|
It's not so simple. If I have something like this:
ptr = new Myclass();
I want to get into the constructor of Myclass and see what happens, I don't want to debug the code of "new". If I use "step over" I don't see what happens inside the constructor.
Anyway, the Borland 4.52 worked differently. By the way, I tried to run it on windows 2000 and it tells me it cannot find the 32 bit compiler (I set the path). Any idea what's wrong?
Alex
|
|
|
|
|
cnd12001 wrote:
I don't want to debug the code of "new". If I use "step over" I don't see what happens inside the constructor.
Put a breakpoint inside the constructor. Duh.
Christian
I have drunk the cool-aid and found it wan and bitter. - Chris Maunder
|
|
|
|
|
I was about to write "put the breakpoint in the constructor" but you beat me too it. 
Programmers often forget that "contruction" involves more than just calling MyObject::MyObject(). Stepping into new MyObject() may take you to any number of places before you get to the object constructor. If one isn't careful, clean, and clear on this you can introduce bugs through side effects.
Instead of second guessing the runtime, just put break points on the parts you want to inspect.
|
|
|
|
|
|
Hi People!
I’m trying to start an Anti-spam community project. The general purpose of it is to protect and prevent spam from getting in side of users mailboxes; also I want to be able to trace an original spammer, so it can be punished in one way or in another. If things will work out I also would want to change the identification methods for POP/SMTP science I found that most of the email servers have no identification for sending mail (so you can send spam from any user account on the server, and stay undetected)
So if any of you would like to help me please email me or send a post to this treat, thanks.
Also send any ideas if you have!
Alex,
tech@lrcommunications.net
|
|
|
|
|
tempgp wrote:
I also would want to change the identification methods for POP/SMTP science I found that most of the email servers have no identification for sending mail (so you can send spam from any user account on the server, and stay undetected)
This must be solved, if the server cannot identify itself during the mime transfer then refuse the mime.
Yeah its a good idea....
|
|
|
|
|
Is it a server id?
I don’t think so; it is more user identification process for access… If the server is on MS Server OS w/ Exchange than you can just use firewall like ISA to block any one who is not on AD…
|
|
|
|
|
tempgp wrote:
The general purpose of it is to protect and prevent spam from getting in side of users mailboxes;
As long as there are users of computers, this is an impossible task. Spammers use social engineering to help spread their germs.
tempgp wrote:
also I want to be able to trace an original spammer, so it can be punished in one way or in another.
If it was as simple as creating a project, the folks at FTC would have already "been there, done that."
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
Yes, and they can, one thing thay have to think diferently....
If you update the protocols so you can trace the location, than you can go after Spammers, that will reduce spamming ....
|
|
|
|
|
tempgp wrote:
If you update the protocols...
While this is being worked on, there is no guarantee that it will be implemented. You cannot foce computers to be upgraded. It's up to the owner/administrator as to what gets updated/patched and what does not.
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
And If microsoft will release this update?
+ Other big companies?
|
|
|
|
|
I spent a week at Microsoft HQ a few years back, talking with some folks in the development and research departments. Out of all that was talked about, what I remember most was what was said about Microsoft's product output. It was said that 90% of the projects presented to Microsoft are terminated very early on (i.e., they only bother creating 10% of all the ideas they receive). Some of them are scraped at the concept stage while others might make it to the early development stage. When asked why, they responded that it costs very little money to determine if a product is worthwhile or not, whereas once a product goes into development, the costs start going up quickly. They would rather spend a few million up front shooting holes in a concept rather than 10 times that later on supporting a product that was not well received. I said all that to say, you'd have to present Microsoft with a very solid case if you wanted them to release it. They already have an initiative in place that pays a reward if certain virus-writers are caught, so the desire to thwart spammers can't be too far off.
