Interface for Visual C++ in VS 2005?

I've been tasked to understand and develop an easy interface to query (insert, select, update, etc) a sql server compact database on an x86 win ce 5 machine. I'm using VS 2005 and have all the necessary SDK's installed. The problem is I can't find any good tutorials or documentation on sql server compact coding in C++ (only in C# and VB). Do y'all have any suggestions on where I can get started to learn the basic c++ sql interface coding?

Thanks!
Jeff

Edit: I'd like to note that I can run the sample northwind project on a win ce 5 x86 emulator, so it's not the setup that I need help with, but the actual coding.

Here are a couple of good places where you can start:

http://www.codeproject.com/ce/#Database

http://www.pocketpcdn.com/articles/articles.php?&atb.set(c_id)=74&atb.perform(list_folder)=&

|||Thanks for the articles. I like what I see with the ATL OLE DB Consumer Templates articles you wrote, but are they still valid for Visual Studio 2005 or just with eVC++ 3 and 4? I tried converting the sample projects, but I got a ton of compile errors as if the atl library was now very different than the one used in the sample project.

Thanks for any light you can shed!

Jeff
|||

Hi ,

I'am also looking for the same thing.If you get the information, please post it.My requirement is Windows Ce 5 as the mobile device and sql ce 3 and above as the backend.I have tried with the articles posted in the codeproject.com resulting with no luck. I have also tried with ADO and ATL OLE DB but resulting in tons of errors.Please help me in this regard.

|||

I am still using the same headers in VS2005. It's a bit of a kludge, but the Windows Mobile 5 SDK do not ship with the newer versions of the consumer templates (you can see these headers in the Win32 SDK). So far these have worked without issues.

|||Hi,

Thank you very much for quickly reacting to my problem.I have tried to convert a pocket pc 2003 Database application which is shipped with Sql ce 2.0 named Northwindoledb.

I have taken a win32 smart device application and copied all the files from Northwindoledb application and made the following changes .

I have removed the following files from "stdafx.h" header file

"oledb.h"
"oledberr.h"

I have included the following header file to my application.

"ssceoledb.h"

After this I have added the code right after

// Microsoft SQL Server for Windows CE 2.0 Provider (Microsoft.SQLSERVER.OLEDB.CE.2.0)
//
extern const OLEDBDECLSPEC GUID CLSID_SQLSERVERCE_2_0 = {0x76A85B2E,0x9DE0,0x4ded,{0x8E,0x69,0x4D,0xEF,0xDB,0x9C,0x09,0x17}};

in the ssceoledb.h header file.

// Microsoft SQL Server Lite for Windows 3.0 Provider (Microsoft.SQLSERVER.OLEDB.CE.3.0)
//
// {32CE2952-2585-49a6-AEFF-1732076C2945}
//
extern const OLEDBDECLSPEC GUID CLSID_SQLSERVERCE_3_0 = {0x32ce2952, 0x2585, 0x49a6, {0xae, 0xff, 0x17, 0x32, 0x7, 0x6c, 0x29, 0x45}};

and after this add the following line to the "ssceoledb.h" header file

typedef DWORD DBROWSTATUS;

And installed the sqlce 3.0 on windows CE 5.0 mobile.

The application has started running.

But question to you is how can I convert the Same project to MFC Dialog Based application.

Please help me in this regard.

Bye..

