ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Donald <>
Subject Re: Order of attributes versus elements
Date Sun, 03 Mar 2002 04:19:23 GMT
On Sun, 3 Mar 2002 12:02, Conor MacNeill wrote:
> Peter Donald wrote:
> > On Sat, 2 Mar 2002 21:05, Conor MacNeill wrote:
> >>>>Should this
> >>>>be specified in Ant1? How about Ant2? Alternatively, should it be left
> >>>>undefined? In that case Build files which rely on any particular order
> >>>>would be considered defective even if they happened to work
> >>>>
> >>>>Anyway, IMHO, I think this buildfile is broken.
> >>>
> >>>I wouldn't say that. I would say the task is broken. BTW I think you
> >>> will find that sort of construct in a few places (myrmidons build file
> >>> for one and it also used to be in ants build file).
> >>
> >>The construct is not broken and neither is the task. I'm happy to have
> >><src> elements and srcDir attributes at the same time. What is broken is
> >>having two Java files for one class in the set of paths given to javac
> >>and expecting one of those paths to take precedence. The use of
> >><exclude> elements is an indicator that this is a bit of a hack.
> >
> > So let me get this correct. You are saying that it is a **hack** that if
> > you specify java files in compile list and then the compiler compiles
> > these files rather than those that happen to be on it search path?
> NO. Reread what I said.


> Have a close look at the construct. It explicitly ***excludes*** the
> files in question. They are **not** in the compile list.

They are included by <src/> element?

> In fact if they
> were in the compile list it wouldn't work because javac would be given
> two Java files for the same class. Don't you think it is odd that even
> though the build just went to some trouble to filter these files it
> specifically excludes them from compilation? 

no seems fine to me.



| "Common sense is the collection of prejudices        |
|  acquired by age 18. " -Albert Einstein              |

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

View raw message