<div>
It seems that your C++ DLL is compiled as UNICODE which causes this effect. You can either set that to compile as ANSI manually (go to &quot;Project Settings|General&quot; and set &quot;Character Set&quot; to &quot;Use Multibyte Character Set&quot; or strictly stick to ANSI datatypes, i.e.<br />

<pre><code class="code-1 nolinks" id="code-260227"><span>extern &quot;C&quot; __declspec(dllexport) BOOL WINAPI MYTEST(LPSTR PW, DWORD BlockNumber, DWORD Date, WORD Options, DWORD ProductCode) //'LPSTR' for ANSI charset</span>
</code></pre>
</div>


<div>
Unfortunately, I do not have the source code to the DLL. I only have the declaration for it. So somehow I need to make it work in C#.
</div>


<div>
you can use <br />
UnmanagedType.LPWStr instead of UnmanagedType.LPTStr <br />
or <br />
use CharSet = CharSet.Auto in Dllimport.<br />

</div>


<div>
&gt;&gt;&gt;&gt; MyPassword is being returned as Chinese characters but they should be ANSI.<br />
<br />
You probably mean your password was &quot;passed&quot; as Chinese UNICODE while the dll expected it to be ANSI. AFAIK C# has only wide char strings (Microsoft: UNICODE), i. e. the String and C# - marshalling type LPTSTR both were wide char. <br />
<br />
C++ dlls on the other hand default to be ANSI text what means that the LPTSTR in the C++ API is supposed to match to LPSTR what is char* (and not wchar_t* as it was sent by the caller).<br />
<br />
If I am right, you simply could exchange the LPTSTR by LPSTR in both interfaces and it firstly should compile without errors at the C++ and secondly should pass the password as ANSI cause the marshalling can do the conversion. 
</div>


<div>
&gt;&gt;&gt;&gt; So somehow I need to make it work in C#.<br />
<br />
If I was right with my analysis you only need to change to<br />
<br />
&nbsp;[DllImport(&quot;MyTest.dll&quot;, CharSet = CharSet.Ansi)]<br />
&nbsp; &nbsp; &nbsp; &nbsp; public static extern bool MYTEST([MarshalAs(Unmanage<wbr />dType.LPSt<wbr />r)] string password, UInt32 ID, UInt32 Date, UInt16 Options, UInt32 ProductCode);<br />
<br />
<br />
(LPSTR instead of LPTStr)
</div>


<div>
Hi itsmeandnobodyelse:<br />
<br />
I changed the line as you suggested and now it returns nothing. The string is blank just as it was prior to call. &nbsp;<br />
<br />
I just did a test in VB6 and it works fine. So I know the DLL works, just not from C# due to something in my declaration.<br />
<br />
Here is my VB6 declaration (it works) just in case it might help:<br />
Public Declare Function MYTEST Lib &quot;MyTest.dll&quot; (ByVal Password As String, ID As Double, _<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; MyDate As Double, Options As Long, ProductCode As Double) As Boolean<br />
<br />
Here is my VB6 call:<br />
MyPassword = String(255, 0)<br />
Call MYTEST (MyPassword, ID, MyDate, Options, ProductCode)
</div>


<div>
If your DLL returns a UNICODE string and you don't have the source code, you'll have to convert that manually on the C# side... yet that sounds odd, since .NET AFAIK is UNICODE by default.
</div>


<div>
unmeshdave:<br />
<br />
I tried all combinations of your LPWStr suggestion with .Auto and all returned Chinese characters.<br />
<br />
Just a reminder to all experts, I do NOT have the source to the DLL, only the declaration. 
</div>


<div>
I don't think the DLL is returning a UNICODE string. Why do you think that?
</div>


<div>
Because chinese characters cannot be encoded in ANSI.
</div>


<div>
The DLL is only designed to use ANSI so the Chinese are bizzare. Could it be some setting in my C# project that is messing it up?<br />
<br />
Also, it works fine from a VB6 project. See my note above.
</div>


<div>
u shoould not combined both the suggestions...<br />
Did u <br />
Try this,<br />
UnmanagedType.LPWStr instead of UnmanagedType.LPTStr &nbsp;with &nbsp;CharSet = CharSet.Ansi <br />
OR THIS,<br />
CharSet = CharSet.Auto in DllImport with &nbsp;UnmanagedType.LPTStr in MYTEST.<br />
?<br />

</div>


<div>
Here is a sample output of the password from debugger:<br />
&quot;NePnKmiBSTAJ &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&quot;
</div>


