|
hello
Heres the generic situation ...
lets say i have a two machines connected via a socket and messages have been passing though all groovy like. Using send(theSocket etc etc
now, someone pulls out a cable
a Send from machine 1 will place the message in the queue for sending and send(theSocket... will report that it has done it's job.
How can i tell the socket/winsock to fail if the message didnt go through to the target right away?
i guess i'm asking how can i tell if a TCP message actually made its way across the network to the target PC.
Hmmmm
cheers
Bryce
---
Publitor, making Pubmed easy.
http://www.sohocode.com/publitor
|
|
|
|
|
first of all let me suggest a good winsock book. Any socket book will do.
Okay...basically you have machine 1 that will listen for a connection to it. This is a server. Machine 2 is the client. The client creates a socket connection to the server which is listening for a client. When the cable is pulled between the two machines and the client tries to send to the server the socket returns back an error. What ever the last thing was that you were trying to send is what was not sent to the server.
Hope this helps.....but it might be better if you got a good book and read up on sockets... I afradi that I can't recommend one because I cna't name any off the top of my head, but there are two that I have in my office that are really good and if you want to email me I can emails back the titles and the authors.
Tom Wright
tawright915@yahoo.com
|
|
|
|
|
well no the send( ) doesnt return an error when the client's cable is pulled out
on a network
the tcp message is queued and it tcp tries sending it again until it times out i suppose
I'll hafta think on this
bryce
---
Publitor, making Pubmed easy.
http://www.sohocode.com/publitor
|
|
|
|
|
Not true. If the cable is disconnected, then the socket is dropped. If it is dropped on a Send then you will get back an error. You will probabaly get an error 10054
Tom
Tom Wright
tawright915@yahoo.com
|
|
|
|
|
sorry tom old chap
network with N pc's
One running as a server and one as a client,
Do the socket connecty thing, fire a tcp message or two from the server
then you pull out the receiver's cable , the server's send(...) function will not necessarily return an error, it will return that it performed its job correctly (i know coz i just did it here at work  )
from msdn
"The successful completion of a send does not indicate that the data was successfully delivered."
cheerypip 
Bryce
---
Publitor, making Pubmed easy.
http://www.sohocode.com/publitor
|
|
|
|
|
I would say, that in that case you would have to add your own ack/nak layer. You would have to have the receiver acknowledge every message you sent by sending back an "ack" message. 
|
|
|
|
|
Hi,
How to get a coloured button?
Thanks,
Deepak Samuel.
|
|
|
|
|
Hi,
Your question has been posted several times! You should search the comments and you'll get your answer!
A student knows little about a lot.
A professor knows a lot about little.
I know everything about nothing.
|
|
|
|
|
checkout for CxShadeButton class its a 'Class' of its own
|
|
|
|
|
Is there a way to convert the value of a string to an 'enum' value? For example:
string pos = "first";
enum ePOS {first, second, third, etc} POS;
POS = pos; OK, I know "first" as a string, is different than "first" as an 'enum' value (the 'enum' value being the equivalent of integer 0). But if the user enters "first" I'm trying to equate that to the 'enum' value "first" (without having to do a lot of "if - else" testing).
William
Fortes in fide et opere!
|
|
|
|
|
Yes
Could you use a CStringArray instead / or in addition (like build from)? If so find returns the pos of "First" and then enum from there.
Best Wishes and Happy Holiday's,
ez_way
|
|
|
|
|
Not without using some secondary data structure, like a map, to facilitate the conversion.
map<string, ePOS> theMap;
theMap["first"] = first;
theMap["second"] = second;
string s = "first';
ePOS p = theMap[s]; // error checking needed here, obviously
--Mike--
Ericahist | CP SearchBar v2.0.2 | Homepage | RightClick-Encrypt | 1ClickPicGrabber
Four fonts walk into a bar. The bartender says "Hey - get out! We don't want your type in here."
|
|
|
|
|
Thanks for replying.
I decided to use an array of the values, and check what the user entered against the values in the array, then use the relative address of the value in the array (if there was a match) and equate that with the relative value of the 'enum' variable. (Almost the same thing you were suggesting.)
William
Fortes in fide et opere!
|
|
|
|
|
Yes, it is quite same. Only when you have a lot of strings, the Michael Dunn's idea is better, because the map algorithm of matching string-type keys is quicker.
Robert-Antonio
|
|
|
|
|
I have spent the last 24 hours tracking down a heap corruption problem. I think I now know vaguely what the problem is, but it has made it clear I have no idea what I'm doing when it comes to heap usage with DLLs under windows. I am hoping some seasoned windows programmer out there can first look at my sad story below, and then either educate me as to the do's and don'ts of heap usage with DLLs under windows or point me to an article or book that will educate me. I'm not looking to be an expert, but I don't want to hit any more snafus like this one.
MY SAD STORY:
I'm working on my first windows DLL. I have an exported function that looks much like this one:
void __declspec(dllexport) getCommandString(std::string& commandString);
In case it's not obvious, usage of this function is for the application to pass in a string by reference, then some code in the DLL sets it for you.
