Comments by CodeWraith (Top 46 by date)

Thanks. I already took a first look and it looks like I'm going in circles. The problem always is that I have to ask for what objects I want to see from the document and 'Contents' only yields the indirect references. I would be very happy to get to the point directly, but have no idea where the text is actually stored in the document or how to ask for this.

As you wish, but I doubt it will help very much. Everything so far is ok, but what can i do with the indirect references from there on?

You forgot to mention that posting the question in the Lounge will bring more interesting responses. :-)

Nah, only 4 for not mentioning 'one's complement' or 'two's complement' in the lecture and only hinting at the difference between those two.

It's not exactly 'on the fly', but combining conventional learning methods with an evolutionary algorithm may do the trick.

Just what I thought. :-)
That's what some people mean by saying 'programming'.

Maybe. I told him that he should take a good look at the Main() function and figure out what to pass to the constructors and functions. As they say in Texas: You can bring a horse to the water, but you can't drink for it.

You could have told him that the code in the Main() function would not compile as it is. He does not need to add any new lines, but he has to complete the existing lines in such a way that it would compile and running the program would produce the given output. He mostly has to deduce the right parameters for the calls to do that.

Saw a good modern interpretation just last week. With submarines and nuke torpedos. In the end they nuke an island instead, the captain resorts to try the traditional way with a harpoon and the submarine is sunk by MD.

Short inlined functions can make your executable smaller. Don't forget the hidden overhead of passing parameters and cleaning up the stack when calling regular functions.

Bonus points for looking into the stack trace to find the line which throws this error.

I would pick something simple and traditional. How about Frank?

He is wrestling with the same problems as many others: No understanding of the data types and how they are stored in memory, stringly typing and then parsing or converting to some not too well chosen datatype, only to cast it to another not well chosen datatype.

In the end it looks like he simply wants to initialize a variable with a literal 64 bit address and then mask out (guessing!) some part with a checksum.

Now I saw this:

p=new test [3];

That does not work because the type of p is wrong. This should work better:

test** p; // that's a pointer to a pointer of 'test'!
p= new test*[4];

This should give you an array with four pointers to objects of the type 'test'. Caution: These four pointers have not yet been initialized!

Something like this little loop (and don't forget to clean up later by deleting the four objects in a similar loop):

int i;
for(i = 0; i < 4; i++)
{
p[i] = new test(i, i + 1);
}

'test' is a class name, that's right. Declaring a class always introduces a new self defined complex datatype. To get an instance of the class (anywhere you need it, not just in main(), you must first declare a pointe to your class type and assign it a value by using the 'new' operator. Let's put these two things into two separate lines to make it more clear:

test* p; // this is just a normal variable declaration
// a pointer to an object of the type 'test'

p = new test(1, 2); // here we actualy give p a value.
// memory is allocated for the new object
// and the parameters 1 and 2 are passed to the
// constructor

Now you can use p to access the new object and call methods and do whatever you wanted to do with it. But there is one more important thing you should do when you don't need it anymore:

delete p; // this frees the memory that was used by the object

If you don't clean up your program will have massive memory leaks. In the main() function this may not be a problem, but it would still be sloppy. In othe functions or methods this is othing less than a bug.

These errors do not correspond to the updated code you have posted, at least as far as I can see. So we are back at the beginning. Now, we can play this endlessly or you can try to read the error messages and try to resolve them one by one. What, for example, is so hard to understand about something like 'test' was not declared in this scope? Do you really need us to tell you that you must declare variables (with the correct type!) before you use them?

These are all very basic mistakes. I know that this can be frustrating at the beginning, but you will not get very far if you don't learn to read the error messages. look at the code and fix it.

Start with the topmost error. Read the error message, look it up or ask us if you still don't understand what it means. And then you fix it. Recompile and repeat this procedure until you have no more errors or warnings.

Then it may be a good idea not to keep us guessing. What error did the compilers show? And usually both CodeBlocks and VS tell you at what code ine they have a problem.

I would suggest to look at the error messages, look up the meaning of the error and then taking a good look at the code line. If that does not help, come back and post the error message and don't forget to mention which code line here.

the number of true is 2

That explains everything.

