- From: David Lee <dlee@calldei.com>
- Date: Wed, 3 Sep 2014 03:14:58 +0000
- To: Norman Walsh <ndw@nwalsh.com>
- CC: XProc Dev <xproc-dev@w3.org>
By coincidence I have this weekend replaced xmlsh''s logging from log4j1 to log4j2 I did it brutally fast and now reading up on why the easy part was hard and the hard part easy, RTFM? Naw .... Just twist the code until it's inconceivable obtuse ... It's got to fit. Then RTFM and <facepalm> I have to say it's an elegant and finely tuned piece of art/engineering ,,, In the rare style of actually learning from a decade of good solid work and improving on it. Sent from my iPad (excuse the terseness) David A Lee dlee@calldei.com > On Sep 2, 2014, at 8:53 PM, "Norman Walsh" <ndw@nwalsh.com> wrote: > > Hello world, > > I find the subject of logging in Java to be extraordinarly > complicated. I've been trying to work through some weirdness in the > logging behavior of XML Calabash. > > Complicated or not, there are clearly some benefits to "doing it > right." In particular, I've removed all those .printStackTrace() calls > and things are moving towards having reasonable info/debug/trace level > messages. > > FWIW, I've settled, perhaps mostly by accident, on using SLF4J as a > front end on LOG4J2 with commons-logging-1.1.1 thrown in for good > measure because the Apache HTTP libraries seem to require it. > > I've poked at the XML Resolver library code to use those loggers as > well. > > If you're curious to see the effects, there's a log-hacking branch > checked into github now (it's going to be transitory, but it'll be > there at least until I can figure out why Travis CI can't get a > passing build with it). > > Comments and suggestions most welcome. > > Be seeing you, > norm > > -- > Norman Walsh > Lead Engineer > MarkLogic Corporation > Phone: +1 512 761 6676 > www.marklogic.com
Received on Wednesday, 3 September 2014 03:15:57 UTC