Perhaps you already know my problem. If you do, YOU are the person I hope to hear from. But I'll finish my sad story for those of you who don't.
So I had some application code that called this function, much like this:
void doSomething()
{
std::string commandString;
getCommandString(commandString);
return;
}
doSomething() fails at the return statement with an assertion failure when memory for commandString was being freed. Digging into the stack trace I found the assertion failure was occuring somewhere in the implementation of this system call:
HeapValidate( _crtheap, 0, pHdr(pUserData) );
I read the docs for HeapValidate and discovered you could set the thrid arg to NULL to force the entire heap to be validated. So I immediately had the idea of sprinkling my own calls to HeapValidate throughout my code to track down the time when I was corrupting the heap. But then I couldn't figure out where to get the symbol _crtheap. I could see it in the debugger, but couldn't find it declared in any headers that came with VC6. So I says to myself, "I'll just hard code the address since surely _crtheap won't change values from run to run." But this caused me problems too. Eventually I figured out that that value of the global symbol _crtheap was changing within a single run! After some additional contemplation, I says to myself, "Perhaps there is more than one heap for my process!" Through additional experimentation, I now believe that my DLL is allocating and freeing memory from one heap, and my application is using another. The above functions cause memory to be allocated in the context of the DLL and freed in the context of the application (and vice-versa).
Voila! Crash!
MY QUESTIONS:
I would like someone to confirm my theorys and perhaps provide me any additional information as to the usage of heaps in DLLs. Does every DLL have its own heap? I currently believe that I must eliminate any interfaces where memory is allocated in one library context and freed in another. What if I changed the signature of my fucntion to:
std::string __declspec(dllexport) getCommandString();
Would this eliminate my problem?
Thanks much,
Matt
|
|
|
|
|
You've got it essentially right. The CRT new/delete operators alloc/free from a heap set up by the CRT. If the DLLs are linked statically to the CRT code, each DLL does its own CRT initialization, thus each gets a different heap.
To alloc memory that's availble to the whole process, use CoTaskMemAlloc() and free it with CoTaskMemFree(). This prevents you from passing a string as a param, however.
--Mike--
Ericahist | CP SearchBar v2.0.2 | Homepage | RightClick-Encrypt | 1ClickPicGrabber
"That probably would've sounded more commanding if I wasn't wearing my yummy sushi pajamas."
-- Buffy
|
|
|
|
|
If I understand correctly, you can only pass string objects to exported functions if the dll was explicitly loaded via LoadLibrary?
|
|
|
|
|
Hi
I have a small app that is included with my main app's. All it does is collet the users feedback and emails the same to me.
When the mod goes to email should I ?
Use my send mail account to carry the mail?
or
Use the cleints email account?
Since send does not require a password either will be easy.
I need some input on this as a friend said if I used my mail server in the U.S. from the other side of the world it might not work.
Best Wishes and Happy Holiday's,
ez_way
|
|
|
|
|
i would prefer using the client's email account. why should u use ur email acc for other users?
Don't try it, just do it! 
|
|
|
|
|
Thank you for your input.
Best Wishes and Happy Holiday's,
ez_way
|
|
|
|
|
I too would go 'client side' - as long as it was stipulated somewhere that 'use of this program/xyz requires a working email setup'
I remember writing a commercial email interface for an acquantance, and one of his customers came back and said 'it doesnt work' .. when I received the error log, the customer didnt have the right smtp account at his isp .. groan
'G'
|
|
|
|
|
Yes, I understand. What if I used both, orr let me correct that. If one failed use the other.
If I use my own I know the ports, the authinication,an all the bells and whistles. If it timed out I would know that there is a long hop between the user and my mail server, and grab the default email account from the reg and send it that way.
Best Wishes and Happy Holiday's,
ez_way
|
|
|
|
|
the problem with this, as I see it, is, what if you user has to go through a firewall to get to you ??
I can imagine the smirk (and answer) from our comms team If I said 'Ive got this app that instead of using our mail client set up, wants you to open a port on the firewall for it to speak directly home' ...
See, it depends if you're talking small or large .. in our case, every desktop is going to have a valid mail client - (we use Novell Groupwise), so, we would invoke a 'mailto' - ditto any clients I have that use Outlook, anything Mapi compliant etc ... only in rare circumstances do we allow 'things' to speak direct smtp to our mail server, these are usually automated programs (and they have to traverse an inner firewall to get to the smtp server)
'G'
|
|
|
|
|
That is why I love you guy's! Great Answer! Thanks
Best Wishes and Happy Holiday's,
ez_way
|
|
|
|
|
You don't need an email account to send email.
Take a look PJ Naughter's SMTP software http://www.naughter.com/smtp.html[^]. It will allow you to send the email directly to your email servers without having to go through any email forwarding servers. However, it does mean that your email server will have to be available 24/7 since there is no in between server to retry if the email fails to be delivered.
Tim Smith
I'm going to patent thought. I have yet to see any prior art.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
