2016-11-23 12:14 GMT+01:00 [email protected] <[email protected]>: > Hi > > > In the past I have to build a few applications with C/C++ and I found #ifdef > directives everywhere to be ugly and confusing. For example, using ifdef > DEBUG to enable/disable debug messages.
depends on the use case (you can't do everything without #ifdef), but full ack on debug message. > > That's why I opt to use functions which behavior was defined depending on > compile headers/source files. For example, define log_debug(...) as NOOP or > sleep() depending on platform. > > Have you considered doing something like this? yup, the function is named DBGPRINTF() and some helpers. Being used since ~1970 (the original dbgprintf() call stems back to the original code ;)). Sometimes we use #ifdef DEBUG, but only if we actually need to compute something beyond simple output generation (e.g. iterate over a structure). This can't be done efficiently without #ifdef. > > > Also, some time ago I started playing with "component" debugging, which > enabled debug only for specific components. For example, setting > RSYSLOG_DEBUG_MODULES_C will enable debug just for modules_c file, but it > will be very verbosed > > Have you considered doing something like this? similar, a verbosity level, and partial support is available via the RSYSLOG_DEBUG variable (you can turn on/off some debugging features). But it's still in its infancy. Would really be good to have. Rainer _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