<div>
Hmmm, &nbsp;when I posted the string to EE it changed it to ANSI. &nbsp;I do not know why??? &nbsp;When I paste it into EE, it looks like Chinese, but they when I press &quot;SUBMIT&quot; it changes it to ANSI. &nbsp;Trust me it looks like random Chinese chars in the string.
</div>


<div>
when u run your program instead of debug, do u get same chinese characters or u get proper ANSI string? try to put some message box to show the string u get.<br />
sometimes, C++ dll comes in two versions, one for debug &amp;&nbsp;another for release version. I dont think it should matter with Interop services, but u can just give it a try.<br />

</div>


<div>
jkr: <br />
<br />
Yes, I am trying to get the password returned from the DLL. &nbsp;I tried it using your StringBuilder suggestion and in the debugger, the password is still showing up as Chinese chars as well. &nbsp;
</div>


<div>
The MessageBox.Show result is attached to this msg. It also shows Chinese chars.<br />
<a href="https://filedb.experts-exchange.com/incoming/2009/02_w09/110437/junk.jpg" target="_blank" class="file-inline" title="MessageBox result with chinese chars (6 KB)"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path d="M67.508 468.467c-58.005-58.013-58.016-151.92 0-209.943l225.011-225.04c44.643-44.645 117.279-44.645 161.92 0 44.743 44.749 44.753 117.186 0 161.944l-189.465 189.49c-31.41 31.413-82.518 31.412-113.926.001-31.479-31.482-31.49-82.453 0-113.944L311.51 110.491c4.687-4.687 12.286-4.687 16.972 0l16.967 16.971c4.685 4.686 4.685 12.283 0 16.969L184.983 304.917c-12.724 12.724-12.73 33.328 0 46.058 12.696 12.697 33.356 12.699 46.054-.001l189.465-189.489c25.987-25.989 25.994-68.06.001-94.056-25.931-25.934-68.119-25.932-94.049 0l-225.01 225.039c-39.249 39.252-39.258 102.795-.001 142.057 39.285 39.29 102.885 39.287 142.162-.028A739446.174 739446.174 0 0 1 439.497 238.49c4.686-4.687 12.282-4.684 16.969.004l16.967 16.971c4.685 4.686 4.689 12.279.004 16.965a755654.128 755654.128 0 0 0-195.881 195.996c-58.034 58.092-152.004 58.093-210.048.041z"/></svg>junk.jpg</a>
</div>


junk.jpg

<div>
r u getting this while running? (not debug).
</div>


<div>
unmeshdave:<br />
<br />
I added MessageBox and rebuilt as RELEASE and then ran the EXE file. Same result as before with Chinese chars.<br />
<br />
I tried the LPWStr with ANSI with same bad result.<br />
<br />
I don't have the souce code to DLL so I can't try your suggestion <br />
&quot;CharSet = CharSet.Auto in DllImport with &nbsp;UnmanagedType.LPTStr in MYTEST&quot;<br />
<br />

</div>


<div>
CharSet = CharSet.Auto in DllImport with &nbsp;UnmanagedType.LPTStr in MYTEST<br />
this you have to try in your call,<br />
like,<br />
[DllImport(&quot;MyTest.dll&quot;, CharSet = CharSet.Auto)]<br />
&nbsp; &nbsp; &nbsp; &nbsp; public static extern bool MYTEST([MarshalAs(Unmanage<wbr />dType.LPTS<wbr />tr)] string password, UInt32 ID, UInt32 Date, UInt16 Options, UInt32 ProductCode); <br />

</div>


<div>
unmeshdave:<br />
<br />
Okay. I just tried it with same bad result.<br />
<br />
[DllImport(&quot;MyTest.dll&quot;, CharSet = CharSet.Auto)]<br />
&nbsp; &nbsp; &nbsp; &nbsp;public static extern bool MYTEST([MarshalAs(Unmanage<wbr />dType.LPTS<wbr />tr)] string password, UInt32 ID, UInt32 Date, UInt16 Options, UInt32 ProductCode); <br />
<br />

</div>


<div>
&gt;&gt;&gt;&gt; Hmmm, &nbsp;when I posted the string to EE it changed it to ANSI.<br />
<br />
That would mean your output 'interprets' the ansi string as wide char string and outputs garbage that way. Your browser - on the contrary - was more clever and printed it correctly. <br />
<br />
Could you print out the hex code of each char in the return string. Unfortunately I have no idea ho to do that in C#. Maybe you could do that in the debugger somehow. <br />
<br />

