Re: WSAAsyncSelect and FD_WRITE - C / C++ / MFC Discussion Boards

Re: Learning VS 2005 C++ - recommendation

Rolf Kristensen10-Jan-10 1:40

10-Jan-10 1:40

Very seldom you need a high-performance CButton og CEdit. Use the right tool for the job, and C++/MFC is not the best tool for building GUI applications.

If there are parts of the application that are performance critical, then one can implement these parts in unmanaged C++, and interop with them from .NET.

Btw. MFC has also become quite bloated after including the BCG library, so unless you want to go WTL then it is difficult to deploy a standard GUI application without dependencies.

Re: Learning VS 2005 C++ - recommendation

LunaticFringe10-Jan-10 1:53

LunaticFringe

10-Jan-10 1:53

Snakefoot wrote:
C++/MFC is not the best tool for building GUI applications.

That is a matter of opinion. With practice and a decent knowledge base, it is possible to create compelling UIs quite quickly w/ MFC. The class library is designed in a way that makes it fast and easy to implement custom UI components, unlike Builder's.

Your own experience may have been negative with MFC; mine has been fun and productive.

[edit]
Truth be told, though, I do love WTL. Any new native C++ project I've started in the last couple of years or so has been a WTL app. It's not a very beginner-friendly framework, though - I wouldn't recommend it in this case.
[/edit]

L u n a t i c F r i n g e

Re: Learning VS 2005 C++ - recommendation

Rolf Kristensen10-Jan-10 1:59

Rolf Kristensen

10-Jan-10 1:59

I'm working with MFC every day, so yes you can build nice applications with this framework.

I'm just saying there are better frameworks out these, especially if used to the GUI library of C++ Builder.

Re: Learning VS 2005 C++ - recommendation

Denny10110-Jan-10 20:28

Denny101

10-Jan-10 20:28

Thank you all for your input - my experience with Builder is - fast to make a simple project, easy to make a user interface, but of course limitations when you get deeper and the size of the final exe is bigger because you have to ship the VCL.
Sorry, but I am still hazy about the book to learn VS 2005 Smile | :)

. I think I will stay with the Builder. Big Grin | :-D

Thanks again,
Denny

Re: Learning VS 2005 C++ - recommendation

Rolf Kristensen15-Jan-10 1:20

Rolf Kristensen

15-Jan-10 1:20

Btw. in your question about documentation about using MFC, then this is a good place to start (See "Fundamentals"):

http://www.functionx.com/visualc/index.htm[^]

Re: Learning VS 2005 C++ - recommendation

Denny10115-Jan-10 9:32

Denny101

15-Jan-10 9:32

Thank you for the link, a very good site, indeed,
Regards,
Denny Thumbs Up | :thumbsup:

WSAAsyncSelect and FD_WRITE

zoopp9-Jan-10 0:35

zoopp

9-Jan-10 0:35

First of all...I can't tell you how frustrated i am about this FD_WRITE event. I've been searching for quite some time explanations about how it works yet I'm still not able to understand it properly. D'Oh! | :doh:

Q1: Do i use FD_WRITE to communicate with the client application?

If yes...

Q2: FD_WRITE is triggered when the network buffer becomes full. What is this network buffer and why does it need to become full before sending the data?
Q3: In order to fill it up I should make continuous calls to send() until i receive WSAEWOULDBLOCK. But if i don't have enough data to fill it up, what should i do? And...why continuous calls to send()?

If no..

Q2: What should i do when the FD_WRITE event triggers?
Q3: Should i use the FD_READ event to send a response based on the data received from the client?

Re: WSAAsyncSelect and FD_WRITE

Moak9-Jan-10 5:50

Moak

9-Jan-10 5:50

zoopp wrote:
Q1: Do i use FD_WRITE to communicate with the client application?

Maybe, but you don't have to use it. If you are sending (a lot of) data you might be interested in this event. It is used for handling partial sends, that is if you send more data then the socket can push over the network at a given time.

zoopp wrote:
Q2: FD_WRITE is triggered when the network buffer becomes full.

No, the opposite. If is triggered when the socket can be written to again (after a send with WSAEWOULDBLOCK), it tells you that the socket is ready to send more data. This is a simplified answer, it is also triggered in other cases, for the full documentation see MSDN[^]

zoopp wrote:
Q3: In order to fill it up I should make continuous calls to send() until i receive WSAEWOULDBLOCK.

Yes, until that happens or you have sent all your data. Data will be copied in the socket's send buffer first, the WSAEWOULDBLOCK case only arises if you pump in data faster than the socket can handle (lot of data or slow connection).

zoopp wrote:
Q3: Should i use the FD_READ event to send a response based on the data received from the client?

