:
"If you did not enable MDS integration with DQS yet, the Data Quality group of the Master Data tab might not be activated within Excel. If this is the case, open the Master Data Services Configuration Manager console. You can find it in All Programs | Microsoft SQL Server 2012 | Master Data Services. Click "Web Configuration" in the left pane. In the Website drop-down in the right pane, select MDS. Lastly, click the "Enable integration with Data Quality Services" button in the bottom of the right pane. Click the "Exit" button. "

This step verifies the column mappings for an OLE DB destination adapter. It's in the middle of a set of instructions for configuring a flat file destination adapter, whose mappings are verified in step 20.
Are the mappings applicable to some other adapter, or is this step superfluous?

Note from the Author or Editor:
Please remove step 17, on page 346:

17. On the Mapping tab of the OLE DB Destination Editor, check to make sure that all destination columns are mapped to input columns. Click OK.

Regards

Grega

The note for bulk insert task and permissions states that "The bulk insert task requires the user who is executing the ssis package that contains this task to be a member of the sysadmin fixed server role."

This should state the user needs to be a member of the "bulkadmin" server role. Sysadmin permissions are not necessary.

Note from the Author or Editor:
The text is based on the official documentation. I will have to verify this with the vendor.


ML


[UPDATE, Oct. 24 2013]:
I am still investigating this issue.


ML


[UPDATE, Nov. 26 2013]:

I have asked my contacts at the vendor several times, but unfortunately have not received any definite answers. I will give them a few more days, and then prepare a correction based on my own observations, which may or may not be 100% correct, it will, however, be more accurate than what is said in the vendor documentation.


ML


I will send the amended text via email.


ML


UPDATE from ML on March 18, 2014: The ?Important? reader aid should be replaced with this one:

IMPORTANT THE BULK INSERT TASK AND PERMISSIONS
According to vendor documentation, the Bulk Insert task requires the user who is executing the SSIS package that contains this task to be a member of the sysadmin fixed server role. However, based on practical experience with SSIS 2012, it seems that the minimum set of permissions required to execute the task is far less than what is said in the documentation.

The user executing an SSIS package, in which the Bulk Insert task is used, actually needs the following permissions:
a) the user must be a member of the bulkadmin fixed server role; and
b) must either have ALTER TABLE permissions on each destination table, or be a member of the db_ddladmin database role; and
c) the user must have INSERT permissions on each destination table, or be a member of the db_datawriter database role.

"... to the TK463 database."
should be:
"... to the TK463DW database."

Assertion is support by text inside parenthesis in step 11.

Note from the Author or Editor:
On page 312, step 3 of the exercise 2, please change the sentence from:

Add a new ADO.NET connection to the TK463 database.

to:

Add a new ADO.NET connection to the TK463DW database.

GJ

"... to the TK463 database."
should be:
"... to the TK463DW database."

This is assertion is supported by Figure 7-10.

Note from the Author or Editor:
On page 310 in step 9 please change the sentence from:

In the Configure ADO.NET Connection Manager dialog box, select New and create the connection to the TK463 database.

to:

In the Configure ADO.NET Connection Manager dialog box, select New and create the connection to the TK463DW database.

GJ

In the included testing application question 463P_5.20_02 includes numerous entries of "qds_" when those all should be replaced wtih "dqs_"

For exmple one of the possible answers is "Only members of the qds_administrators role." qds_administrator should be dqs_administrator.

As currently written none of the answers are correct because they call include references to roles that do not exist.

Note from the Author or Editor:
Confirmed; however, the writers cannot do much about this. Please inform the questions author about this error.

The original is:

12. Drag a data flow task from the SSIS Toolbox onto the Sequence Container.

It is better to change it to:

12. Drag a data flow task from the SSIS Toolbox onto the Event Handler design area.

Note from the Author or Editor:
Please change the text of step 12 from exercise Use Event Handlers from

12. Drag a data flow task from the SSIS Toolbox onto the Sequence Container.

to:

12. Drag a data flow task from the SSIS Toolbox onto the Event Handler design area.

Thank you

Grega

The CREATE TABLE stg.CustomerInformation statement in the book creates the Gender and MaritalStatus as data type CHAR(1).

The CREATE TABLE stg.CustomerInformation statement in the file TK463Chapter05.sql creates the Gender and MaritalStatus as data type NCHAR(1).

Please synchronize the table DDL between the book and the SQL file.

Note from the Author or Editor:
Fixed in the sql file.

I know it's nit-picky but the last two words in the answer are "Not Supported" with a space between them. The correct property is "NotSupported" without the space.