J.V.Sivaram

interesting update/insert trigger problem (null issue)

Ok so here is the issue. I am thinking I somehow have to clear the old data
after executing the trigger.
So I add a user Joe Brown with his info on a users table, a trigger fires
and dumps the duplicate data into a users-dup table (for other justifiable
purposes). Update does the same basic thing.
Works fine. Here is the problem. I then add another name. Jack Black and
his info, but he has some null values... like i don't know his address. So
now the trigger fires and all of the duplicate data is carried across to the
dup table... except where there was no data (NULL) in a field... it is
adding the last "real" data set in replace of the null. SO Jack Black has
Joe Brown's address in his field... since it was the last "not null" value
entered in that column.
Looking for ideas. Figured it is something simple, I am just missing. Like
some sort of purge call.
Below is the code I am using:
for insert:
CREATE TRIGGER insertUserMrktg ON [dbo].[USERS]
FOR INSERT
AS
insert into user_marketing (greeting, fName, lName, title, compName,
address, city, provState, fk_country,
zip, email, phone, phoneext, fax, fk_language, fk_segment, fk_job,
emailType, addedBy, fk_userID)
select greeting, fName, lName, title, compName, address, city, provState,
fk_country,
zip, email, phone, phoneext, fax, fk_language, fk_segment, fk_job,
emailType, addedBy, pk_userID
FROM Inserted
for update:
CREATE TRIGGER updateUserMrktg ON [dbo].[USERS]
FOR UPDATE
AS
update a
set a.greeting=b.greeting,
a.fName=b.fName,
a.lName=b.lName,
a.title=b.title,
a.compName=b.compName,
a.address=b.address,
a.city=b.city,
a.provState=b.provState,
a.fk_country=b.fk_country,
a.zip=b.zip,
a.email=b.email,
a.phone=b.phone,
a.phoneext=b.phoneext,
a.fax=b.fax,
a.fk_language=b.fk_language,
a.fk_segment=b.fk_segment,
a.fk_job=b.fk_job,
a.emailType=b.emailType,
a.addedBy=b.addedBy
FROM user_marketing a, users b
where a.fk_userID=
(SELECT pk_userID
FROM Inserted)
Thanks!Two questions.
1 - Why do you want to do this?
2 - How can we identify the last "real" data inserted in the table?, How do
you know it is "real" and not a fake like you are trying to do?
AMB
"cheezebeetle" wrote:

> Ok so here is the issue. I am thinking I somehow have to clear the old da
ta
> after executing the trigger.
> So I add a user Joe Brown with his info on a users table, a trigger fires
> and dumps the duplicate data into a users-dup table (for other justifiable
> purposes). Update does the same basic thing.
> Works fine. Here is the problem. I then add another name. Jack Black an
d
> his info, but he has some null values... like i don't know his address. S
o
> now the trigger fires and all of the duplicate data is carried across to t
he
> dup table... except where there was no data (NULL) in a field... it is
> adding the last "real" data set in replace of the null. SO Jack Black has
> Joe Brown's address in his field... since it was the last "not null" value
> entered in that column.
> Looking for ideas. Figured it is something simple, I am just missing. Li
ke
> some sort of purge call.
> Below is the code I am using:
> for insert:
> CREATE TRIGGER insertUserMrktg ON [dbo].[USERS]
> FOR INSERT
> AS
> insert into user_marketing (greeting, fName, lName, title, compName,
> address, city, provState, fk_country,
> zip, email, phone, phoneext, fax, fk_language, fk_segment, fk_job,
> emailType, addedBy, fk_userID)
> select greeting, fName, lName, title, compName, address, city, provState,
> fk_country,
> zip, email, phone, phoneext, fax, fk_language, fk_segment, fk_job,
> emailType, addedBy, pk_userID
> FROM Inserted
> for update:
> CREATE TRIGGER updateUserMrktg ON [dbo].[USERS]
> FOR UPDATE
> AS
> update a
> set a.greeting=b.greeting,
> a.fName=b.fName,
> a.lName=b.lName,
> a.title=b.title,
> a.compName=b.compName,
> a.address=b.address,
> a.city=b.city,
> a.provState=b.provState,
> a.fk_country=b.fk_country,
> a.zip=b.zip,
> a.email=b.email,
> a.phone=b.phone,
> a.phoneext=b.phoneext,
> a.fax=b.fax,
> a.fk_language=b.fk_language,
> a.fk_segment=b.fk_segment,
> a.fk_job=b.fk_job,
> a.emailType=b.emailType,
> a.addedBy=b.addedBy
> FROM user_marketing a, users b
> where a.fk_userID=
> (SELECT pk_userID
> FROM Inserted)
>
> Thanks!
>|||Hi
Triggers are executed per statement, which can update multiple rows. Using
where a.fk_userID= (SELECT pk_userID FROM Inserted) will return just one
arbitrary value. You are also not relating user_marketing to users
To keep user_marketing in step try:
update a
set a.greeting=b.greeting,
a.fName=b.fName,
a.lName=b.lName,
a.title=b.title,
a.compName=b.compName,
a.address=b.address,
a.city=b.city,
a.provState=b.provState,
a.fk_country=b.fk_country,
a.zip=b.zip,
a.email=b.email,
a.phone=b.phone,
a.phoneext=b.phoneext,
a.fax=b.fax,
a.fk_language=b.fk_language,
a.fk_segment=b.fk_segment,
a.fk_job=b.fk_job,
a.emailType=b.emailType,
a.addedBy=b.addedBy
FROM dbo.user_marketing a
JOIN Inserted b ON b.pk_userID = a.fk_userID
John
"cheezebeetle" wrote:

