|
It is entirely possible that finalSalary will not be assigned a value before it is used in execution (consider if experience is not 1, 2, or 3). This is what the compiler is complaining about.
You could assign finalSalary a value right after you declare it. You could also make the else branch assign it.
|
|
|
|
|
Your problem is that it is possible in your if statement to never assign a value to finalSalary since you declared finalSalary at the beginning and did not initialize it. So if someone enters a "4" at your prompt the code will execute the final "else" statement and write out to the console "That is not a valid entry" but will continue on to the final Console.WriteLine which will try to execute with an uninitialized variable.
Visual studio is smart enough to detect when people try to initialize variables inside of "if" statements but not smart enough to know if all paths assign a valid result to the variable so it always throws a flag when you try to do this.
The answer is to do one of the following:
1) initialize the variable at instantiation
2) move the Console.WriteLine statement to inside each "if" clause that you want it to print the final salary (probably your best bet since you don't want it to write out if the users response is invalid)
Hope this helps
|
|
|
|
|
I have an application that I've written using C#. I want to incorporate trialware functionality so that a user can run my program for X many days. Any suggestions on this?
I don't consider simply reading/writing to the registry an acceptable solution. Nor do I like the notion of a license file, As i am not maintaining any server. Perhaps some combination of these would prove an adequate solution.
I want to prevent the following:
1. user setting back the clock to get more free trial.
2. Uninstall - reinstall to get a new free trial
3. hacking a file or registry to turn off the trial. <- I don't think this is likely, given the intended customer base.
Thanks for your help.
|
|
|
|
|
Read http://www.codeproject.com/dotnet/xmldsiglic.asp[^] but even this won't satisfy your requirements. In fact I don't know of any technology available that will satisfy your list for .Net Framework.
Simply put, the old mechanisms and behavior necessary to do licensing on C/C++ runtime binaries will not work for anything that uses the .Net runtime. I do not know of anything that functions like FlexLM for .Net simply because of this, although I admit I haven't looked recently either. If you want rigorous license enforcement you might want to consider using C/C++ under Win32 instead.
|
|
|
|
|
thanks tom,
yes that is true, every code is hackable one way or the other.
i have looked into some of the product's out there that just do licensing. but they are very costly. but the one i liked is
www.xheo.com
thanks for the help.
Samar Aarkotti
|
|
|
|
|
I've done a bit research on the subject where the primary problem is that even signed binaries and assemblies can be subverted to avoid any license check. That is just the nature of the .Net Framework since it doesn't obey the classic rules for binary handling. The reason why classic C/C++ can be strongly licensed is that the runtime rules are tied deeply into the OS behavior. In .Net the rules are enforced only by .Net itself and therefore are much easier to subvert.
I've mentioned this before in similar questions on the topic: The .Net Framwork is wonderful for many things but being hardened isn't one of them. If the requirement is have a hardened, highly unhackable app you need to chose some other technology.
|
|
|
|
|
I beleive that is true, not anything like C/C++, even if i store the license file or something in the Isolated storage with Rigdal or some kind of encryption, it is still prone to hack.
but i guess average developer or user wouldn't try hard to read the assemblies and extract the code.
i guess i have to write a hack proof license management software 
the best way to go is to have a server authentication , when requested for license, server can create an encrypted license which is digitally signed. and pass it over to the user , and every time user run's it , validate against the server.
well i can keep on writing on the issues.
thanks
Samar Aarkotti
|
|
|
|
|
I wish you luck. If you find a bulletproof solution then you should sell it. This is the level of stuff Macromedia is wrestling and I don't think they've got a solution yet.
To hack that scheme you've just described I would recommend: Disassemble the binary executable to figure out the when the call happens to check the license file. It becomes a matter of simply removing the places it is called in the IL or altering the function to always return "yes I'm licensed" depending on which has a larger impact on the IL. Copy all of the rest of the IL into a new hacked version. If you really want to be clean you'd also remove the call to request an encrypted license file from the server using the same methods.