You are joking, right? You will have better chances trying to sell pork chops in Arabia or refrigerators to the Eskimos than this surface primer for old ladies around here.

Gone are the days when we sent our response in the name of the Greek goddess of victory, also known under the profane name of MIM-14. µSAMs? More like bottle rockets and firecrackers.

What? Have they not been replaced by multicopters that sneak up on you and poke webcams through your window?

Or messengers on horseback. Pigeons also could work.

Dang! I hoped you would dig out a more exotic solution. I love to use a message broker for that. Or one form could send out an email and the other form processes incoming mails. Or we could try interprocess communication with the same process on both ends. One form sends, the other one receives. Or...

It's hard to argue with an error message. In such situations it's best not to assume anything, read the documentation and check the configuration. Maybe there is some error or something is still missing.

You write: I have 2 dll files. Say x.dll and y.dll . Both are in same folder.

And also: in ue4.16 ,it gives error "unable to load x.dll"

Does 'it' know in which folder you have your libraries? Obviously not, so read the documentation and try to find out where to set this path in the configuration.

That's the problem he is facing, and this is the face I'm palming. :-)

Perhaps he's from Laconia. Then you guys are in trouble because This. Is. Sparta!

Fine. Then set a breakpoint there and examine every part of Registered_Students.Registered.CurrentRow.Cells(0).Value, And how do you know that Cells contains any items at all? Examine the count property of Cells. If there is no Cells(0), you can't access its Value property. Something must be wrong there, otherwise there would be no exception.

Was it the line where you put together the sql string?

Did you set a breakpoint at the line shown in the stack trace of the exception and use the debugger to examine the values of the variables?

I can't tell you any more. This information is on your computer and I can not see it. You must learn how to get it.

YOur text and the SQLstatements don't fit together at all. There is not a single match. What is your question?

How about not trying to do everything at once? Loading and processing less rows at a time and doing this in a loop until all rows have been processed may help a lot.

Ah, ok.

Back to the topic: For a beginner it may be too easy to mistake an ASP .Net page and its lifecycle with something like WindowsForms. Perhaps it would already help him to learn why he can't simply put the code from the console application into the webpage and expict this to work, even if he starts a thread.

By the way, I would avoid that as well. If I remember correctly, you can also let the listener run asynchronously and you don't have to mess with the threads yourself. I think he copied this from some sample without really understanding it.

Yes, please do. You wrote 'My response kinda assumed you'd done all that.' - in your first response to anything I wrote.

So you obviously are a psychic and knew beforhand what I was going to write or you mistook me for the OP.

Then I don't quite get what you meant with 'My response kinda assumed you'd done all that.'

I'm not the OP. :-)

Windows also could shut down the console application any time it wants to, just for updates. Actually, it's a very good argument against Windows when you can't afford some service to be down.

True, but there may be an alternative. It may work to store the listener in a cache where it does not get destroyed when the page is unloaded. Now it can keep listening between page requests and can be retrieved any time the page is requested.

Now we only need a good place for the callback method that receives the incoming data and stores it in a way that allows to retrieve it whenever it is needed. And last we also need a scrap of code that reliably cleans up the listener when it's not needed anymore.

It may not be intended to be used this way, but I think I could pull it off without any sinister unsafe hacks.

No, we don't. I could easily write code that specifically constructs a clone of the entity and all further entities that are mapped to it. Then, one day someone has to implement a new feature by mapping yet another entity to this schema.

This would be the worse case, because it would probably not break my code as it is, but go on without an error, ignoring the change. That's a bug report waiting to happen.

What I would like is something like an entity.Clone() method, which automatically clones the entity and all entities that are mapped to it and storing them as new rows in the respective tables.

That is what I feared. I must write specific code that must be adapted when there are changes in the data schema. Forget to do that and we will have a new bug report and an excited customer. Too much of this can make maintaining an application hell.

Perhaps the Sysinternals Suite? You would have to run one of the tools on the server to monitor the IIS processes.

https://technet.microsoft.com/de-de/sysinternals/bb842062