RE: Secure popen

From: Jeff Dafoe (jeffdat_private)
Date: Thu Jun 21 2001 - 08:55:58 PDT

  • Next message: John Viega: "Re: Secure popen"

    > The real issue is a) above. Scripting languages, by their nature, make
    > heavy use of "in-band signalling", where language constructs are
    > embedded within data. Unless care is taken to prevent the problem, the
    > risk is that user-supplied "data" can effectively become user-supplied
    > code.
    
    	I can't figure out what you are talking about, in what scripting language
    does user-supplied data become user-supplied code without the programmer
    explicitly calling some function to execute the data as code?  You seem to
    think that substitutions triggered by metacharacters are triggered on
    user-inputted data as well.  I can't think of any commonly used scripting
    language that will expand user-inputted metacharacters or execute user input
    as if it were code without the programmer explicitly doing such.
    	I have seen a lot of recent bugs that arose from the use of C format
    strings, though, which readily expanded user inputted macros.
    
    
    Jeff
    



    This archive was generated by hypermail 2b30 : Thu Jun 21 2001 - 20:34:41 PDT