Has anyone translated LMDFS

[Ronald Robinson]

Has anyone translated the LMDFS Header file?

or have a sample of using NetDfsGetInfo API call they would not mind sharing?

Thanks in Advance

[William Burns]

José Roca has translated that header and a lot of others in to powerbasic inlcude files. His translations are based on the C headers. He has done a great job of it. I just checked and he does have the LMDFS header converted and ready to use.

You can find those files here:

http://www.jose.it-berater.org/smfforum/index.php?board=344.0

Note: You must register and sign in before you will be able to download the files. But his site is loaded with good advice and information, so it is well worth it.

Hope that helps,

[Ronald Robinson]

Thank you and Jose' as well

[Ronald Robinson]

Here is the somewhat working test code. Only nagging question is why it fails with invalid paramter unless in the MultibytetoWideChar you include either
ACODE$(StrToUnicode(FolderPath))
or "\\DFSSERVER\DFSROOT\DIR" States, looking for a pointer if I do not include acode$ I get 1722 error invalid parameter.


Test code for NetDFSGetInfo:

'Includes Contributed by: Jose Roca
#INCLUDE "WIN32API.inc" '<-- JRoca includes
#INCLUDE "LM.inc"    '<-- NetAPIBufferFree declare
#INCLUDE "LMDFS.inc"  '<-- NetDFS include


'From Powerbasic Forum (www.powerbasic.com)
'Contributed by George Bleck 12/1/2001
FUNCTION UnicodePtrtoStr(BYVAL dwdConvert AS DWORD) AS STRING
   LOCAL lngLength AS LONG
   LOCAL strBuffer AS STRING
      lngLength = lstrlenw(BYVAL dwdConvert)
      strBuffer = SPACE$(lngLength)
         widechartomultibyte 0, %NULL, BYVAL dwdConvert, lngLength, BYVAL STRPTR(strBuffer), _
            LEN(strBuffer), BYVAL %NULL, BYVAL %NULL
   FUNCTION = strBuffer
END FUNCTION

'From Powerbasic Forum (www.powerbasic.com)
'Contributed by George Bleck 12/1/2001
FUNCTION StrToUnicode(BYVAL sConvert AS STRING) AS STRING
   LOCAL lngLength AS LONG
   LOCAL strBuffer AS STRING
      lngLength = LEN(sConvert)
      strBuffer = SPACE$(lngLength * 2)
         MultiByteToWideChar 0, %NULL, BYVAL STRPTR(sConvert), lngLength, BYVAL STRPTR(strbuffer), LEN(strBuffer)
   FUNCTION = strBuffer + CHR$(0,0)
END FUNCTION

FUNCTION PBMAIN () AS LONG
     LOCAL DFSData AS DFS_INFO_3 PTR
     LOCAL Result AS DWORD
     LOCAL FolderPath AS STRING, wstrEntryPath AS ASCIIZ * (2 * %MAX_PATH)
     LOCAL dwdResult AS DWORD
     Folderpath = "\\DFSServer\DFSROOT\DIR1" '<-- This must point to valid DFS server will not use drive letter mapping
    
     multibytetowidechar %CP_ACP, 0, ACODE$(StrToUnicode(FolderPath)), -1 , BYVAL VARPTR(wstrEntryPath), %MAX_PATH
     dwdResult = VARPTR(wstrEntryPath)

     Result = NetDFSGetInfo(dwdResult, %NULL, %NULL, 3, DFSData)

     IF result = 0 THEN
         MSGBOX  "\\" & UnicodePtrtoStr(@[email protected]) & _
            "\" & UnicodePtrtoStr(@[email protected])
         NetAPiBufferfree(DFSData)
     ELSE
         MSGBOX "Failure Reason: " & STR$(Result)
     END IF

END FUNCTION

[José Roca]

Folderpath = "\\DFSServer\DFSROOT\DIR1" '<-- This must point to valid DFS server will not use drive letter mapping

multibytetowidechar %CP_ACP, 0, ACODE$(StrToUnicode(FolderPath)), -1 , BYVAL VARPTR(wstrEntryPath), %MAX_PATH
dwdResult = VARPTR(wstrEntryPath)

Result = NetDFSGetInfo(dwdResult, %NULL, %NULL, 3, DFSData)

You are doing it too hard. Just use:

    Folderpath = UCODE$("\\DFSServer\DFSROOT\DIR1")
    Result = NetDFSGetInfo(STRPTR(FolderPath), %NULL, %NULL, 3, DFSData)

[Michael Mattias]

Only nagging question is why it fails .... if I do not include acode$ I get 1722 error invalid parameter

If you reference a PB "ASCIIZ" datatype (wStrEntryPath), it is assumed to terminate on the first $NUL. If your actual string is unicode, that means it will probably terminate on the second byte. Then when you convert that, you don't convert the whole string

You might have better luc with those unicode paths using something like

%UNICODE_BUFFER_SIZE =   %MAX_PATH * 2 + 2 
  
LOCAL wStrEntryPath   AS STRING*  %UNCODE_BUFFER_SIZE

Except that won't work either come to think of of because PB treats simple assignment to fixed string data type as though it were "LSET USING $SPC" , which is not what you want.... (why the compiler does that is a mystery to me. if I wanted LSET using $SPC, I'd LSET USING $SPC, ok? Please don't 'help' me like this!)

What you want is "TYPE SET wStrEntryPath = UCODE$(ansi_string) USING $NUL"

The compiler really could use a "WSTRINGZ * size" data type so we would not have to screw around dealing with the Unicode nulls ourselves.

MCM

[Ronald Robinson]

Thanks again Jose' and Michael for your insight

Jose' I went with your way much cleaner than the way I was doing it