From: Junio C Hamano Date: Thu, 24 Feb 2011 20:50:57 GMT Subject: Re: [PATCH 4/8] trace: refactor to support multiple env variables Message-ID: <7v39nd17im.fsf@alter.siamese.dyndns.org> In-Reply-To: <20110224194828.GA5616@sigill.intra.peff.net> Jeff King writes: >> > I think I covered that pretty well above, but you lose the flexibility >> > of pushing different trace types to different places if you want to. >> >> That statement illustrates the inflexibility that all packet traces going >> to a same place rather clearly, doesn't it, though? That "two-tier" thing >> was my primary concern. > > Sorry, I don't quite understand what you're saying here. I was contrasting these two and called that "two-tier thing": - GIT_TRACE_SETUP and GIT_TRACE_PACKET can be used to give you flexibility of sending their output to the same location or to different destination. Nice and flexibile. - Even though the trace identities like "clone", "fetch-pack", etc. are clearly marked in the code, you cannot take advantage of the marking and send the output from these packet sources to different destination, because GIT_TRACE_PACKET is just one and single variable. Clearer? Not that I think the same flexibility in the latter is absolutely necessary, but it somewhat feels inconsistent to have these two classes.