Note from the Author or Editor:
On page 332 in the quick check box please correct the sentence from:

You should remove the legacy task from the sequence container or set the TransactionOption property to Not Supported.

to:

You should remove the legacy task from the sequence container or set the TransactionOption property to NotSupported.

Thank you

Grega

FWIW when I ran the package 15 rows, not 5, were redirected to the ErrorRows flatfile.

Note from the Author or Editor:
Please change the text in step 19 on page 321 from:

Execute the package. Notice that five rows were transferred through the error path, and the package ran successfully.

to : Execute the package. Notice that fifteen rows were transferred through the error path, and the package ran successfully.

GJ

Exercise 2 Step 5 reads: "5. In the Variables pane, create a new, project-scoped integer (Int32) variable named lastRecordCount."

The package solution provided from the TK70463_CodeLabSolutions.zip file has the lastRecordCount variable created as a Char.

Note from the Author or Editor:
Thank you for noticing it. I will change the lab files.

BR

Grega

The answer to Lesson 3 Question 1 Choice A is "Correct: All source adapters have an error flow."

This contradicts the answer to Question 3 Choice B, which reads: "incorrect: Raw File source does not have an error flow."

Note from the Author or Editor:
The answer to Lesson 3 Question 1 Choice A is "Correct: All source adapters have an error flow."

It should be "Correct: The OLE DB source adapter has an error flow."

GJ

The "DefaultBufferMaxRow" property is actually "DefaultBufferMaxRows" (Note the added "S"). The following locations need to be corrected:

page 263 - Step 3
and
page 263 - Text next to LIsting 6-2

Note from the Author or Editor:
Both occurrences of the expression "DefaultBufferMaxRow" on page 263 must be corrected to "DefaultBufferMaxRows".


ML

Excel 2012 should be Excel 2010.

Note from the Author or Editor:
The Quick Check answer should be corrected from
"Quick Check Answer
■■ Use Excel 2012 with MDS Add-in for Excel to update batches of data in an MDS
model."
to
"Quick Check Answer
■■ Use Excel 2010 with MDS Add-in for Excel to update batches of data in an MDS
model."

The text refers to a column in the DimSalesReason dimension table named "SalesReasonType", but the column name is actually "SalesReasonReasonType" in the AdventureWorksDW2012 database (and is correctly displayed in the screenshot on p25).

Note from the Author or Editor:
The last sentence of the fourth bullet of Exercise 2 (top of the page) should be modified from
"In the DimSalesReason dimension, it seems that there is a natural hierarchy:
SalesReasonType ➝ SalesReasonName."
to
"In the DimSalesReason dimension, it seems that there is a natural hierarchy:
SalesReasonReasonType ➝ SalesReasonName.".

In addition to re-running the script from chapter 5 code folder you also need to open the TK463Chapter13.sql file and run the first section to create the stg.SalesTerritory table. This step is missing from the instructions.

Note from the Author or Editor:
Exercise 1, Step 1 on page 508 should have an additional statement after the existing text "If you are missing the database objects from Chapter 5, ?Designing and Implementing Data Flow,? execute the needed SQL code from that chapter to have all the stage and dimension tables available in the TK463DW database. "

Should have an additional sentence at the end:

Start SSMS and connect to your SQL Server instance. Open a new query window by clicking the New Query button. Select the TK463DW database in the database drop-down list. Create the stg.SalesTerritory staging table by executing the following SQL statement.
CREATE TABLE stg.SalesTerritory
(
TerritoryID INT NULL,
Name NVARCHAR(50) NULL,
CountryRegionCode NVARCHAR(10) NULL,
[Group] NVARCHAR(50) NULL,
ModifiedDate DATETIME NULL
);

"PowerShell_Example.ps2" should be "PowerShell_Example.ps1"

Also it may be worth mentioning that some some computers have scripts disabled for powershell. I eventually got it to run by launching the PowerShell ISE and then executing the following commands

> get-executionpolicy -CurrentUser
That returned "Undefined"

>set-executionpolicy Unrestricted -scope CurrentUser
That gave me a warning which I clicked OK to accept

once those two commands were executed I was able to run the script within the PowerShell ISE

Note from the Author or Editor:
Regarding the typo:
Please, replace the text "PowerShell_Example.ps2" with "PowerShell_Example.ps1".

Regarding the operating system privileges to execute PowerShell scripts:
An appropriate warning should be added to the chapter. I will send the text through email.


