First, Thanks for the thoughtful responses. I'm beginnning to understand how the Ant logging mechanism works now :)
Second, I agree 1.7 cannot be changed at this late date to fit in a logging cludge! Third, I came across the problem with a particularly long-running build, and it was less of a CPU load problem and more of a max JVM mem limit problem. I understand some builds may be the other way around and introducing guard statements would add CPU overhead, whereas the memory isn't so tightly constrained. This is why I wanted to spark a discussion, to see if there is a way we can get Ant to only do the necessary work *when* it's necessary. In the example/article D uses lazy or partial evaluation to sort of get around this. In other languages there are other mechanisms. In Java perhaps there's a way of refactoring the logging such that Strings are re-used (use static final String and build logging from them?), or a way of getting the building code to short-circuit before constructing the string (as in D). I'd like to put logging enhancements/improvements up there with distributing a default scripting language for 1.8 timeframe Yes, let's improve logging (if possible) to reduce memory/cpu load for 1.8 [ ] No, it's not a big problem (and besides your use-case is weird!) [ ] Thanks, Kev --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
