On Mon, Mar 10, 2008 at 7:43 AM, Jonathan Mast <[EMAIL PROTECTED]> wrote:
> What benefit does flushing have? Would it help prevent memory > leaks as per the above scenario or what? Flushing gets the data out right away, in case there's anybody waiting for it. In the case of log files, it means that the person debugging the long-running program gets to see the up-to-date log. It also means that a program that suffers a serious fault (such as running out of memory, or an uncaught signal) won't be so likely to die with the last few lines of the log still unwritten -- and those last few lines are probably trying to tell you something important, since the program just crashed. As for memory leaks, if we learned that technique X were necessary to prevent memory leaks in Perl release Y, we would track down the leak so that technique X is no longer needed in the next release. The exception is if Perl's architecture prevents us from stopping the leak; then we document it. The primary cause of memory leaks in Perl code is circular references, in which a programmer has (intentionally or not) built a ring-shaped data structure out of (non-weak) references. It is important to break all such rings when data structures are to be discarded, else their memory can't be reclaimed before the program quits. This is rarely a problem for most programmers, though, since ring-shaped data structures are hardly ever needed in Perl; hashes and arrays are more convenient ways to work with most data. Cheers! --Tom Phoenix Stonehenge Perl Training -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] http://learn.perl.org/