ML

Not necessarily a mistake but you may want to point out to the readers that if they have more than one instance of SQL Server installed on their computer that they will need to specify the full path for the DTExec.exe file. I had to edit the .cmd file to change the path to "c:\program files\Microsoft SQL Server\110\DTS\Binn\DtExec.exe" (with the quotes) Once I did that the package ran fine.

Note from the Author or Editor:
A corrected version of the CMD file has been sent to the publisher.


ML

Step 14 and 15 refers to FillStageTables.dtsx. It should be DimCustomer.dtsx as shown in Figure 10-4

Note from the Author or Editor:
The discrepancy between text and figure is in this particular case of no real importance. An appropriate replacement image file for Figure 10-4 has been sent to the publisher.


ML

Step4 needs an extra first step:

Set "Evaluation operation" to "Expression".
Click the ellipsis (...)

Note from the Author or Editor:
The text for Step 4 needs to be replaced with the following paragraph:

Change the Evaluation operation constraint option from ?Constraint? to ?Expression and Constraint?, leave ?Success? as the evaluation value, then click the ellipsis (?) at the right of the Expression text box to open the Expression Builder dialog box, and build the expression shown in Listing 6-5.

The text has:

"In this example, you cannot use all foreign keys together as a composite primary key ... and the primary key would be duplicated. You could use ... however, in order to show how you can autonumber a column with the IDENTITY property, this exercise has you add your own integer column with this property. This will be your surrogate key.

CREATE TABLE dbo.InternetSales
(
InternetSalesKey INT NOT NULL IDENTITY(1,1),
CustomerDwKey INT NOT NULL,
ProductKey INT NOT NULL,
DateKey INT NOT NULL,
OrderQuantity SMALLINT NOT NULL DEFAULT 0,
SalesAmount MONEY NOT NULL DEFAULT 0,
UnitPrice MONEY NOT NULL DEFAULT 0,
DiscountAmount FLOAT NOT NULL DEFAULT 0,
CONSTRAINT PK_InternetSales PRIMARY KEY (InternetSalesKey)
);"

But FIGURE 2-1 does not have InternetSalesKey as a column of the InternetSale table and shows CustomerDwKey, ProductKey, and DateKey as the primary key.

Note from the Author or Editor:
I already sent the corrected figure to the publisher.

This question "463P-2.4_02C" has an incorrect answer.

The "Merge Join" transformation is partial blocking yet when I add it to the 'Partial blocking' node on the transformation tree as an answer it marks it as incorrect.

The explanation states:

The transformation types are:

* Non-blocking:
* Multicast
* Derived Column
* Row Count
* Export Column

Partial blocking:
* Union All

Blocking:
* Aggregate
* Sort
* Fuzzy Lookup

It should be:

The transformation types are:

* Non-blocking:
* Multicast
* Derived Column
* Row Count
* Export Column

Partial blocking:
* Union All
* Merge Join

Blocking:
* Aggregate
* Sort
* Fuzzy Lookup

Note from the Author or Editor:
Merge join is partly blocking. Please inform the practice test questions author about this issue.
Dejan Sarka

The @updateusage parameter of the system stored procedure sp_spacedused is a varchar datatype not nvarchar so all references to @updateusage = N'TRUE' should be changed to @updateusage='TRUE'

Note from the Author or Editor:
I checked, and this is correct. The sys.sp_spaceused system procedure's second parameter, the @updateusage parameter, is VARCHAR(5) data type. However, it seems to me that this is an inconsistency from MS. Nevertheless, the error is really minor, so I guess it makes no sense to change it all over the chapter.

The first sentence starts out with "The query returns 6343 rows..." Actually it returns 6434. This can be confirmed by running the following:

Select count(*) from dbo.FactInternetSales where CustomerKey <= 12000;

The result of that query is 6434.

Also not that it really matters but my logical read counts are significanty different than yours:

FactInternetSales logical reads - 4534
DimCustomer logical reads - 196
Worktable logical reads - 21314

What could cause such big differences?

Note from the Author or Editor:
The beginning of the first sentence of the last paragraph before the note should be modified from
"The query returns 6,343 rows and performs..."
to
"The query returns 6,434 rows and performs...".

About different numbers for logical IO: you cannot guarantee that you would always get the same execution plan. This depends on your HW, database settings, indexes and more. There is actually a note at the bottom of the page: "Note Number of Logical Reads
You might get slightly different numbers for logical reads; nevertheless, you should get
many more logical reads from the first query than from the second query." The important thing here is that you definitely should get less logical reads from the second query, which uses the Window functions.

gacutil fails with "Failure adding assembly to the cache: Attempt to install an assembly without a strong name."

