Re: Problem with backslash/newline on DOS/Windows/OS2/etc.?

[Paul D. Smith]

%% "Earnie Boyd" <address@hidden> writes:

  >> The "prior code" is gone; I rewrote that entire section (and it needed
  >> it, too!)  We'll have to reconstitute the correct behavior, which
  >> shouldn't be that difficult to do, actually.  This code is easier to
  >> understand and work with than it was before.

  eb> Is there some macro that identifies a POSIX compliant system?  It
  eb> would be a matter of concantenation before the process spawn.

Well, if it's a macro that means it's a compile-time thing (I assume you
mean with ifdef).

In an ideal world this wouldn't be a compile-time thing: it would be a
runtime thing.  Consider that on Windows you can, and many people do,
have a full-fledged UNIX shell that make invokes.  Any shell like that
will properly handle backslash-newline issues, and so make could behave
"properly" according to POSIX on those systems.

Only when the shell make invokes is not capable of dealing with this
properly would we need to revert back to the old behavior.


I don't know the details, but I believe that's what the unixy_shell
variable is for (see job.c): it's true if the thing make invokes is a
UNIX-y shell, and false otherwise.

So, assuming I'm correct about this, it seems like conditionalizing the
code on that value would be The Right Thing.

Yes?  No?  I need guidance on this from the DOS/Windows/OS2 folks.

-- 
-------------------------------------------------------------------------------
 Paul D. Smith <address@hidden>          Find some GNU make tips at:
 http://www.gnu.org                      http://make.paulandlesley.org
 "Please remain calm...I may be mad, but I am a professional." --Mad Scientist