Problem to remove \0 in the array in the best way

[chrisben]

Hi,

I have a windows C# app which read data through socket from a C++ app in a
linux box. After a marshal a pointer and get an array of string ends with \0,
I did something like this

string s=new string(array);
s = s.Replace("\0","");
if(s == "ABC")
success
else
fail

However, if I did it in my test all in C#, it works. If I plug into the
production and start to listen to the linux box messages, it always fail. If
Studio's output window, I can see s becomes ABC~ instead of ABC.

Does anyone here know what is the most efficient way to solve this problem?
I do can iterate the array to detect \0, but I do want the most efficient way
since I am facing a very heavy traffic from that linux box.

Thanks a lot

Chris

[Jon Skeet [C# MVP]]

chrisben <ch******@discussions.microsoft.com> wrote:

I have a windows C# app which read data through socket from a C++ app in a
linux box. After a marshal a pointer and get an array of string ends with \0,
I did something like this

string s=new string(array);
s = s.Replace("\0","");
if(s == "ABC")
success
else
fail

However, if I did it in my test all in C#, it works. If I plug into the
production and start to listen to the linux box messages, it always fail. If
Studio's output window, I can see s becomes ABC~ instead of ABC.

Does anyone here know what is the most efficient way to solve this problem?
I do can iterate the array to detect \0, but I do want the most efficient way
since I am facing a very heavy traffic from that linux box.

Well, your code is definitely removing \0 characters, so the rogue
character in s is something different. I suggest you print out its
Unicode value.

--
Jon Skeet - <sk***@pobox.com>
http://www.pobox.com/~skeet
If replying to the group, please do not mail me too

[Jon Skeet [C# MVP]]

chrisben <ch******@discussions.microsoft.com> wrote:

I have a windows C# app which read data through socket from a C++ app in a
linux box. After a marshal a pointer and get an array of string ends with \0,
I did something like this

string s=new string(array);
s = s.Replace("\0","");
if(s == "ABC")
success
else
fail

However, if I did it in my test all in C#, it works. If I plug into the
production and start to listen to the linux box messages, it always fail. If
Studio's output window, I can see s becomes ABC~ instead of ABC.

Does anyone here know what is the most efficient way to solve this problem?
I do can iterate the array to detect \0, but I do want the most efficient way
since I am facing a very heavy traffic from that linux box.

Well, your code is definitely removing \0 characters, so the rogue
character in s is something different. I suggest you print out its
Unicode value.

--
Jon Skeet - <sk***@pobox.com>
http://www.pobox.com/~skeet
If replying to the group, please do not mail me too

[chrisben]

Thanks. Actually, I checked the input again and realize there is more than /0
in the input. That is a 5 byte array, after converting to char array, it
shows as below in debu window

{Length=5}
[0]: 65 'A'
[1]: 73 'B'
[2]: 89 'C'
[3]: 0 ''
[4]: 732 '˜'
Any idea about what is going on?

Thanks

"Jon Skeet [C# MVP]" wrote:

chrisben <ch******@discussions.microsoft.com> wrote:

I have a windows C# app which read data through socket from a C++ app in a
linux box. After a marshal a pointer and get an array of string ends with \0,
I did something like this

string s=new string(array);
s = s.Replace("\0","");
if(s == "ABC")
success
else
fail

However, if I did it in my test all in C#, it works. If I plug into the
production and start to listen to the linux box messages, it always fail. If
Studio's output window, I can see s becomes ABC~ instead of ABC.

Does anyone here know what is the most efficient way to solve this problem?
I do can iterate the array to detect \0, but I do want the most efficient way
since I am facing a very heavy traffic from that linux box.

Well, your code is definitely removing \0 characters, so the rogue
character in s is something different. I suggest you print out its
Unicode value.

--
Jon Skeet - <sk***@pobox.com>
http://www.pobox.com/~skeet
If replying to the group, please do not mail me too

[chrisben]

Thanks. Actually, I checked the input again and realize there is more than /0
in the input. That is a 5 byte array, after converting to char array, it
shows as below in debu window

{Length=5}
[0]: 65 'A'
[1]: 73 'B'
[2]: 89 'C'
[3]: 0 ''
[4]: 732 '˜'
Any idea about what is going on?

Thanks

"Jon Skeet [C# MVP]" wrote:

chrisben <ch******@discussions.microsoft.com> wrote:

I have a windows C# app which read data through socket from a C++ app in a
linux box. After a marshal a pointer and get an array of string ends with \0,
I did something like this

string s=new string(array);
s = s.Replace("\0","");
if(s == "ABC")
success
else
fail

However, if I did it in my test all in C#, it works. If I plug into the
production and start to listen to the linux box messages, it always fail. If
Studio's output window, I can see s becomes ABC~ instead of ABC.

Does anyone here know what is the most efficient way to solve this problem?
I do can iterate the array to detect \0, but I do want the most efficient way
since I am facing a very heavy traffic from that linux box.

Well, your code is definitely removing \0 characters, so the rogue
character in s is something different. I suggest you print out its
Unicode value.

--
Jon Skeet - <sk***@pobox.com>
http://www.pobox.com/~skeet
If replying to the group, please do not mail me too

[Jon Skeet [C# MVP]]

chrisben <ch******@discussions.microsoft.com> wrote:

Thanks. Actually, I checked the input again and realize there is more than /0
in the input. That is a 5 byte array, after converting to char array, it
shows as below in debu window

{Length=5}
[0]: 65 'A'
[1]: 73 'B'
[2]: 89 'C'
[3]: 0 ''
[4]: 732 '?'
Any idea about what is going on?

It looks like you're potentially using the wrong encoding to convert
the data into characters, and also potentially you're not decoding the
right amount of data in the first place.

What does the code you're using to read from the socket look like?

--
Jon Skeet - <sk***@pobox.com>
http://www.pobox.com/~skeet
If replying to the group, please do not mail me too

[Jon Skeet [C# MVP]]

chrisben <ch******@discussions.microsoft.com> wrote:

Thanks. Actually, I checked the input again and realize there is more than /0
in the input. That is a 5 byte array, after converting to char array, it
shows as below in debu window

{Length=5}
[0]: 65 'A'
[1]: 73 'B'
[2]: 89 'C'
[3]: 0 ''
[4]: 732 '?'
Any idea about what is going on?

It looks like you're potentially using the wrong encoding to convert
the data into characters, and also potentially you're not decoding the
right amount of data in the first place.

What does the code you're using to read from the socket look like?

--
Jon Skeet - <sk***@pobox.com>
http://www.pobox.com/~skeet
If replying to the group, please do not mail me too