</div>


<div>
&gt;&gt;Could you print out the hex code of each char in the return string. Unfortunately I have no idea ho to &gt;&gt;do that in C#. Maybe you could do that in the debugger somehow. <br />
<br />
I have attached some output from debugger, I am still looking into a hex conversion...<br />
<br />
<a href="https://filedb.experts-exchange.com/incoming/2009/02_w09/110459/junk.jpg" target="_blank" class="file-inline" title="Output from Debugger (16 KB)"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path d="M67.508 468.467c-58.005-58.013-58.016-151.92 0-209.943l225.011-225.04c44.643-44.645 117.279-44.645 161.92 0 44.743 44.749 44.753 117.186 0 161.944l-189.465 189.49c-31.41 31.413-82.518 31.412-113.926.001-31.479-31.482-31.49-82.453 0-113.944L311.51 110.491c4.687-4.687 12.286-4.687 16.972 0l16.967 16.971c4.685 4.686 4.685 12.283 0 16.969L184.983 304.917c-12.724 12.724-12.73 33.328 0 46.058 12.696 12.697 33.356 12.699 46.054-.001l189.465-189.489c25.987-25.989 25.994-68.06.001-94.056-25.931-25.934-68.119-25.932-94.049 0l-225.01 225.039c-39.249 39.252-39.258 102.795-.001 142.057 39.285 39.29 102.885 39.287 142.162-.028A739446.174 739446.174 0 0 1 439.497 238.49c4.686-4.687 12.282-4.684 16.969.004l16.967 16.971c4.685 4.686 4.689 12.279.004 16.965a755654.128 755654.128 0 0 0-195.881 195.996c-58.034 58.092-152.004 58.093-210.048.041z"/></svg>junk.jpg</a>
</div>


<div>
Something interesting, the Chinese password appears to be only half the correct length of the ANSI password. &nbsp;I hope that helps....
</div>


<div>
I think the problem is that the DLL is passing back a long pointer to a string of 1 byte chars (LPTStr). However, C# is converting that to UNICODE which is 2 bytes for each char. Thus, while the DLL has written an ANSI string of 1 byte per char, C# is using 2 bytes per char and thinks it is a Chinese string of chars. &nbsp;This would also explain why the Chinese password is only half the correct length.<br />
<br />
So, how do I get C# to receive the string as a LPTStr and not Unicode? &nbsp;I've tried many things but everything is converting the DLL string to UNICODE rather than ANSI.<br />
<br />
Ex.<br />
DLL returns: &nbsp;&quot;abcd&quot; &nbsp;4 ANSI chars (dec 97 98 99 100)<br />
C# interepets as &quot;??&quot; 2 chinese chars (dec 97 98 &nbsp;&amp; &nbsp;99 100 )<br />

</div>


<div>
In summary, the DLL is passing a password string LPTStr to my C# app. The C# app is treating every 2 bytes in the string as a single char, however, the DLL has created the string using 1 byte per char. Therefore, the C# string is half the correct length and interpreted as Chineses chars.<br />
<br />
So how can this be fixed without the DLL souce code? &nbsp;My VB6 app works fine with the DLL, so I'm sure C# can too.....somehow.....<br />
<br />
<br />
<br />

</div>


<div>
well this is what <b>itsmeandnobodyelse </b>said above...<br />
Moreover, C++ dll does is not passing LPTSTR(its a generic type), it is returning LPSTR. but as u said, u already tried that.<br />
Now to analyse this, can you put your code here which shows how you are calling this function &amp;&nbsp;where you are seeing this chinese characters?<br />
2nd, u told that it works with VB, VB normally deal with variants while dealing with DLL, so eventhough it is not related, just curious to see how you are declaring and calling this function in VB.<br />

</div>


<div>
It is as I assumed. The 19810 is hex 4D62 what are the ANSI chars 0x4D=77 and 0x62==98 what is 'M' and 'b'.<br />
<br />
So, the LPTSTR contains ANSI single-char string (as suggested in my firstpost) which unfortunately was not converted to wide string by the Marshaler. <br />
<br />
Are there bit operations possible in C#? If yes we could do the conversion by something like <br />
<br />
&nbsp; &nbsp;NewPassword[0] = ((Int)MyPassword[0])) &amp;&nbsp;0x00ff; &nbsp;// takes only lowest 8bits<br />
&nbsp; &nbsp;NewPassword[1] = (((Int)MyPassword[0])) &amp;&nbsp;0xff00) &gt;&gt;8; &nbsp;// high 8bits
</div>


<div>
Try the code in the code snippet to see if it shows the returned string:<br />

<pre><code class="code-1 nolinks" id="code-260583"><span>// Update to the password string returned</span>
<span>byte[] pw = System.Text.Encoding.ASCII.GetBytes(YourPWString);</span>
<span> </span>
<span>MessageBox.Show(System.Text.Encoding.ASCII.GetString(pw));</span>
</code></pre>
</div>


