|
Mycroft has given you some good advice.
In terms of your original question, yes this will be reliable if you serialize the transactions.
Incidentally, this will reset before 1000, not after. I assume that is what you want, so you are generating your business numbers in the range 1001... to 1999...
|
|
|
|
|
I have a stored procedure that pivots data so that I have a row containing an ID column and a number of other columns which are named by a Product Code.
For example the original table would look something like this :-
OrderId
ProductCode
Quantity
with data like so...
1000, ProdA, 100
1000, ProdB, 200
1001, Prodc, 50
1002, ProdB, 200
Thus my stored procedure would produce a result set of :-
OrderId, ProdA, ProdB, ProdC
----------------------------
1000, 100, 200, 0
1001, 0, 0, 50
1002, 0, 200, 0
So far so good.
Now, the problem is that I need to attach this to the result of a view which contains other related details.
I've researched on the internet, but all the examples I find assume I know the structure (field names) that the Stored Procedure will return. All I know for sure is that there will be an Id column of type BigInt (which is common to both the view results and the Stored Procedure results), and a varible number of columns, with names that change, that are of type Decimal.
Any pointers, gratefully received.
Steve Jowett
-------------------------
Real Programmers don't need comments -- the code is obvious.
|
|
|
|
|
Hi,
You could try inserting the stored Proc's results into a temp table and joining your view from there.
|
|
|
|
|
Q1: Is the "other related data" related to the Order ID or to the Products ?
Q2: I know I could get alot of grief for this but, What about a Temp table ?
Create your Pivot Analysis, store it in a temp table, then process this temp table to add any additional associated data.
I could probably offer a better answer if Q1 was answered.
Regards,
David
|
|
|
|
|
David Mujica wrote: Is the "other related data" related to the Order ID or to the Products
The OrderId is the common key.
David Mujica wrote: Create your Pivot Analysis, store it in a temp table, then process this temp table to add any additional associated data
Can I define a temp table without knowing the column names or the number of columns?
My other thought was to create a .NET Extension and have the produce the data resultset required and return a datatable. (never done it before, but I am will to try.)
Steve Jowett
-------------------------
Real Programmers don't need comments -- the code is obvious.
|
|
|
|
|
While you can join the 2 sets of data (you say the productid is common) by building sub selects and inner joins it will look horrible and will be difficult to support.
Go with the temp table or the table var, I often use a table var for pivot stuff because you are usually using dynamic SQL to build the pivot and therefore building the target table code is trivial. Also pivot data tends to be small in volume!
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
More reason not to use stored procedures. Try a table-valued function instead.
|
|
|
|
|
hi to all,
I have an application that requires a lot of calculation, there are a lot of clients that uses that application having separated DB’s. Now I am going to develop another web-application that will get the records of all those databases and on the basis of filter will display the results. I want to ask you a question that
1. Should I change the structure of the application in which all of my clients share a same database and from where I could get all the records I needed and perform calculation on them?
2. Would I keep the all databases separate (as it is condition) and develop a service that will fetch the records from the entire database into one from where I would perform my calculation and display the data.
Do also tell me if there is some other solution to this application architecture.
|
|
|
|
|
What is the rationale for every client to have their own database? Would always be easier to have all of the information in one database but by no means a must - it would depend on the design and needs of the application.
"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." Red Adair.
nils illegitimus carborundum
me, me, me
|
|
|
|
|
My vote is for a centralized DB where you store all of the data. This has the advanatage of 1 backup, transactions, no import logic and overall less headaches.
The only advantage of each client having a local DB is that they can operate while disconnected from the network.
Do your clients have a reliable connection to a central DB ?
Good luck.
|
|
|
|
|
Thnx for your reply, but what if the Database grows big coz of the data and the client. Also the backup will also become an issue due to larger size. All the users databases are also managed by us.
|
|
|
|
|
Unless there are very good reasons not to do so, I would go for a central DB. It may take some doing to get there, but it should be worth it. Remember, your app will grow, and maybe the number of clients will increase too, so delaying the decision probably will only make matters worse.
Good reasons not to do so would be special circumstances, such as a relatively slow network. If so, I'd suggest to remedy them.
|
|
|
|
|
adilsardar wrote: Would I keep the all databases separate (as it is condition)
If that's the condition, then yes, keep them separate. Who came up with this restriction, and why?
I can imagine some clients not feeling happy about mingling their valuable data with the data of their competitors. It would also mean that things go over "a line" outside the building, and some customers don't trust SSL.
Do you need live access to their database, or could you suffice with a backup that they send you? Have you explained to them in detail what information you're going to pull from 'their' database?
Do you need much of their data? Often? If so, it might be a better idea to replicate or synchronize the data. If not, could the query run during the night? Could it run at a lower priority when run during the day?
Happy thinking
I are Troll 
|
|
|
|
|
The database contains approx 260 tables and what we need is the data from appox 150 tables.
Is ssis packages are good solution if we use separate db's and then sych them.?
|
|
|
|
|
adilsardar wrote: The database contains approx 260 tables and what we need is the data from appox 150 tables.
That's over 50%. How much data would that be, in megabytes? And how often will you run this query?
adilsardar wrote: Is ssis packages are good solution if we use separate db's and then sych them.?
I don't have personal experience with SSIS, but replication sounds good.
I are Troll
|
|
|
|
|
Probably one database.
As for making calculations, it depends on the calculations. Are they such that you can keep a summary table from which to calculate them?
|
|
|
|
|
thnx for the reply, this involves a lot of calculation from multiple tables.
|
|
|
|
|
Furthermore, you will need a database engine which is capable of those amounts of calculations - go for Microsoft SQL Server or Oracle, postgres or MySQL (even worse) cannot be used for such a purpose.
|
|
|
|
|
I'm trying to edit the CommandText for the Insert Command in a TableAdapter in the Designer in vb.net. Whenever I make changes to the CommandText, Visual Studio (2008) automagically changes the Parameters to a Collection consisting of one Parameter that existed before editing the Command, immersed in a sea of complete garbage Parameters, and then I either have to undo (unless this was the first thing I tried editing, in which case I have to close the Designer to discard changes and reopen it) or manually change ALL the Parameters back to something that isn't complete garbage.
Has anybody seen anything like this, or know how to stop Visual Studio from murdering my carefully tended Parameters? Any way to change the CommandText without finding yet another reason to curse the name of M$ would be greatly appreciated.
|
|
|
|
|
I expect that's what it's supposed to do.
Never use a DataAdapter / TableAdapter -- they are too much trouble for real work.
|
|
|
|
|
That is why i dont use wizards to do my Data Layer
Vuyiswa Maseko,
Spoted in Daniweb-- Sorry to rant. I hate websites. They are just wierd. They don't behave like normal code.
C#/VB.NET/ASP.NET/SQL7/2000/2005/2008
http://www.vuyiswamaseko.com
vuyiswa@its.co.za
http://www.itsabacus.co.za/itsabacus/
|
|
|
|
|
Nor do I. 
|
|
|
|
|
Parameters are build based on CommandText. So if you are going to change command text, you would need to change the parameters as well.
Abhishek Sur
Don't forget to click "Good Answer" if you like this Solution.
Visit My Website-->www.abhisheksur.com
|
|
|
|
|
Not always, I can have parameters set up according to the columns of a particular table and then wish to set the command as appropriate for a task and not need to alter the Parameters collection. But to do that I need to roll my own -- which I do.
|
|
|
|
|
Yes, I understand this, and I am quite fine with this. The problem is, VS changes the Parameters for me, instead of giving me the option to specify the Parameters myself.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