"The pointy end goes in the other man." - Antonio Banderas (Zorro, 1998)
|
|
|
|
|
Yes, too pruve the MS that the project is woth it you just have to have some expiriense in top level presentations, or wight phone numbers 
My intend to make people afraid of sanding SPAM because FBI can go after them....
Also as I alredy mentioned I can get large companies to cooperate, so I just need people who just want to do it, like for idea and not for money
(Well, yes i can say that because I have a lot of free time and...)
|
|
|
|
|
Hi,
I'm new at C++, and am taking over a Visual C++/MFC app. My first assignment involves parsing a data file and creating a graph. Once created i need to interact with the graphdata - for calibration/zoom/etc.
I was wondering where i could find resources on graphing in VC++.
Any suggestions would be greatly appreciated.
Cheers,
Mike
|
|
|
|
|
|
If you'd like to use an inexpensive, royalty-free commercial product, I recommend Lassalle's AddFlow OCX[^]. We've been using it for a while without any problems.
/ravi
My new year's resolution: 2048 x 1536
Home | Articles | Freeware | Music
ravib@ravib.com
|
|
|
|
|
I am using MFC with Visual Studio 2003. Is there a way to retrieve all of the written data off of a CD, and save it as a file? I don't know weather or not copy protection is going to be an issue, and I am assuming that this is not going to be an easy thing to do, so any help at all is appreciated
-Dev578
|
|
|
|
|
I am trying to run a Dialog based application on different machines(desktop and a laptop) , both having different display settings . I develop in the Desktop and when i run this app on the laptop I am unable to display all components . How do I make my Dialog adjust to this ? I need to do this both for the main Dialog and all the subcomponents too . Please help
|
|
|
|
|
If you are comfortable with C++ coding, you could look for the example (CResizingDialog, CResizeableDialog ?) in Dialog controls in CodeProject. There is an example that handles resizing both the "container dialog" and the proper handeling of the controls within the dialog.
Best of luck!
Johnny
|
|
|
|
|
Im Trying to transfer a 2 color bitmap image from a camera through the parallel port and display this image pixel by pixel to the screen.
is there a way other than bitblt to map the pixel directly to the window everytime i recieve it. this way it will be transfered directly until the whole frame is complete and i see the image developing slowly.
thank you very much for your help
|
|
|
|
|
Have you tried SetPixel()?
onwards and upwards...
|
|
|
|
|
SetPixel is as slow as anything - what you need is for the image you put up to be a DIBSEction ( there's a nice DIBSection wrapper here on CP ), so that you can set the pixel values directly.
Christian
I have drunk the cool-aid and found it wan and bitter. - Chris Maunder
|
|
|
|
|
I have a pool of threads and I need to know if all the threads are suspended so that the pool itself does not cause a deadlock condition. I need to know if there is an easy and fast way to tell if a thread is suspended.
John
|
|
|
|
|
You could have each thread keep track of it's state by having the threads write a blocked marker to a global area just prior to calling a blocking function, and then write a running marker after the blocking call completes.
And if you are using real OVERLAPPED structures, an idea from Platform SDK:
HasOverlappedIoCompleted
The HasOverlappedIoCompleted macro provides a high performance test operation that can be used to poll for the completion of an outstanding I/O operation.
Bottom line, there is not a GetThreadState() WINAPI that I know of or can find.
onwards and upwards...
|
|
|
|
|
basementman wrote:
You could have each thread keep track of it's state by having the threads write a blocked marker to a global area just prior to calling a blocking function, and then write a running marker after the blocking call completes.
In my code this would be to much work because any one of 100+ functions may block... Although I am not throwing this idea away yet.
basementman wrote:
And if you are using real OVERLAPPED structures
No, only in a few of the operations.
basementman wrote:
Bottom line, there is not a GetThreadState() WINAPI that I know of or can find.
I could call SuspendThread then ResumeThread to get the info I need but that seems wasteful. A second idea is to use GetThreadTimes to determine if any work has been done since the last time I checked this again seems wasteful...
Thanks for your help,
John
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