> Ok so here is the issue. I am thinking I somehow have to clear the old da
ta
> after executing the trigger.
> So I add a user Joe Brown with his info on a users table, a trigger fires
> and dumps the duplicate data into a users-dup table (for other justifiable
> purposes). Update does the same basic thing.
> Works fine. Here is the problem. I then add another name. Jack Black an
d
> his info, but he has some null values... like i don't know his address. S
o
> now the trigger fires and all of the duplicate data is carried across to t
he
> dup table... except where there was no data (NULL) in a field... it is
> adding the last "real" data set in replace of the null. SO Jack Black has
> Joe Brown's address in his field... since it was the last "not null" value
> entered in that column.
> Looking for ideas. Figured it is something simple, I am just missing. Li
ke
> some sort of purge call.
> Below is the code I am using:
> for insert:
> CREATE TRIGGER insertUserMrktg ON [dbo].[USERS]
> FOR INSERT
> AS
> insert into user_marketing (greeting, fName, lName, title, compName,
> address, city, provState, fk_country,
> zip, email, phone, phoneext, fax, fk_language, fk_segment, fk_job,
> emailType, addedBy, fk_userID)
> select greeting, fName, lName, title, compName, address, city, provState,
> fk_country,
> zip, email, phone, phoneext, fax, fk_language, fk_segment, fk_job,
> emailType, addedBy, pk_userID
> FROM Inserted
> for update:
> CREATE TRIGGER updateUserMrktg ON [dbo].[USERS]
> FOR UPDATE
> AS
> update a
> set a.greeting=b.greeting,
> a.fName=b.fName,
> a.lName=b.lName,
> a.title=b.title,
> a.compName=b.compName,
> a.address=b.address,
> a.city=b.city,
> a.provState=b.provState,
> a.fk_country=b.fk_country,
> a.zip=b.zip,
> a.email=b.email,
> a.phone=b.phone,
> a.phoneext=b.phoneext,
> a.fax=b.fax,
> a.fk_language=b.fk_language,
> a.fk_segment=b.fk_segment,
> a.fk_job=b.fk_job,
> a.emailType=b.emailType,
> a.addedBy=b.addedBy
> FROM user_marketing a, users b
> where a.fk_userID=
> (SELECT pk_userID
> FROM Inserted)
>
> Thanks!
>|||Thanks John...
That worked. Duh...
"John Bell" wrote:
> Hi
> Triggers are executed per statement, which can update multiple rows. Using
> where a.fk_userID= (SELECT pk_userID FROM Inserted) will return just one
> arbitrary value. You are also not relating user_marketing to users
> To keep user_marketing in step try:
> update a
> set a.greeting=b.greeting,
> a.fName=b.fName,
> a.lName=b.lName,
> a.title=b.title,
> a.compName=b.compName,
> a.address=b.address,
> a.city=b.city,
> a.provState=b.provState,
> a.fk_country=b.fk_country,
> a.zip=b.zip,
> a.email=b.email,
> a.phone=b.phone,
> a.phoneext=b.phoneext,
> a.fax=b.fax,
> a.fk_language=b.fk_language,
> a.fk_segment=b.fk_segment,
> a.fk_job=b.fk_job,
> a.emailType=b.emailType,
> a.addedBy=b.addedBy
> FROM dbo.user_marketing a
> JOIN Inserted b ON b.pk_userID = a.fk_userID
> John
>
> "cheezebeetle" wrote:
>