There should be a reference to instructions for signing the assembly.

Note from the Author or Editor:
Additional steps must be added to Exercise 2 of Lesson 3, Chapter 19, instructing the student how to digitally sign the assembly.

Quick Check Answer reads
"On the SSISDB, project, and package levels"
It should read
"On the SSISDB, Folder, and Package levels"
Predefined reports are not available on the Project level.

Note from the Author or Editor:
Unable to check at the moment, but I suspect the reader is correct.

Will respond with a more definite answer ASAP.

ML


UPDATE: The reader is only partially correct.

The text of the Quick Check Answer should be changed to:
"On the SSISDB catalog, folder, project, and package levels."

ML

There isn't an SSISDB stored proc called "catalog.configure_server".
Perhaps you mean "catalog.configure_catalog"

Note from the Author or Editor:
True. Any and all references to "catalog.configure_server" should be replaced with "catalog.configure_catalog" (that is, once on page 481 and once in the Index, on page 772).


ML

Please retract my earlier report of this error. The error is actually in the Lesson Review section, not the Answers section.


'The Lookup task cache mode property'
should read
'The Lookup transformation SqlCommand property'

Note from the Author or Editor:
Agree. The question 3, answer D should be changed from:
"The Lookup task cache mode property" to
"The Lookup transformation SqlCommand property".

GJ

This note should probably be expanded to include both criteria.
The IsolationLevel property in SSIS objects applies only when you: 1. Set the TransactionOption property to Required
2. The parent container has not already started a transaction

Note from the Author or Editor:
The existing note: "Note Setting Isolation Levels of a Transaction in SSIS
The IsolationLevel property in SSIS objects applies only when you explicitly open a transaction
inside SSIS by setting the TransactionOption property to Required."

Should be modified: "
Note Setting Is olation Levels of a Transaction in SSIS
The IsolationLevel property in SSIS objects applies only when you explicitly open a transaction
inside SSIS by setting the TransactionOption property to Required and the parent container has not already started a transaction."

"The Snapshot value of the IsolationLevel property is incompatible with package transactions."
See http://msdn.microsoft.com/en-us/library/ms137749.aspx
This information affects how Snapshot can be used in SSIS.

Note from the Author or Editor:
The existing text "Snapshot The data read within a transaction will not reflect changes made by other
simultaneous transactions. The transaction uses the data row versions that exist when
the transaction begins. No locks are placed on the data when it is read."

Should be changed to:
"Snapshot The data read within a transaction will not reflect changes made by other
simultaneous transactions. The transaction uses the data row versions that exist when
the transaction begins. No locks are placed on the data when it is read.
The Snapshot value of the IsolationLevel property is incompatible with package transactions. Therefore, you cannot use the IsolationLevel property to set the isolation level of package transactions to Shapshot. Instead, use an SQL query to set package transactions to Snapshot."

The name of the parent package variable must match the name of
the corresponding variable in the parent package.

should be

The name of the parent package variable must match the name of
the corresponding variable in the child package.

Note from the Author or Editor:
The last sentence in paragraph 6 (under Package Configurations) should read:

"The name of the parent package variable must match the name of
the corresponding variable in the child package."


ML

Point 12 tells me to use the suggested data types to import the data from the file. For gender and marital status, SSIS considers unicode strings of length 1. As we created the corresponding table with char(1) for those two attributes at the beginning of this exercise, the execution of this part fails.

Note from the Author or Editor:
Fixed in all scripts and projects to have the gender and marital status of type nchar(5).

Please also change on page 209, Exercise 1, step 1 the new CREATE TABLE for dbo.Customers:

-- Customers dimension with a PK
CREATE TABLE dbo.Customers
(
CustomerDwKey INT NOT NULL,
CustomerKey INT NOT NULL,
FullName NVARCHAR(150) NULL,
EmailAddress NVARCHAR(50) NULL,
BirthDate DATE NULL,
MaritalStatus NCHAR(5) NULL,
Gender NCHAR(5) NULL,
Education NVARCHAR(40) NULL,
Occupation NVARCHAR(100) NULL,
City NVARCHAR(30) NULL,
StateProvince NVARCHAR(50) NULL,
CountryRegion NVARCHAR(50) NULL,
Age AS
CASE
WHEN DATEDIFF(yy, BirthDate, CURRENT_TIMESTAMP) <= 40
THEN 'Younger'
WHEN DATEDIFF(yy, BirthDate, CURRENT_TIMESTAMP) > 50
THEN 'Older'
ELSE 'Middle Age'
END,
CurrentFlag BIT NOT NULL DEFAULT 1,
CONSTRAINT PK_Customers PRIMARY KEY (CustomerDwKey)
);
GO

