C# Discussion Boards - CodeProject

2-Feb-05 23:29

Robert Rohde wrote:
2. A service can do everything a normal app can do.

With restrictions. If the service runs as SYSTEM - and you should really determine if you need SYSTEM privileges - the service needs to be marked for interactive use by bitwise OR'ing the Type registry key for the service name under HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services. In that case, you would typically impersonate the caller or pipe owner, etc.

This posting is provided "AS IS" with no warranties, and confers no rights.

Software Design Engineer
Developer Division Sustained Engineering
Microsoft

[My Articles] [My Blog]

Re: Services

Heath Stewart2-Feb-05 23:39

how can i show the character if a UNICIODE in textbox

2-Feb-05 23:39

You do not necessarily have to stop a service when servicing your application, but a reboot will be required if you don't (to act on the pending file rename operation - or PFRO - that was scheduled since your service executable was in use).

Whether or not you can remotely update the service depends on what remote administrative tools you have set up. There's nothing built into Windows to enable this functionality directly, but you can use WMI to stop the service, use the Admin share (like \\SomeMachine\c$) to copy over the service executable, then use WMI to restart the service. This wouldn't require an interactive login, but your connecting user account would still be authenticated.

A strong word of caution (and this is an area I know well) - really evaluate what the minimum set of permissions are. Too many services often run as SYSTEM, which grants your service full administrative rights. It's highly discouraged to run as SYSTEM unless you have to. Run either as an interactive user (which allows you to create window stations in which to run Windows applications - visible or not), NetworkService (for basic rights + network privileges) or LocalService for a decent amount of privileges.

You can find a lot of information about Windows services - and just about everything else regarding Windows development - at http://msdn.microsoft.com[^].

For information specfic to services, see http://msdn.microsoft.com/library/en-us/dllproc/base/services.asp[^]. To learn more about the privileges granted to LocalSystem (SYSTEM), LocalService, and NetworkService see http://msdn.microsoft.com/library/en-us/dllproc/base/service_user_accounts.asp[^].

You can also create window stations (which host desktops, which are required for Windows applications) by running as SYSTEM and OR'ing the Type registry key with 256 (when I get time to finish it, I'll be releasing an article describing this in more detail) as I mentioned in my indirectly reply to your post. It is recommended in this case to impersonate the caller, typically calling ImpersonateLoggedOnUser which you'd have to P/Invoke.

This posting is provided "AS IS" with no warranties, and confers no rights.

Software Design Engineer
Developer Division Sustained Engineering
Microsoft

[My Articles] [My Blog]

mirzaei2-Feb-05 12:11

mirzaei

2-Feb-05 12:11

Hi Is there any body help me to solve my problem ?
I read a HTML file from web unsing MSWinSock and I need show some strings contaned in it that is UNICODE (such as "س") in a textbox, but i can't convert it to original carachters readable for human. Cry | :((

Please help me
if you can help me please send your solution to my mail box : mirzaei@gmail.com
Thank You

Re: how can i show the character if a UNICIODE in textbox

Heath Stewart2-Feb-05 23:49

Re: how can i show the character if a UNICIODE in textbox

2-Feb-05 23:49

What does this have to do with C#? If you were using C# (or .NET in general) you shouldn't be using Winsock but the System.Net classes, like the HttpWebRequest, or more simply the WebClient, to retrieve a web page. Unnecessarily interoperating with native code (like Winsock) creates security dependencies that may or may not be granted to your code.

What is it exactly that you're asking? Do you want to show a Unicode (not all caps, BTW) character as س, or resolve س to a Unicode character?

If you understand HTML (or actually SGML that defines entities like this), it should be simple. The representation above is a decimal (base 10) representation. If you see an x, like س, then it's hexadecimal (base 16). In either case you could parse the number, convert it to hex (if necessary), then split it into 2 characters (2 hex digits). Use the UnicodeEncoding.GetString or UnicodeEncoding.GetChars to covert that byte[2] array into a char (in .NET, a char is always stored in Unicode) or a string (again, always stored in Unicode internally).

mirzaei wrote:
if you can help me please send your solution to my mail box : mirzaei@gmail.com

All forums on Code Project are message boards. Direct email is not encourage and should not be expected. If you entered your address correctly into your profile you will receive notification email by default when someone replies to you.

This posting is provided "AS IS" with no warranties, and confers no rights.

Software Design Engineer
Developer Division Sustained Engineering
Microsoft

[My Articles] [My Blog]

Heath Stewart3-Feb-05 0:07

Jpeg images, YCbCr encoding, and .Net

3-Feb-05 0:07

If you need to convert a Unicode character to that representation, then it's quite easy. Use UnicodeEncoding.GetBytes to get a byte[] array, then loop over that byte[] array and print either decimal or hexadecimal.