<div>
well, one more thing,<br />
try using, IntPtr instead of LPSTR and after getting that pointer convert it back to string as follows,<br />
&nbsp;public static extern bool MYTEST([MarshalAs(Unmanage<wbr />dType.IntP<wbr />tr)] IntPtr password, UInt32 ID, UInt32 Date, UInt16 Options, UInt32 ProductCode); <br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; IntPtr iPass;<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&lt; your call&gt;<br />
String strPassword = Marshal.PtrToStringAnsi(iP<wbr />ass); JUST TRY IT.<br />

</div>


<div>
Okay, problem solved! You are all amazing and I thank you for the support. Here is what finally worked:<br />
<br />
Declaration:<br />
-------------------<br />
[DllImport(&quot;MyTest.dll&quot;, CharSet = CharSet.Ansi)]<br />
public static extern bool MYTEST(StringBuilder password, UInt32 ID, UInt32 Date, UInt16 Options, UInt32 ProductCode);<br />
<br />
Call to function:<br />
-------------------<br />
StringBuilder sb = new StringBuilder(256);<br />
Class1.MYTEST(sb, ID, Date, Options, ProductCode);<br />
<br />
This is 99% what jkr recommended, the only diff is the inclusion of &nbsp;CharSet = CharSet.Ansi. So I am giving him the points.<br />

</div>


<div>
<div class="content wysiwyg-content">
I have a DLL written in C++ and I am trying to call it from C#.<br />
<br />
Here is my C++ DLL:<br />
--------------------------<wbr />---<br />
extern &quot;C&quot; __declspec(dllexport) BOOL WINAPI MYTEST(LPTSTR PW, DWORD BlockNumber, DWORD Date, WORD Options, DWORD ProductCode)<br />
<br />
Here is what I tried in C#:<br />
--------------------------<wbr />--------<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; UInt32 ID = 654321;<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; UInt32 ExpireDate = 20090228;<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; UInt16 Options = 0;<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; UInt32 ProductCode = 1;<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; String MyPassword = &quot;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&quot;;<br />
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Class1.MYTEST(MyPassword, ID, ExpireDate, Options, ProductCode);<br />
<br />
Where Class1.MYTEST is:<br />
--------------------------<wbr />----------<wbr />--<br />
&nbsp; &nbsp; &nbsp; &nbsp; [DllImport(&quot;MyTest.dll&quot;, CharSet = CharSet.Ansi)]<br />
&nbsp; &nbsp; &nbsp; &nbsp; public static extern bool MYTEST([MarshalAs(Unmanage<wbr />dType.LPTS<wbr />tr)] string password, UInt32 ID, UInt32 Date, UInt16 Options, UInt32 ProductCode);<br />
<br />
PROBLEM: MyPassword is being returned as Chinese characters but they should be ANSI. &nbsp;What did I do wrong?<br />

</div>
</div>


I have a DLL written in C++ and I am trying to call it from C#.

Here is my C++ DLL:
-----------------------------
extern "C" __declspec(dllexport) BOOL WINAPI MYTEST(LPTSTR PW, DWORD BlockNumber, DWORD Date, WORD Options, DWORD ProductCode)

Here is what I tried in C#:
----------------------------------
            UInt32 ID = 654321;
            UInt32 ExpireDate = 20090228;
            UInt16 Options = 0;
            UInt32 ProductCode = 1;
            String MyPassword = "                                                               ";
            Class1.MYTEST(MyPassword, ID, ExpireDate, Options, ProductCode);

Where Class1.MYTEST is:
--------------------------------------
        [DllImport("MyTest.dll", CharSet = CharSet.Ansi)]
        public static extern bool MYTEST([MarshalAs(UnmanagedType.LPTStr)] string password, UInt32 ID, UInt32 Date, UInt16 Options, UInt32 ProductCode);

PROBLEM: MyPassword is being returned as Chinese characters but they should be ANSI.  What did I do wrong?


How to call C++ LPTStr in C#

The .NET Framework is not specific to any one programming language; rather, it includes a library of functions that allows developers to rapidly build applications. Several supported languages include C#, VB.NET, C++ or ASP.NET.

.NET Programming

C# is an object-oriented programming language created in conjunction with Microsoft’s .NET framework. Compilation is usually done into the Microsoft Intermediate Language (MSIL), which is then JIT-compiled to native code (and cached) during execution in the Common Language Runtime (CLR).

C++ is an intermediate-level general-purpose programming language, not to be confused with C or C#. It was developed as a set of extensions to the C programming language to improve type-safety and add support for automatic resource management, object-orientation, generic programming, and exception handling, among other features.