GJ

This is a minor tweak to a great exercise, in a very helpful introductory chapter.
The calculation for Age
Age AS
CASE
WHEN DATEDIFF(yy, BirthDate, CURRENT_TIMESTAMP) <= 40
THEN 'Younger'
WHEN DATEDIFF(yy, BirthDate, CURRENT_TIMESTAMP) > 50
THEN 'Older'
ELSE 'Middle Age'
END
does not really deal with NULL values for BirthDate. Regardless of whether there are NULL values currently in the data, the BirthDate field allows them, so the code should be prepared to deal with them.
Age AS
CASE
WHEN BirthDate IS NULL THEN NULL
WHEN DATEDIFF(yy,BirthDate,CURRENT_TIMESTAMP) > 50
THEN 'Older'
WHEN DATEDIFF(yy,BirthDate,CURRENT_TIMESTAMP) > 40
THEN 'Middle Age'
ELSE 'Younger' END

Note from the Author or Editor:
I agree with the reader. Just to make clear what needs the change: the code starts on page 50, at the bottom, bullet 2. The complete code for creating the dbo.Customers table, which spans over page 51 as well, is now:

CREATE TABLE dbo.Customers
(
CustomerDwKey INT NOT NULL,
CustomerKey INT NOT NULL,
FullName NVARCHAR(150) NULL,
EmailAddress NVARCHAR(50) NULL,
BirthDate DATE NULL,
MaritalStatus NCHAR(1) NULL,
Gender NCHAR(1) NULL,
Education NVARCHAR(40) NULL,
Occupation NVARCHAR(100) NULL,
City NVARCHAR(30) NULL,
StateProvince NVARCHAR(50) NULL,
CountryRegion NVARCHAR(50) NULL,
Age AS
CASE
WHEN DATEDIFF(yy, BirthDate, CURRENT_TIMESTAMP) <= 40
THEN 'Younger'
WHEN DATEDIFF(yy, BirthDate, CURRENT_TIMESTAMP) > 50
THEN 'Older'
ELSE 'Middle Age'
END,
CurrentFlag BIT NOT NULL DEFAULT 1,
CONSTRAINT PK_Customers PRIMARY KEY (CustomerDwKey)
);


and it should be changed to:

CREATE TABLE dbo.Customers
(
CustomerDwKey INT NOT NULL,
CustomerKey INT NOT NULL,
FullName NVARCHAR(150) NULL,
EmailAddress NVARCHAR(50) NULL,
BirthDate DATE NULL,
MaritalStatus NCHAR(1) NULL,
Gender NCHAR(1) NULL,
Education NVARCHAR(40) NULL,
Occupation NVARCHAR(100) NULL,
City NVARCHAR(30) NULL,
StateProvince NVARCHAR(50) NULL,
CountryRegion NVARCHAR(50) NULL,
Age AS
CASE
WHEN BirthDate IS NULL THEN NULL
WHEN DATEDIFF(yy,BirthDate,CURRENT_TIMESTAMP) > 50
THEN 'Older'
WHEN DATEDIFF(yy,BirthDate,CURRENT_TIMESTAMP) > 40
THEN 'Middle Age'
ELSE 'Younger'
END
CurrentFlag BIT NOT NULL DEFAULT 1,
CONSTRAINT PK_Customers PRIMARY KEY (CustomerDwKey)
);
GO

"However, in the DimDate dimension, if you know the month, you obviously know the calendar quarter, and if you know the calendar quarter, you know the calendar year."
There is a functional dependency between month and quarter.
This is NOT true for quarter and year. For example, knowing it is the 2nd quarter does not identify the year, as each and every year has a 2nd quarter.

Note from the Author or Editor:
The last statement in th e4th paragraph reads:
However, in the DimDate dimension, if you know the month, you obviously know the calendar quarter, and if you know the calendar quarter, you know the calendar year.

It should be:
However, in the DimDate dimension, if you know the month, you obviously know the calendar quarter, and if you know the calendar quarter, you know the calendar semester.

The specified link for the source code zip TK70463_CodeLabSolutions.zip is displayed on front page but the link does not provide access to the file, just access to the sample databases.

Note from the Author or Editor:
O'Reilly is informed about this issue and it should be resolved soon.
Thank you for reporting.
Dejan Sarka

You should be able to download the companion content from http://examples.oreilly.com/9780735666092-files/.

Thank you for your patience,
Dejan Sarka