The following example loops in 2-byte increments and prints a decimal format:

using System;
using System.Text;
 
class HtmlEntities
{
  static void Main(string[] args)
  {
    if (args.Length > 0)
    {
      Console.WriteLine(HtmlFormat(args[0]));
    }
  }
 
  static string HtmlFormat(string value)
  {
    StringBuilder sb = new StringBuilder(value.Length * 7);
    Encoding enc = Encoding.Unicode;
 
    foreach (char c in value)
    {
      byte[] buffer = enc.GetBytes(new char[] {c});
      int i = BitConverter.ToInt16(buffer, 0);
      sb.Append("&#" + i.ToString() + ";");
    }
 
    return sb.ToString();
  }
}

This posting is provided "AS IS" with no warranties, and confers no rights.

Software Design Engineer
Developer Division Sustained Engineering
Microsoft

[My Articles] [My Blog]

object882-Feb-05 12:09

object88

2-Feb-05 12:09

I have to render to screen images that come from a variety of sources (such as faxes, etc.), and some of those images are TIFF files with YCbCr encoded Jpegs. Unfortunately, .Net doesn't handle these kinds of images.

I need some help understanding how to decode a jpeg. I have enough code to take apart the TIFF, and I have some sample code which will translate YCbCr values into RGB values, but I could really use some help with decoding the jpeg itself.

Any pointers?

Christian Graus2-Feb-05 12:11

2-Feb-05 12:11

You need to read it manually ? If you just read it in, the byte array won't read in with YCbCr values ?

Christian

I have several lifelong friends that are New Yorkers but I have always gravitated toward the weirdo's. - Richard Stringer

object882-Feb-05 12:32

object88

2-Feb-05 12:32

Well, even if I read it in manually, it's compressed. That's the trick: how do I handle jpeg compression? Jpegs aren't stored in a simple pixel-by-pixel format. Or... so I gather. Confused | :confused:

Christian Graus2-Feb-05 12:36

2-Feb-05 12:36

C# can read a number of formats, including jpeg, and give you a byte array. Read my image processing articles for more info.

So, if the jpeg just contains YCbCr instead of RGB, it would be a simple case of reading the file, iterating over it, and converting each pixel to an RGB value.

Christian

I have several lifelong friends that are New Yorkers but I have always gravitated toward the weirdo's. - Richard Stringer

object883-Feb-05 10:00

object88

3-Feb-05 10:00

I think that there's something broken in the image I was testing with. I have a single-page TIFF which I can extract a YCbCr encoded jpeg from and open with a .Net bitmap. Thanks for the help!

Adding Class file

dasmeher2-Feb-05 11:32

dasmeher

2-Feb-05 11:32

Hi All,

I added a C# class file to the C# project.
I want to use the drop down control at runtime. I get an error when I write
using System.Web.UI.WebControls;

on top of the page. Can't we not use it?? How do I access any control at runtime in a class. Please let me know.

Thanks.

Christian Graus2-Feb-05 11:48

2-Feb-05 11:48

You're probably trying to use web controls in a project that has not imported the dll that contains them

But that's a wild guess, seeing as you've not told us the error.

Christian

I have several lifelong friends that are New Yorkers but I have always gravitated toward the weirdo's. - Richard Stringer

MyThread2-Feb-05 11:55

MyThread

2-Feb-05 11:55

Thanks. The error is

The type or namespace name 'UI' does not exist in the class or namespace 'System.Web' (are you missing an assembly reference?)

Christian Graus2-Feb-05 11:58

2-Feb-05 11:58

Yep - I'm right then. I'm not sure if you can use the web controls outside an ASP.NET project. They work by generating HTML in conjuction with the ASP.NET ISAPI filter, you realise that, right ? What's wrong with the Windows controls ?

Christian

I have several lifelong friends that are New Yorkers but I have always gravitated toward the weirdo's. - Richard Stringer

MyThread2-Feb-05 12:07

MyThread

2-Feb-05 12:07

Thanks. Ok, now I just converted a code from VB.net class file to a C# class file. This works fine in VB.net. But it gave error here.

Christian Graus2-Feb-05 12:10

2-Feb-05 12:10

OK, in that case your converter sucks. Just add the reference ( there's a references section on the top of the project, just right click and choose add reference, and look for System.Web, or whatever it is you need ) and you should be fine.

Christian

I have several lifelong friends that are New Yorkers but I have always gravitated toward the weirdo's. - Richard Stringer