Getting username under windows

Hi List, I'am trying to get the the current username under windows. After a little searching I found the the following api call: BOOL GetUserName( LPTSTR lpBuffer, LPDWORD lpnSize ); In REBOL I tried the following: lib: load/library %advapi32.dll ; no error user: make routine! [arg1 [string!] agr2 [integer!] return: [integer!]] lib "GetUserNameA" ; no error name: make string! 1000 buffersize: 1000 user name buffersize ; crashes rebol After hitting <enter> rebol crashes. I tried different values for name and buffersize. I tried arg2 as string! type. Rebol then stops crashing but no name or anything else is returnt. Can someone give me a hint on what I 'am doing wrong? Or maybe is there some other way to get the current user who is logged in? Thanks Tim I have included the official MS info: Parameters lpBuffer [out] A pointer to the buffer to receive the user's logon name. If this buffer is not large enough to contain the entire user name, the function fails. A buffer size of (UNLEN + 1) characters will hold the maximum length user name including the terminating null character. UNLEN is defined in Lmcons.h. lpnSize [in, out] On input, this variable specifies the size of the lpBuffer buffer, in TCHARs. On output, the variable receives the number of TCHARs copied to the buffer, including the terminating null character. If lpBuffer is too small, the function fails and GetLastError returns ERROR_INSUFFICIENT_BUFFER. This parameter receives the required buffer size, including the terminating null character. If this parameter is greater than 32767, the function fails and GetLastError returns ERROR_INSUFFICIENT_BUFFER. Return Value If the function succeeds, the return value is a nonzero value, and the variable pointed to by lpnSize contains the number of TCHARs copied to the buffer specified by lpBuffer, including the terminating null character.

> I'am trying to get the the current username under windows.

here is full working code... I have this library in use in apps and its fully functional: the first line was to implement different implementations based on os, but I never got to it, by lack of need so far. hope this helps! -MAx ;----------------------- rebol [] isWin?: (system/version/4 = 3) ;- OS LIBS libsysinfo*: context [ libinfo: make library! switch system/version/4 [ ; other OSes will have to be filled in by other REBOLERS, as I have absolutely no idea. 3 [%Advapi32] ] username: make routine! [ "Get the current logon name of thread which launched REBOL" buffer [char*] count [string!] return: [integer!] ] libinfo "GetUserNameA" ] ;- ;- REBOL ACCESSORS ;- username username: func [/local buffer count][ buffer: head insert/dup copy "" #"^-" 100 ; this should cover any username! ;buffer: copy "..............................." count: third make struct! [val [integer!]] [100] unless isWin? [to-error "username only implemented for windows at this point"] either 0 < libsysinfo*/username buffer count [ return copy/part buffer (to-integer reverse count) - 1 ][ to-error "no username or invalid username" ] none ] probe username ask "..." ;-------------------------------- On 3/18/07, th72-dds.nl <th72-dds.nl> wrote:

hi, its also a huge security hole, and is used much to often by many (non rebolers). -MAx On 3/18/07, Christopher Ross-Gill <chris-ross-gill.com> wrote:

> its also a huge security hole, and is used much to often by many (non > rebolers).

Chris I think that MAx is pointing out that it is very easy for somebody to fake the user name using the environment variable. Regards Peter

Hi Tim, On Sunday, March 18, 2007, 10:49:15 PM, you wrote: tdn> user: make routine! [arg1 [string!] agr2 [integer!] return: tdn> [integer!]] lib "GetUserNameA" ; no error The problem is that the second argument should be a pointer to a integer, not an integer. Just create a struct with an integer in it and pass it to the function. Regards, Gabriele. -- Gabriele Santilli <gabriele-rebol.com> --- http://www.rebol.com/ Colella Chiara software division --- http://www.colellachiara.com/