If there is no SSL, an even simpler way to do is just hang onto the encrypted license file and continually serve it to the application.
I've worked with a licensing scheme somewhat like you've described (btw, the server bit is a bit more complex than I suspect you think ). It is "good enough" but hardly hardened and light years away from bulletproof.
I would love for someone to come up with a .Net way of doing this kind of hardened licensing but all indications it is not possible inside of the .Net Framework. Instead to do this it has to come from *outside* where the very nature of assembly loading is tied in. This means that Fusion itself must be changed to do some more strenious validation. It must be done at that step because the moment you allow any IL to run on the CLR you've opened a place where you can subvert the license mechanism.
-- modified at 16:59 Friday 21st October, 2005
|
|
|
|
|
well in that case, create the license on the serverside for every new instance of the application and have it check against it, the only difference would be is to sign a new license everytime when the client connects to server with the minor difference of have append some kind of random string along with the computer name, or CID name or something like that.
changing license on the FLY for every invocation.
|
|
|
|
|
here is the application i am developing
http://speechtoolscenter.com/Products.htm
This is the dumbest website i have everdone, for some reason i hate doing webpages.
|
|
|
|
|
Hi folks!
I'd like to create a configurable hotkey assignment for my application. When a hotkey is pressed, a menuitem in my application should be invoked.
Creating these assignments within code is trivial, but I want the assignments to be configurable by the user and to be serializable, so that the user can save his/her assignment after configuration and load it again.
Unfortunately, MenuItem isn't marked as Serializable, so I don't know how to save the MenuItem the user wants to invoke with a given hotkey... 
Any ideas how to approach this issue?
Thanks in advance,
mav
|
|
|
|
|
Hi!
Try creating your menus at run-time instead of design-time. And you can assign the hotkeys you want for each one of them based on the previously saved hotkeys (only these need to be saved).
Regards
|
|
|
|
|
Thanks for the reply, but unfortunately that's not the point.
When I want the user to be able to configure the assignment I have to save the information which hotkey is assigned to which menu item. The hotkey part is easy, I just don't know how to uniquely identify a MenuItem.
One idea was to save the member name of the MenuItem and then, when loading a set of hotkey assignments, to retrieve the correct instance using reflection, but that's rather clumsy and cannot be encapsulated well.
Right now I'm trying to save the path to the MenuItem (something like "\E&xtras\&Format\&Defaults") and then, when a hotkey is pressed, find the correct menu item by iterating through the main menu items and their subitems using this path.
This does work and since the path is just a string the whole assignment can be serialized nicely, but it also makes the assignment dependent on the exact menu item names. Using the same assignment for different languages is not possible.
Because it's also possible to customize the menu itself (for example to hide unneeded menu items) I can't use the MenuItem's index within the parent menu instead of the item text, because this can get messed up as well...
It's too bad that MenuItem doesn't have a Name or Tag property or else I could use this. That's why I was asking for any other glorious idea on how to handle the assignment in a robust and flexible way...
Thanks,
mav
|
|
|
|
|
Ive set up the simplest of all remoting examples, a simple service hosted by a server (single call) and a client that calls a method on the service object. That works great I get my results, but when I quit the client, the server crashes with a null pointer exception. Other examples I've downloaded and compiled and run behave in the same way.. Anyone?
|
|
|
|
|
Without code snipets and more detail there is any number of reasons why the server could and would crash. I suspect you are doing something you shouldn't to the remote object on the "application exit event".
|
|
|
|
|
Sorry for not posting any code. As I said before, all the samples from tutorials fail in the same manner.
