C# Discussion Boards - CodeProject

30-Jul-09 21:28

What difference I see between using statement and Dispose/Finalize pattern is

- “Dispose/finalize” have upper hand if any object which implements this pattern and supporting different interfaces for different operations which performed on underneath unmanaged object, can take care of releasing that object even if user forgets to call dispose on that object
- However with “Using” statement, the scope of the using statement will be restricted to a single method, i.e. we will have to open the unmanaged object which will closed automatically by using’s finally block in the same method (which is, in most of the cases will not be feasible and efficient as we don’t do all the operation at a time on the unmanaged objects)

For Example:
1.Dispose/Finalize:
Suppose, we are using MS word COM to write doc files, we dedicate one class to do all the activities related to doc file and provide different interfaces to write to doc file like WriteTest() , InsertTable(), InsertImage(), …and so on . In this situation user will keep calling these interfaces according to their needs, and might forgets to close the word doc object. In this case, we can very well implement Dispose/Finalize pattern where finalize method will be act as back up plan to release unmanaged word doc object.
This approach will be efficient as long as we are using the same unmanaged object.

2.Using statement
Suppose, we are writing to a text file and we have interface to do this operation which takes text data and file name as parameter. In this case we can use “Using” statement to open the file and write data to it (again supposing that we will not any further operation on this file). What Using statement we do is it add try and finally block to this text file call. Which I guess efficient solution.

Hey guys, anybody know other differences other than what I mentioned here

Or you can correct me, If I am wrong in my above made conclusions!!!!

Smile | :)

Christian Graus30-Jul-09 22:00

Christian Graus

30-Jul-09 22:00

You basically have it in one. using is not a seperate pattern. It's a piece of shorthand that allows you to define an object, and AT THAT MOMENT, know that it will be cleaned up, so you don't need to go to the bottom and have a list of objects, and check that each has Dispose called on it. It's shorthand, to make code more readable and to make it clearer that you're not leaking anything. But, it only works when an object is in the scope of a single method.

Christian Graus

Driven to the arms of OSX by Vista.

Read my blog to find out how I've worked around bugs in Microsoft tools and frameworks.

N a v a n e e t h30-Jul-09 22:07

30-Jul-09 22:07

Christian Graus wrote:
It's a piece of shorthand that allows you to define an object, and AT THAT MOMENT, know that it will be cleaned up, so you don't need to go to the bottom and have a list of objects, and check that each has Dispose called on it.

I liked how C++/CLI implemented this using Stack Semantics. You only need to write a destructor and compiler is smart enough to implement IDisposable on the type and ensure disposal when scope ends. IMO, it is much superior than using block.

Navaneeth

How to use google | Ask smart questions

N a v a n e e t h30-Jul-09 22:04

30-Jul-09 22:04

You can't compare using statement and dispose pattern. using is just a syntactic shortcut which helps to call Dispose method. There is no efficiency difference in calling Dispose directly or wrapping in a using block.

Leaving disposable objects to finalization is inefficient. If the type can be disposed, you should take care about calling dispose on it once you done with the object.

It looks like you are confused with managing life time of objects.

Smile | :)

Navaneeth

How to use google | Ask smart questions

Cracked-Down30-Jul-09 22:59

30-Jul-09 22:59

@Navneeth
No,I am not confuse!! However I wanted to know more about it!!! Smile | :)

I just wanted to see if there are some other differences other than which I have mentioned!
I agree that using statement is not pattern and cant be compared with the Dispose/Finalize. However, if you think from perspective of reclaiming unmanaged objects from memory, then I am sure we can comapre them[as both do the same thing]

Apart from that, what exactly pattern? its only some set of standard which anybody can define( with some advantage), so saying something like this that Using is not pattern and cant be compared with Dispose/Finalize pattern will be a harsh!! isn't it?

I just posted this for discussion, which might be useful for people who do not know about these things.

Anyways thanks for your valuable reply from N a v a n e e t h and Christian!

Not Active30-Jul-09 23:46

Not Active

30-Jul-09 23:46

I'll agree with Navaneeth, you seem a bit confused.

"a design pattern is a general reusable solution to a commonly occurring problem in software design. A design pattern is not a finished design that can be transformed directly into code. It is a description or template for how to solve a problem that can be used in many different situations."
http://en.wikipedia.org/wiki/Design_pattern_%28computer_science%29[^]

If your aurgument is that the using statement is a pattern, then you are also implying that statements like foreach are also patterns.

only two letters away from being an asset

Cracked-Down30-Jul-09 23:58

30-Jul-09 23:58

not at all!!!!
my argument was not that whether the using is a pattern or not

I said, saying - "using is not pattern and cant be compared with other pattern" is wrong!!!!!! [only because its not a pattern]

PIEBALDconsult31-Jul-09 4:58

PIEBALDconsult

31-Jul-09 4:58

So it's a technique, what of it?

N a v a n e e t h31-Jul-09 7:02

31-Jul-09 7:02

SaveTigers wrote:
I just wanted to see if there are some other differences other than which I have mentioned!

Points mentioned by you are correct. It all depends on your object's lifetime. You can use using block when your object have very small lifetime, say in a function.

Consider the example 1 provided in your first post. I assume you have your word wrapper class used in a windows form.

class WordOperations : Form
{

    WordWrapper wrapper = /* ... */

    protected override void Dispose(bool disposing)
    {
        wrapper.Dispose();
        /* ...... */
    }

}

In the above example, we have used form's dispose method to dispose your word wrapper. This will help you to *avoid* the finalization. Now a user of your form can use it like

using(WordOperations op = new WordOperations())
{
    op.ShowDialog();
}

This will ensure the disposal of form and the wrapper.

Finalization has a cost and try to avoid it when possible. So, if you own the code base, there is no need to implement a finalizer on the wrapper class. Just make sure you call Dispose().

SaveTigers wrote:
Apart from that, what exactly pattern? its only some set of standard which anybody can define( with some advantage), so saying something like this that Using is not pattern and cant be compared with Dispose/Finalize pattern will be a harsh!! isn't it?

No.

Implementing IDisposable in a recommended way is called as dispose pattern. using is just a syntactic sugar to ease the use of disposable objects. Read Implementing IDisposable and the Dispose Pattern Properly[^] to get a good understanding about this.

Smile | :)

Navaneeth

How to use google | Ask smart questions

Cracked-Down31-Jul-09 19:46

How to have more than one key and one value in any structure(like hash table) in C#?

31-Jul-09 19:46

Thanks Navaneeth for your reply !!

akashsthakkar30-Jul-09 20:46

akashsthakkar

30-Jul-09 20:46

I want to store two keys and one value associated with those keys in an object and want to send that object to some method.
As hash table can have key value pair,What is the way to do this?? Smile | :)

Re: How to have more than one key and one value in any structure(like hash table) in C#? [modified]

dan!sh 30-Jul-09 20:50

dan!sh

30-Jul-09 20:50

You can use a <code>Dictionary</code> object with key set as an array. Something like this:

<pre>Dictionary<string[], string> oDictionary = new Dictionary<string[], string>();</pre>

modified on Friday, July 31, 2009 3:01 AM

Re: How to have more than one key and one value in any structure(like hash table) in C#?

PIEBALDconsult30-Jul-09 20:59

PIEBALDconsult

30-Jul-09 20:59

A Dictionary of Dictionaries?

Or write a class to contain the two keys.

Lots of ways.

Re: How to have more than one key and one value in any structure(like hash table) in C#?

N a v a n e e t h30-Jul-09 22:10