Maybe, you use this event instead of constantly polling the socket "is there more data for me? are there more bytes?". This event is triggered when the socket can be read, it tells you that the socket has more data.

For further reading have a look at examples, e.g. Winsock FAQ[^] (asynchronous sockets). Btw, I tried to simplify the answers... there is much more to talk about when it comes to efficient networking strategies.

Hope this helps and happy coding.
/M

Webchat in Europe Java | [Coffee]

(only 4K)

Re: WSAAsyncSelect and FD_WRITE

zoopp9-Jan-10 11:18

zoopp

9-Jan-10 11:18

Thank you for your reply. It made things clearer now.

I'll ask another question if that's ok. You said that FD_WRITE is used for handling partial sends but, according to MSDN, it will also trigger after a call to connect() or accept(). The event for these calls should be handled different than partial sends and the way I'd handle both situation would be by disabling FD_WRITE before accept()/connect(), enabling it back after and by moving the code that was supposed execute in FD_WRITE when a new client connects in the FD_ACCEPT event.

Is this a good way of doing it?

Re: WSAAsyncSelect and FD_WRITE

Moak9-Jan-10 12:00

Moak

9-Jan-10 12:00

zoopp wrote:
You said that FD_WRITE is used for handling partial sends but, according to MSDN, it will also trigger after a call to connect() or accept().

MSDN can be misleading, you can use it as a reference but it's an epic fail as a tutorial! The FD_WRITE event will also be fired once a client connects. You can interpret it as a "now it is the first time to send data", but it must not be confused with e.g. FD_CONNECT which signals when a client is connected. Check MSDN[^] where it says "Here is a summary of events and conditions for each asynchronous notification message", it describes the different meanings each event may have. Here is a simplified overview of the corresponding events:

* FD_READ informs when (more) data can be received
* FD_WRITE informs when (more) data can be send
* FD_ACCEPT informs when incoming socket is waiting at listening socket, server side only
* FD_CONNECT informs when socket has been connected or attempt failed, client side only

Okay, to answer your question, I would not disable FD_WRITE before connecting and instead let the application logic decide how it handles the event (you can always ignore it when there is no data to be sent). Let me suggest two more things, have a look at Effective TCP/IP Programming: 44 Tips to Improve Your Network Programs[^] (really can recommend to buy it) and maybe join the webchat in the link below.

Hope this helps Smile | :)

Webchat in Europe Java | [Coffee]

(only 4K)

How to check if a tree is AVL tree?

Cowork8-Jan-10 22:17

Cowork

8-Jan-10 22:17

Hi Is there any algorithm to check if a binary tree
(not BST,just a binary tree with max two childs for each parent)
is AVL tree?

Regards

Re: How to check if a tree is AVL tree?

CPallini9-Jan-10 4:42

CPallini

9-Jan-10 4:42

Cannot you use the AVL tree definition, i.e. compute the balance factor(s)?
Smile | :)

If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.

This is going on my arrogant assumptions. You may have a superb reason why I'm completely wrong.
-- Iain Clarke

[My articles]

Re: How to check if a tree is AVL tree?

Cowork9-Jan-10 13:50

Cowork

9-Jan-10 13:50

Can you write some urls with code?
This information is not enough for me

Regards

Re: How to check if a tree is AVL tree?

CPallini9-Jan-10 22:50

CPallini

9-Jan-10 22:50

Cowork wrote:
Can you write some urls with code?

Nope.

Cowork wrote:
This information is not enough for me

That is a shame: the AVL definition (and some way to traverse the tree) in my opinion is enough.
Smile | :)

Re: How to check if a tree is AVL tree?

Saurabh.Garg9-Jan-10 14:30

Saurabh.Garg

9-Jan-10 14:30

Try standardavl/[^]

-Saurabh

regular expression library for win32 api c or c++

Jayapal Chandran8-Jan-10 7:55

Jayapal Chandran

8-Jan-10 7:55

Hi,
I would like to know whether there is any library for regular expression to use in c win32 api.
I am using Dev C++ IDE to develop windows programs (not applications). It is time to change to vc 6.0 for win32 api programming. anyway... even if it could be for vc 6.0 then that would also holds good. ?

Today's Beautiful Moments are
Tomorrow's Beautiful Memories

Re: regular expression library for win32 api c or c++

LunaticFringe8-Jan-10 8:57

LunaticFringe

8-Jan-10 8:57

ATL contains an implementation called CAtlRegExp. It's a little awkward to use, though. The late Paul DiLascia created an excellent wrapper called CRegex (naturally enough). It's part of PixieLib and can be found on his web site[^].