Hi all, Thanks for the positive feedback from you all. This is great! Tonight I'am going to try all your suggestions. Thanks! Tim Quoting th72-dds.nl:

peter, exactly, its even worse, when techie users know they can hide this within a script. in medium-large shops, sometimes its nice to automate the user procedure, not for explicit login purposes, but for simple tracking of who did what. in such circumstances, these often become logs and unfortunately they get used for many purposes... including HR and other depts. if/when any one in the shop gets to understand that an env var allows him to ghost his user or act on someone else's behalf... you've got very touchy consequences... and if its ever understood that the software allowed someone to get wrongfully accused of something (or worse)... well that person (and admins) can look to the author, for the focus of his anger (for allowing this to happen). obviously depends on work environment, but any nice environment can become political nightmare in an instant, with but the change of one manager in the corporate ladder. -MAx On 3/18/07, Peter Wood <pwawood-gmail.com> wrote:

Just getting back on the ML after being unsubscribed (it seems). Wil; this work for you? --Gregg REBOL [ file: %WNetGetUser.r Author: "Gregg Irwin" EMail: greggirwin-acm.org ] win-lib: load/library %mpr.dll integer-struct: make struct! [ value [integer!] ] none null-buff: func [ {Returns a null-filled string buffer of the specified length.} len [integer!] ][ head insert/dup make string! len #"^-" len ] WNetGetUser: make routine! compose/deep [ lpName [integer!] ; change to string if you need to use device name lpUserName [string!] lpLength [struct! [(first integer-struct)]] return: [integer!] ] win-lib "WNetGetUserA" rtn-name: null-buff 256 ; If you wanted to find out exactly how much space you need first, ; you can all with a length of 0 and it will return the amount ; of space you need in the lpLength parameter. len: make struct! integer-struct reduce [0] print ["res: " res: WNetGetUser 0 rtn-name len] print ["len: " len/value] print "" len: make struct! integer-struct reduce [length? rtn-name] print ["res: " res: WNetGetUser 0 rtn-name len] print ["name: " rtn-name] print "" free win-lib halt

Hi Greg and others, Yes this wil work! I try to understand it. Never heard of integer-struct. What does this mean: lpLength [struct! [(first integer-struct)]] Sometimes Rebol leaves me in great mysteries :-) Tim Quoting Gregg Irwin <gregg.irwin-gmail.com>:

Hi List, get-env "USERNAME" also works here. Yet another rebol-word I learned today. For my purpose this seems good enough. The external library interface is still a little bit of a mystery to me. I need some more practice on it. Thanks for al the help. Tim Quoting th72-dds.nl:

Hi Tim, tdn> Yes this wil work! I try to understand it. Never heard of integer-struct. It's just a struct with one int element in it. tdn> What does this mean: tdn> lpLength [struct! [(first integer-struct)]] Using structs with routines is a bit of a pain. What this does is define a struct as the param for the routine, and you need to provide the struct spec when you do that, which is what FIRST does.

>> integer-struct: make struct! [val [integer!]] none >> first integer-struct

hi tim, btw, on my machine (WinXP) for some reason, the username var was not setup correctly, when I first tried using the get-env trick. maybe some software install or whatever.. so Tim, be carefull that you are sure all your users have the username env when you do the install on other people's machines. -MAx On 3/19/07, Gregg Irwin <gregg.irwin-gmail.com> wrote:

To explain what the following means...

> integer-struct: make struct! [ > value [integer!] > ] none > WNetGetUser: make routine! compose/deep [ > lpLength [struct! [(first integer-struct)]] > ] win-lib "WNetGetUserA"

Hi Maxim, Thanks for the "warning". I tested it also on a XP machine and it's working fine here. I'll keep it in mind. Tim Quoting Maxim Olivier-Adlhoch <moliad-gmail.com>:

Hi Anton and others, Thanks for the explanation. I tried it to understand it in the same way. The problem was that I could't find a explanation of doing: first integer-struct. I now understand that this is simply some lazy (and smart) way to reduce some typing. Thanks Tim Quoting Anton Rolls <anton-wilddsl.net.au>: