xbarns
17th December 2008 13:18 UTC

nsExec Unicode Strange Characters returned
Hi all,

one of my programmers has provided me with a little tool to encrypt a password we use.

I use NSIS Unicode 2.38 and when executing

nsExec::ExecToStack '"$PLUGINSDIR\Crypt.exe" $R0'
Pop $0
Pop $1
DetailPrint "Crypt Password: $0 - $1"

in $1 i get "H#B777973B2483E148B95B1086E9D8B3DD3C52â–¡â–¡"

where â–¡â–¡ is what irritates me because that should not be there.

Any Ideas?

Anders
17th December 2008 14:15 UTC

what are those exactly, newlines? http://www.decodeunicode.org/u+25A1 ?

does Crypt.exe output unicode?

xbarns
17th December 2008 15:24 UTC

what are those exactly, newlines? http://www.decodeunicode.org/u+25A1 ?

xbarns
17th December 2008 15:47 UTC

OK, i saw that you can try the character in that webpage, but sadly i cannot paste it in, notepad cuts it off and puts the next character (put there on purpose) on the next line.

I have tried with UltraEdits Hex view and the Character is "0D0A" which represents a "carriage return line feed" :(

Anders
17th December 2008 17:49 UTC

its probably just the output of Crypt.exe, DetailPrint does not support newlines, can't you just strip it out with StrCpy ?

Afrow UK
17th December 2008 18:10 UTC

nsExec dumps all the output into one stack item.
If you use ExecDos::Exec /TOSTACK, it will output line by line to the stack.

Just Push /End before executing Crypt.exe and keep popping until you Pop /End.

Stu

xbarns
18th December 2008 08:24 UTC

It was the EXE itself that produced that CRLF, the guy programming it fixed it.

Thanks for your help :)