[Edit]

Update - CAtlRegExp is no longer part of the core ATL distribution, but is included in the ATL server classes package. It can be found here[^]. The regex class is in atlrx.h.
[/Edit]

L u n a t i c F r i n g e

Re: regular expression library for win32 api c or c++

Jayapal Chandran9-Jan-10 5:37

Jayapal Chandran

9-Jan-10 5:37

Thank you for the reply. I am new to this env so it would take some time to use it.

Today's Beautiful Moments are
Tomorrow's Beautiful Memories

Re: regular expression library for win32 api c or c++

LunaticFringe9-Jan-10 11:56

LunaticFringe

9-Jan-10 11:56

You should probably go with the TR1 implementation mentioned below; it'll be less likely to be deprecated.

L u n a t i c F r i n g e

Re: regular expression library for win32 api c or c++

Jayapal Chandran10-Jan-10 8:14

Jayapal Chandran

10-Jan-10 8:14

Yeah. i have to. thank you.

Today's Beautiful Moments are
Tomorrow's Beautiful Memories

Re: regular expression library for win32 api c or c++

Graham Shanks8-Jan-10 10:39

Graham Shanks

8-Jan-10 10:39

Boost[^] contains a regular expression library called Regex.

The advantage of using this is that Regex is also part of TR1[^], so that it will be part of the C++ standard in the future and will be included as part of the standard library in upcoming C++ compilers (it was include in the April 2008 Visual Studio 2008 Feature Pack and I believe that gcc 4.0 supports it as well).

Graham

Librarians rule, Ook!

Re: regular expression library for win32 api c or c++

Nemanja Trifunovic8-Jan-10 12:07

Nemanja Trifunovic

8-Jan-10 12:07

If you are using VC++ 2008 (or even 2005 with the feature pack), you already have TR1 regex installed. Here is a small tutorial.[^]

utf8-cpp

Re: regular expression library for win32 api c or c++

Jayapal Chandran9-Jan-10 5:39

Jayapal Chandran

9-Jan-10 5:39

Thank you for all the above replies... this it totally new so it would take some time for me to ...

Today's Beautiful Moments are
Tomorrow's Beautiful Memories

Question about reading and writing to a file

amichai228-Jan-10 6:25

amichai22

8-Jan-10 6:25

I am trying to write a triple vector to a file and then read the file back into the triple vector. When I try to read the file back after its been saved the first fifty values come out correct but the rest of the values are garbage. I'd be really appreciative if someone could help me out here. Thanks a lot!

File declaration:

fstream memory_file("C:\\Users\\Amichai\\Pictures\\output.txt", ios::in | ios::out);

The write function:

void save_training_data(fstream &memory_file, vector<vector<vector<long double> > > &training_data)
 {
   int sizeI = training_data.size();
   memory_file.write((const char *)&sizeI, sizeof(int));
   for (int i=0; i < sizeI; i++)
   {
       int sizeJ = training_data[i].size();
       memory_file.write((const char *)&sizeJ, sizeof(int));
       for (int j=0; j < sizeJ; j++) 
       {
           int sizeK = training_data[i][j].size();
           memory_file.write((const char *)&sizeK, sizeof(int));
           for (int k = 0; k < sizeK; k++)
           {
               int temp;
               temp = (int)training_data[i][j][k];
               memory_file.write((const char *)&temp, sizeof(int));
           }
       }
   } 
 }

The read function:

void upload_memory(fstream &memory_file, vector<vector<vector<long double> > > &training_data)
{
     memory_file.seekg(ios::beg);
     int temp=0;
     int sizeK, sizeJ, sizeI; 
     memory_file.read((char*)&sizeI, sizeof(int));
     training_data.resize(sizeI);
     for (int i=0; i < sizeI; i++)
     {
           memory_file.read((char*)&sizeJ, sizeof(int));
           training_data[i].resize(sizeJ);           
           for (int j=0; j < sizeJ; j++)
           {
               memory_file.read((char*)&sizeK, sizeof(int));
               training_data[i][j].resize(sizeK);
               for (int k = 0; k < sizeK; k++)
               {
                    memory_file.read((char*)&temp, sizeof(int));
                    training_data[i][j][k]=temp;
               }
           }
     } 
}

Thanks again,
Amichai

Re: Question about reading and writing to a file

amichai228-Jan-10 6:28

amichai22

8-Jan-10 6:28

Just solved the problem.
The file declaration needs the ios::binary specification.

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.