DJ Delorie wrote:
I'm not sure how I can "fix MinGW"; see above. Also, if a MinGW application wants to invoke some other Windows program, the behavior should be the same as if I compiled that application with Visual C, or Intel's C compiler, or whatever; if we were using magic to pass command-line arguments, we'd be breaking things.
My suggestion is to take the patch you are proposing, and add it instead to MinGW's crt0 code. That way, *ALL* applications built with MinGW would support @file on the command line, not just gcc.
But, that would change the behavior for applications that may not want that. That's what I was trying to communicate above. One of the benefits of MinGW is that you can write code that compiles and behaves identically with MSVC, ICC, or GCC. If we did as you suggest, we would violate that invariant.
A similar argument would apply to the shell-script stuff that was just added to libiberty. Why not just have MinGW wrap CreateProcess and do shebang handling there? Because only some applications what that functionality.
I agree that it would be technically feasible to create a different version of MinGW crt0 code that users could optionally select when they want @-file behavior in an application. (Of course, we could do that for shebang handling too.)
However, there's demonstrable interest in this feature for GNU/Linux as well, from the lists, and for Java on all operating systems.
-- Mark Mitchell CodeSourcery, LLC [EMAIL PROTECTED] (916) 791-8304
