Gennadiy Rozental wrote:
What about this:
time_t slow_time() { BOOST_LOG(x) << "going to sleep at " << time(0); sleep(1); time_t t; BOOST_LOG(x) << "waking up at " << t = time(0); return t; }
void f() { BOOST_LOG(x) << "slow_time returned " << slow_time() << " then " << slow_time(); }
Resulting in (I ran this test a long time ago, boot time isn't bad either :-) :
going to sleep at 0 waking up at 1 going to sleep at 1 waking up at 2 slow time returned 1 then 2
I have several questions about this:
1. How do you know in general which log present trace of first invocation, and which one second:
void acess_db( ... ) { if( some error ) log << "invalid account key" << key; }
int get_balance(...){ ... if( !access_db(...) ) return 0; }
...
log << "Mary has balance: " << get_balance() << " and John has balance: " << get_balance()
I would assume you meant something like: log << "Mary has balance: " << get_balance("mary") << " and John has balance: " << get_balance("john"); If so, when you have a bad key, you should have some extra information to log (the name, or something).
2. How does the log system know when to dump the cash? Does it
introduce any scope guards? Have you actually taken a look at the code? A temporary variable is created, and in its destructor I perform the logging.
4. What kind of overhead it introduces? Now instead of directly
stream we keep second copy of every piece of log essentially doubling
writing into the
memory usage and increasing log overhead, isn't it?
Have you actually tested it in a multi-threading environment, or is this just your guts? Best, John -- John Torjo, Contributing editor, C/C++ Users Journal -- "Win32 GUI Generics" -- generics & GUI do mix, after all -- http://www.torjo.com/win32gui/ -v1.6.3 (Resource Splitter) -- http://www.torjo.com/cb/ - Click, Build, Run!