Hello,
> the
> (statistic) result was that reusing safes more than half of the time (related to
> starting a new one).
Sorry, this was not correct. Due to an failure of mine, constructors and reused
watchers did not run under exactly the same conditions (different io watcher
timeouts). So, the new result is only a saving of about 1/4 of time by reusage in a
100 watcher test, and of almost nothing if the number is increased to 1000 a loop.
Maybe my test script is not optimal, but it seems to me that the advantage of
watcher reusage strongly depends.
I apalogize for this part of the thread ...
Jochen