Interesting UPDATE STATEMENT for SQL

This Following statement executes perfectly in SQL2000
but not in SQL7.0 it gives the message
Server: Msg 147, Level 16, State 2, Procedure spTEST1, Line 57
An aggregate may not appear in the WHERE clause unless it is in a subquery contained in a HAVING clause or a select list, and the column being aggregated is an outer reference.

Create Procedure "spTEST1"
As

--Update Processed Flags
UPDATE tblTransaction
SET TransProcessed = 1
FROM dbo.tblTrustGroupTrust
INNER JOIN dbo.tblTrust ON dbo.tblTrustGroupTrust.TgtTrustID = dbo.tblTrust.TrustID
INNER JOIN dbo.tblTransaction ON dbo.tblTrust.TrustID = dbo.tblTransaction.TransTrustID
INNER JOIN dbo.tblShareholder ON dbo.tblTransaction.TransShID = dbo.tblShareholder.ShID
INNER JOIN dbo.tblTransType ON dbo.tblTransType.TransTypeID = dbo.tblTransaction.TransTypeID
WHERE (dbo.tblTrustGroupTrust.TgtTrustGroupID = 3) AND (dbo.tblTransaction.TransProcessed = 0)
AND (dbo.tblShareholder.ShPaymentMethod = 1) AND (dbo.tblShareholder.ShDeceased = 0) AND tblShareholder.ShBankAccount IS NOT NULL
AND EXISTS
(
SELECT dbo.tblTransaction.TransShID
FROM dbo.tblTrustGroupTrust
INNER JOIN dbo.tblTransaction ON dbo.tblTrustGroupTrust.tgtTrustID = dbo.tblTransaction.TransTrustID
WHERE (dbo.tblTransaction.TransProcessed = 0) AND (dbo.tblTrustGroupTrust.TgtTrustGroupID = 3) AND (dbo.tblShareholder.SHID = dbo.tblTransaction.TransShID)
GROUP BY dbo.tblTransaction.TransShID
HAVING SUM(tblTransaction.TransAmt) >= 10
)

When using a straight select on this statement in SQL7.0 it works fine:
SELECT tblTransaction.TransProcessed
FROM .....

Can anyone shed some light on this?The aggregate is on the table being updated - although it is in a subquery this is correllated and so still fails the test.

If you can change the having clause to
HAVING SUM(dbo.tblTransaction.TransAmt) >= 10

so that it is using the copy of the table in the subquery then it should work.|||It might be clearer if you use an alias for the subquery table like
tblTransaction t2.|||Thanks, I very nearly tried to add the missing dbo prefix's in but I presumed that would be too trivial and the error was more complicated than something like that.

Obviously SQL2000 is a little more relaxed on this sort of thing, as it works without error.

Interesting scenario - Multiple files updating one table

Hi,

i have a scenario where I have to read 2 files that update the same table (a temp staging table)...this comes from the source system's limitation on the amount of columns that it can export. What we have done as a workaround is we split the data into 2 files where the 2nd file would contain the first file's primary key so we can know on which record to do an update...

Here is my problem...

