|
Or possibly as some sort of plug-in system where each utility has its own Panel or TabPage in its DLL.
|
|
|
|
|
Hi,
I have a function that converts r(red), g(green) and b(blue) values to a uint (same as RGB macro in C++).
If I implement as follows I get no errors or warnings:
private uint rgb(byte r, byte g, byte b)
{
return ((uint)((r|((uint)(g)<<8))|(((uint)b)<<16)));
}
However, if I implement like this:
private uint rgb(byte r, byte g, byte b)
{
return ((uint)((r|((ushort)(g)<<8))|(((uint)b)<<16)));
}
I get a warning "Bitwise or operator used on a sign extended operand".
But ushort is un-signed, so why am I getting the warning?
Thanks - John.
|
|
|
|
|
ushort will be being expanded to uint because b is still being shifted as a uint (as it must be). That will cause the ushort to be 'sign extended' i.e. the new bits to be filled with the top bit, I suppose, though that doesn't really make sense (an unsigned type should always be zero extended).
|
|
|
|
|
It would appear that the logical expression is yielding an int value which is,by definition, signed. You should cast each part of the expression to uint as in the first case. you should also review your use of parentheses around your expressions, a better expression would be:
return (uint)(r | (((uint)g) << 8) | (((uint)b) << 16));
return (uint)(r | (g << 8) | (b << 16));
|
|
|
|
|
Thanks Guys,
I would have expected an expression containing unsigned variables to yield an unsigned, but I must be wrong and by default it yields an int.
return (uint)(r | (g << 8) | (b << 16));
...Seems like the best option.
Thanks again.
|
|
|
|
|
The documentation[^] is a little bit ambiguous, but the implication is that a left shift will return int (Int32) by default, hence the problems you encountered. I am sure there is a more explicit rule somewhere in the depths of the language specification if you care to look (but hey, life is short!).
|
|
|
|
|
hi
i had an set of x and y coordinates , it is a closed loop of polygon, so how we can find the area so please give me the idea
|
|
|
|
|
This may not be easy depending on the polygon[^].
"Last year I went fishing with Salvador Dali. He was using a dotted
line. He caught every other fish." Steven Wright
|
|
|
|
|
Triangulate the polygon (you can find this algorithm in lots of places) and add up the area of all the triangles.
|
|
|
|
|
Exactly what I would have recommended. The better algorithms can even deal with concave polygons or 'holes'.
And from the clouds a mighty voice spoke:
"Smile and be happy, for it could come worse!"
And I smiled and was happy
And it came worse.
|
|
|
|
|
There is a better algorithm than adding triangles
It is quite straight forward you need calculate two sums
sum1 = sum of ( pt[i].x * pt[i+1 % n].y)
sum2 = sum of ( pt[i+1 % n].x * pt[i].y)
then the area is sum1 - sum2
http://www.mathopenref.com/coordpolygonarea.html[^]
|
|
|
|
|
Excellent.
|
|
|
|
|
Be aware that this method does not work with polygons with crossovers. Mind you, that is true of simplistic triangulators too.
This is a good link if the OP does not need the triangulated form for other purposes (e.g. collision detection).
|
|
|
|
|
Excellent answer, and an interesting site, thanks +5 !
It appears this algorithm will work on polygons that have areas that "go inward," as long as they do not cross.
best, Bill
"Last year I went fishing with Salvador Dali. He was using a dotted
line. He caught every other fish." Steven Wright
|
|
|
|
|
I need to get the free physical memory available so I don't get OutOfMemory Exception.
I have tried GlobalMemoryStatusEx but it seems the windows swaps out ram to virtual and the value is unreliable and ultimately results in out of memory also.
Any thoughts?
Its the man, not the machine - Chuck Yeager
If at first you don't succeed... get a better publicist
If the final destination is death, then we should enjoy every second of the journey.
|
|
|
|
|
You might want to have a look at MemoryFailPoint Class[^].
Creating an instance of a MemoryFailPoint class creates a memory gate. A memory gate is a check for sufficient resources prior to initiating an activity requiring a large amount of memory. Failing the check throws an InsufficientMemoryException[^] that avoids starting an operation, reducing the possibility of an application failing during execution due to lack of resources. This allows an application to decrease its performance in an effort to avoid an OutOfMemoryException[^] and any state corruption that may result from improper handling of an OutOfMemoryException[^] in arbitrary locations in code.
0100000101101110011001000111001011101001
|
|
|
|
|
Getting the free Physical memory will not necessarily help avoid an OutOfMemory exception - .NET is perfectly happy using virtual memory.
Have a look at this article: it covers the OOM problem in some depth. http://msdn.microsoft.com/en-us/magazine/cc163528.aspx[^]
Ideological Purity is no substitute for being able to stick your thumb down a pipe to stop the water
|
|
|
|
|
And running out of physical memory is not the only way to throw an OutOfMemory exception. Try running the handle pool dry and you'll do the same thing, even though there's plenty of free memory.
You can do this by creating and NOT disposing of Brushs, Pens, Graphics objects, Forms shown with ShowDialog, ...
|
|
|
|
|
I'm a relatively new c# programmer (transitioning from C++)and am working with a grid application. I've successfully bound the grid to a data structure but would now like to get rid of 0s in the rows. There must be an easy way to tell the grid to show nothing if the default value (or user input is 0) but I haven't found it. Also, I'd like to know if there's an easy way to do data validation on individual columns (ie non-negative int, double, etc.) Thank you.
|
|
|
|
|
Tom Paronis wrote: There must be an easy way to tell the grid to show nothing if the default value
(or user input is 0) but I haven't found it.
There's an event that you could use; CellFormatting[^], and there's a how to[^] article on MSDN explaining formatting that would be helpfull.
There's also a walktrough[^] available on MSDN.
Bastard Programmer from Hell 
|
|
|
|
|
Thanks. This is the second time I've been bailed out by the Bastard programmer from Hell. I owe you a drink.
|
|
|
|
|
You're welcome, and thanks for the coffee
|
|
|
|
|
Using 0 for your default value is quite dangerous (0 is often a valid input). If you're using a floating point type, NaN is quite good for 'missing data', and this should be what you use in your data model. For integer types you can either specify a special value or use int? and use null for missing. You can then use the CellFormatting tip to show a blank cell when the actual value matches your designated 'null' value.
|
|
|
|
|
Thanks. I've just developed an abhorrence over the years of seeing a whole lot of 0s on the screen when all it means is that the user hasn't filled the data in yet.
|
|
|
|
|
Greetings Gurus,
Come across a problem I don't seem able to solve and was hoping someone could point me in the right direction.
I am building a Windows service which records Event log entries in a sql db. Running it as a console application works fine but when I try and move the code into a service structure I get errors.
My code
public partial class GetInfo : ServiceBase
{
public ServiceName()
{
InitializeComponent();
}
static AutoResetEvent appsignal, syssignal, secsignal;
public static void Main()
{
What should I use in place of "public static void Main"?
I would like to keep this out of the "Onstart" region because I want it to initialise before the service's timer.
Can you help?
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
