[hxdatatype] Adding core debug capability to audrend
Eric Hyche ehyche at real.comAt 04:56 PM 7/17/2003 -0700, Aaron Colwell wrote: >I was wondering if we could settle on 1 mechanism for debug print >statements. I tend to be partial to DPRINTF() myself, but I've noticed >that Eric seems to use MLOG_XX for debug prints. I have also seen a few >other debug print construct floating around the code as well. Can we >please just pick 1. > >I vote for DPRINTF() since it is basically like using printf() and allows >you to classify debug prints so that you can only turn on the debug prints >that you want. The only thing that I think should be changed from the >current implementation is that we make the debug mask a class instead of a >32 bit value. Doing this will allow us to have more than 32 different >debug print levels. I agree that the masking feature of DPRINTF is cool. But the limitation of DPRINTF is that at its crunchy center, it's a fprintf to stderr. The whole reason for MLOG_xxx and common/dbgtool/multilog.cpp is that you want to write one logging statement that may go to 1) stderr/stdout; 2) debugger window (OutputDebugString() on Windows); 3) log file; 4) CoreDebug player window; 5) logged by the logging plugin (sold separately). And, you might want to be able to turn these logging statements on or off at runtime with a regkey. multilog does all of these things. Those are the nice things about multilog. The not-so-nice thing about multilog is you have to define specialized MLOG_xxx macros in the .cpp file or in a new .h file (this is essentially equivalent to the masking feature). Managing these macros is sometimes a pain. Perhaps the best solution would be if we added the nice things about DPRINTF to multilog (masks), and then standardized on that. Eric >Aaron > >On Wed, 16 Jul 2003, Eric Hyche wrote: > > > > > dev at datatype: > > > > This diff (and additional added file) adds core debug capability to > audrend. > > > > I will check this into HEAD and KH at EOD today unless > > there are objections. > > > > Eric > > > > > > ====================================== > > M. Eric Hyche (ehyche at real.com) > > Core Technologies > > RealNetworks, Inc. ====================================== M. Eric Hyche (ehyche at real.com) Core Technologies RealNetworks, Inc. --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscribe at datatype.helixcommunity.org For additional commands, e-mail: dev-help at datatype.helixcommunity.org