The table that needs to be updated contains 9 columns. File one contains 5 of them and file2 contains 4 of them.

File 1 inserts 100 rows and leaves the other 4 columns as nulls and ready for file 2 to do an update into them.
File 2 inserts 10 rows but fails on 90 rows due to incorrect data.
Thus only 10 rows are successfully updated and ready to be processed but 90 are incorrect. I want to still do processing on the existing 10 but cant affort to try and do processing on the broken ones...

The easy solution would be to remove the incorrect rows from the temp table when ever an error occurs on the 2nd file's package by running a sql query on the table using the primary keys that exist in both files but when the error occurs on the Flat File source, I can't get the primary key.

What would be the best suggestion? Should i rather fail the whole package if 1 row bombs out? I cant put any logic in the following package that does the master file update/insert from the temp table because of the nature of the date. I

Regards
Mike

You have more than one way to accomplish that, I think.

You could add an extra column to the staging table that will act as a flag to indicate whether a row was properly updated by the 2 file or not; then further steps should filter the rows based on the value of that column.

Or...

Why you don't create 2 staging tables; one for each file. Then you can use SQL statements to join them, perform some data quality checks and decide which rows are going to be processed and which ones would be rejected.

|||You could also use a Merge Join transform with an inner join prior to loading the table - you would take the good record pipelines from the two flat file sources into the Merge Join and set up an inner join on the key from each file. The pipeline output from the Merge Join would then have only the ten records that had a key match and all 9 columns, which you can then load to the SQL table desitination.|||Thanks, I used the flag method and its doing fine.

Much Appreciated Rafael
Mike

Interesting scenario - Multiple files updating one table

Hi,

i have a scenario where I have to read 2 files that update the same table (a temp staging table)...this comes from the source system's limitation on the amount of columns that it can export. What we have done as a workaround is we split the data into 2 files where the 2nd file would contain the first file's primary key so we can know on which record to do an update...

Here is my problem...

The table that needs to be updated contains 9 columns. File one contains 5 of them and file2 contains 4 of them.

File 1 inserts 100 rows and leaves the other 4 columns as nulls and ready for file 2 to do an update into them.
File 2 inserts 10 rows but fails on 90 rows due to incorrect data.
Thus only 10 rows are successfully updated and ready to be processed but 90 are incorrect. I want to still do processing on the existing 10 but cant affort to try and do processing on the broken ones...

The easy solution would be to remove the incorrect rows from the temp table when ever an error occurs on the 2nd file's package by running a sql query on the table using the primary keys that exist in both files but when the error occurs on the Flat File source, I can't get the primary key.

What would be the best suggestion? Should i rather fail the whole package if 1 row bombs out? I cant put any logic in the following package that does the master file update/insert from the temp table because of the nature of the date. I

Regards
Mike

You have more than one way to accomplish that, I think.

You could add an extra column to the staging table that will act as a flag to indicate whether a row was properly updated by the 2 file or not; then further steps should filter the rows based on the value of that column.

Or...

Why you don't create 2 staging tables; one for each file. Then you can use SQL statements to join them, perform some data quality checks and decide which rows are going to be processed and which ones would be rejected.

|||You could also use a Merge Join transform with an inner join prior to loading the table - you would take the good record pipelines from the two flat file sources into the Merge Join and set up an inner join on the key from each file. The pipeline output from the Merge Join would then have only the ten records that had a key match and all 9 columns, which you can then load to the SQL table desitination.|||Thanks, I used the flag method and its doing fine.

Much Appreciated Rafael
Mike

Interesting ! Update query with alias ?

Hi,
I have a below Oracle query :
UPDATE test1 a SET a.sno = 3
I need the equivalent SQL Server query for the above along with with alias name 'a' set for the table test1. Please advise.
Thanks,
SamI'd use:UPDATE test1 AS a
SET a.sno = 3-PatP|||I too tried the same, but I am getting the error 'Incorrect syntax near AS' in SQL Server 2000|||How about this :

UPDATE a SET a.sno = 3 FROM test1 a

This is working fine...

