Ok, I respectfully remove my proposal from the offering table. I think one of the main reasons that I made these changes is because I didn't like the fact that I wasn't getting what I asked for. i.e.: I asked you to include all these files, and you didn't.
I think that I will now focus my energies in getting some of the documentation to actually capture some of the functionality of the task. (e.g.: task <javac> actually does some time-stamp checking ...) Thanks for your feedback. Jay -----Original Message----- From: Stefan Bodewig [mailto:[EMAIL PROTECTED] Sent: Wednesday, January 31, 2001 3:43 AM To: [EMAIL PROTECTED] Subject: Re: [Patch] addition of an "include level" to the matching task & jav ac task Jay Glanville <[EMAIL PROTECTED]> wrote: > Why? On a ClearCase file system, file access is slightly slower > then local hard drive access (due to networks, NFS mounts, > repository server loads, etc). Therefore, accessing two different > files in different directories for time stamps can take a while. And this would be worse than always compiling everything? If you remove the target directory upfront (i.e. clean out all old classes) does it improve the situation? > How do I propose to do this? I have added an attribute to the > MatchingTask task called "includelevel". I don't think includelevel is a good name for this - and I don't think MatchingTask is the right place either. If anywhere it belongs to <javac> directly - unless we find it would be useful for other tasks as well - in which case we need to search for a better place. What you want is similar to the overwrite attribute in <copy> which does not inherit from MatchingTask for example. Stefan --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