Heres the service:
public class HelloService : MarshalByRefObject<br />
{<br />
public string hello()<br />
{<br />
Console.WriteLine("Hello!");<br />
return "Hello";<br />
}<br />
}
Here's the server:
class HelloServer<br />
{<br />
static void Main(string[] args)<br />
{<br />
Console.WriteLine("Hit any key to exit at any time");<br />
<br />
try<br />
{<br />
TcpServerChannel channel = new TcpServerChannel(8088);<br />
<br />
ChannelServices.RegisterChannel(channel);<br />
RemotingConfiguration.RegisterWellKnownServiceType(typeof(HelloService), "HelloService", WellKnownObjectMode.SingleCall);<br />
}<br />
catch (Exception e)<br />
{<br />
Console.WriteLine(e.Message);<br />
}<br />
<br />
Console.ReadKey();<br />
}<br />
}
And here's the client:
class HelloClient<br />
{<br />
static void Main(string[] args)<br />
{<br />
Console.WriteLine("Press any key to exit at any time");<br />
<br />
try<br />
{<br />
TcpClientChannel channel = new TcpClientChannel();<br />
<br />
ChannelServices.RegisterChannel(channel);<br />
HelloService service = (HelloService)Activator.GetObject(typeof(HelloService), "tcp://localhost:8088/HelloService");<br />
<br />
if (service != null)<br />
Console.WriteLine(service.hello());<br />
}<br />
catch (Exception e)<br />
{<br />
Console.WriteLine(e.Message);<br />
}<br />
<br />
Console.ReadKey();<br />
}<br />
}
|
|
|
|
|
Are you running them both in the (same) debugger together? (i.e. using debug->start new instance from the context menu of the solution window?)
If so, the console object is destroyed when you exit either one... try running them in separate instances of VS.
Absolute faith corrupts as absolutely as absolute power
Eric Hoffer
All that is necessary for the triumph of evil is that good men do nothing.
Edmund Burke
|
|
|
|
|
Thanks for the replies..
Nah I'm running them separately outside of VS. I've tried both debug and release modes. To me it sounds like the client needs to release something or tell the server about its disposal but it just doesn't make any sense.
|
|
|
|
|
Hi There Friends,
How Can I Scroll an scrollable control (like listview listbox) manually?
they dont have scroll method.
|
|
|
|
|
Hello Ali!
If you set the AutoScroll property to True then you can scroll your control using the following sample code
ConrolName.AutoScrollPosition = new Point(XVal, YVal) // here XVal refers to horizontal scroll and YVal refers to Vertical Scroll value.
When you read the value of AutoScrollPosition property, it returns negative values, but, you must set positive values to scroll the control.
I hope it works.
|
|
|
|
|
|
hey guys! (sorry about my english)
so, i know the great tool dcomperm.exe, but i don't want use this .exe..
what i realy want is: to use a .net assembly with all existing functions on dcomperm.
so i just start to rewrite my own dcomperm in c#, but..! its a LOT of work.. its almost insane...
1- so, anyone know a finished .net component or something like that??
2- i try to convert the C++ unmanaged dcomperm code(from samples VS 6), to
C++.net, but im not so close to C++ code, its a hard work.. it would be a easer workaround??
thanks a lot guys..
if i get finished my own dcomperm, i'll build a codeproject article
to share it!
blitzkrieg bop!!
|
|
|
|
|
Hi
Using C#, how can I break a string into separate words. For instance, " my search string, and some-other-string"
becomes "my" "search" "string" "and" "some-other-string"
Notice, that I have removed spaces and commas, but kept the dashes in my words.
So any idea how I can remove spaces, commas and break a string into separate words??
Kind regards
|
|
|
|
|
You can do it by using Substring() method, or by Split() method, but the last one requires that there is a separator, this separator may be the " " Char.
"Praying." Is this only what I can do for him ?
|
|
|
|
|
If your string is not that large, you can use the Split[^] method on System.String. Something like
string []words = sentence.Split(new char[] {' ', ','});
You need to watch out for empty strings in words though, two successive delimiters will result in an empty string.
If your string is large or your parsing logic is complex, you can try using theRegex[^] class.
Regards
Senthil
_____________________________
My Blog | My Articles | WinMacro
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