Integrity with multiple commands

How can I make sure that a couple of commands are either all executed on the database or none of them. For example right now I have an insert, update and delete command. I'm calling each of them with a SqlCommand. So I am afraid that that one of them might be executed, then there's a bad connection and the other two are not. How can I prevent this so that only all commands or nothing is executed on the database?

Write them in a stored procedure~

Stored procedures are a precompiled collection of SQL statements and optional control-of-flow statements stored under a name and processed as a unit.

In short, it's something atomic, a lot of database use some mechanism such as rolling back when a sp is terminated unexpectedly, so either none or all of your command in a sp will be executed~

|||

A stored procedure has absolutely nothing to do with it. You want to place your code within a transaction. You can do this manaully by specifying a BEGIN TRANSACTION (your statements here) and then either ROLLBACK TRANSACTION or COMMIT TRANSACTION, *OR* you can use the sqltransaction object to put multiple sqlcommands within the same transaction.

You can of course put the whole thing in a stored procedure as well for convience, but putting things in a sp won't guarantee they will either all commit or rollback.

This would be an example of what you can place within the commandtext of the sqlcommand property:

dim cmd as new sqlcommand("SET @.RetVal=0; BEGIN TRANSACTION; INSERT INTO Table1(col1) VALUES (@.col1); IF @.@.ERROR=0 BEGIN INSERT INTO Table2(idcol,col2) VALUES (SCOPE_IDENTITY(),@.col2) IF @.@.ERROR=0 BEGIN COMMIT TRANSACTION SET @.RetVal=1 END END IF @.RetVal=0 ROLLBACK TRANSACTION")

cmd.Parameters.Add("@.col1",varchar).value=something

cmd.Parameters.Add("@.col2",varchar).value=something

cmd.Parameters.Add("@.retval",int).Direction=output

cmd.executenonquery()

if cmd.Parameters("@.retval").Value=0 then

' It failed

end if

You can also do it this way:

dim cmd as new sqlcommand("SET @.RetVal=0; SET XACT_ABORT ON; BEGIN TRANSACTION; INSERT INTO Table1(col1) VALUES (@.col1); INSERT INTO Table2(idcol,col2) VALUES (SCOPE_IDENTITY(),@.col2); COMMIT TRANSACTION; SET @.RetVal=1")

That should also work as well.

Integration Services: ?Table refresh (UPDATE/INSERT)“

Hello

I have a question about the new Integration Services of the MS SQL Server 2005.

Situation:
- SQL Server 2005 (standard edition)

- 2 tables with identical structure (same attributes)

- the table ?TestSource“ will be constantly extend (new records & updates).

- the table ?TestDestination“ will just be refreshed by SSIS (Data Warehouse table)

I would like to create a Integration Service, witch refreshes the table ?TestDestination“ with the data from table ?TestSource“.

Existing records (ID already exists) should be updated (UPDATE), not existing records should be created (INSERT).

I would like to use the IS Data Flow Task, because in future i won’t just copy the data. I also will use Toolbox items like ?Data Conversion“, ?Derived Column“ and so on.

Alike I won’t use an easy SQL-Query, because it would be complicated to make changes and to Log the transactions.

Just clear and refill the whole table is not possible because of performance and availability requests (large data).

Question:
How can I implement this workflow as Data Flow in a Integration Service?
Witch components from the Toolbox do I need?

Greetings

Not sure if you can use this within SSIS, but the tablediff.exe that comes with sql server 2005 is pretty slick. I wrote a c# wrapper for it and I have a config file that holds all the table names that I want to sync as well as the source and destination servers. It then generates the sql change file as well as gives output of the number of rows that are out of sync. It works really well for me especially using the C# wrapper that I wrote.

Just a thought....