Comments by softwaremonkey (Top 23 by date)

Thanks Richard. I may have confused you with my rambling explanation. The main app doesn't include the resource DLL's resource.h file but rather the other way round. I did this in an attempt to keep the IDs the same but Visual Studio always wants to add the IDs of new resources to the DLLs own resource.h.

I may have gone about this the wrong way but all of my translated texts are in a string table and so these IDs need to be 'available' to quite a lot of modules. Perhaps there is a better way maybe?

Thanks to both yourself and Richard for your reply. So are you saying that if I add a new resource to the main app, then copy the resource to the DLL I should then simply 'overwrite' the DLL's resource.h with the one from the main app? That sounds easy enough to do!

Hey thanks, this sounds great but I'm not sure how to do it. Do you know of any examples that I could follow?

Thanks for your reply. If you mean the Resource Symbols dialog from the Edit menu then yes. This allows me to edit the value of resource IDs but since they are not in numerical order, there is no easy way of knowing whether two or more resources have be assigned the same value. Even then, it is a time consuming and error prone task with a large project. I hoped there would be a way to do an automated renumber that I had missed somewhere.

Could it be that I am the only one to have had this problem of duplicate ID numbers?

I guess I might have to stump up the $100+ for the Resorg program but its a bit steep for a home user.

Tony

Thanks. I looked briefly at this article but I wasn't sure if they would allow communication between processes operating under different user accounts on the same PC. I tried WM_COPYDATA a while ago and although it worked between two apps running on the same user account it didn't appear to work between users. It could be something I did wrong though!

Thanks. I checked out the code you suggested and it was extremely useful. My backup program is now working with VSS.

Like I said in my original question, it aint my choice dude, the application I need to pass data to only supports this communication method. I have to work with what I have!

The application is controlling machinery via function keys on the keyboard (customer's requirement, not mine!). The user wants (expects) the machine to respond correctly when he presses the keys at all times. I know Windows isn't the right OS for this kind of thing but again, not my choice ;o)

Although it goes against the Windows concept, my app is the ONLY app that is running and should always have all key presses sent to it when it is running. I have a keyboard hook to prevent some keypress combinations which can upset things (ALT-TAB etc) but every now and again something causes my app to loose keyboard focus. Would system modal do the trick maybe?

Thanks, your advice was useful and helped me find a solution

Why? Why what?

Please explain further.

Hmmm, so let the client run even though it is not receiving valid data and the user has no idea that this is the case. Not sure this is a good idea but its a viewpoint.

Thanks for the suggestion. My app gives a clear message explaining that something has not been configured correctly (wrong port or computer name). These problems need investigating and from past experience, if the client sits there after the user has closed the warning dialog, they walk away thinking everything is fine. These are the reasons I would prefer to exit but I accept there may be other (better?) options.

If my client application does not get a reply from the server within 10 seconds then something is wrong and I would like to exit rather than just stay open and doing nothing.

Thanks for the reply. As I said in my original post, I am calling PostQuitMessage() but it doesn't appear to destroy all allocated objects.

Thanks Richard. I appreciate my question was probably unanswerable, I just wondered if anyone had come across any porting issues from VC++6.0 to VS2008 when using CSockets. It seems the client and server connect OK but data transfers stop after a while and its difficult to know if its the client or server. Do you know of any tools that would allow me to monitor data on a specific TCP port?

Actually, no, I can't get it to work on any PCs at the moment, and its not clear to me why. Thats what got me thinking, is there another way I can get this to work maybe by writing some additional software?

As far as I am aware, the runas command will only work if the user account has a password. Some of the PCs have user accounts without a password.