ant-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Evenson <>
Subject Re: [BUG] emacs cygwin compile.el next-error fails with Ant
Date Thu, 24 Jul 2003 14:18:10 GMT
Conor MacNeill <> writes:

> On Thu, 24 Jul 2003 10:04 pm, Mark Evenson wrote:
> > After some investigation and further reflection, I think this can be
> > classified as a problem with Ant, and not with cygwin and/or Emacs as Ant
> > should emit "UNIX normalized" pathnames if it is running under
> > cygwin/Emacs.  
> Ant is not running under cygwin. It is running in a Win32 JVM so it is running 
> under Windows.

I agree that the ant Java process is within a win32 JVM, but The #!/bin/sh
script named 'ant' definately is running under cygwin: it checks for the
presence of cygwin in the ouput of uname(1), invoking the 'cygpath' utility
to normalize paths.  

So what do we mean by "under"?  Since win32 lacks the UNIX parent/child
process relationship, perhaps it ain't entirely clear from a semantic
relationship.  What I meant by "under" was if thread group A (cygwin) spawns
thread B (emacs) which spawns thread group C (java VM running ant), C is
running under A.  Don't you think this is an appropiate use of "under"?

As Antonine points out, normalizing the output of javac within Ant will
not be particulaly easy, so I wrapped the invocation of ant in suitable
a gawk/sed script that will allow me to function for the time being.  Ant
is definately running "under" cygwin enough for cygwin to modify its

Mark Evenson <>

"A screaming comes across the sky.  It has happened before, but there is
nothing to compare to it now."

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message