Review: std.logger

[Jeremy Powers via Digitalmars-d]

Haven't had a chance to look closely at the new version yet, but looks
pretty good.  Couple initial comments:

* Definite vote for std.experimental.  Should get a bunch of folks to bang
on it before the API has to be locked down.  Having it as a dub package
first has made it much easier to pick up and use, feel like this is a good
path for all std packages to take - get it out there and iterate with user
input before pulling in to std.

* The 'Tracer' doesn't feel like it belongs here, and if I understand
things correctly won't actually work properly anyway, so should probably be
removed.

* What others have said about string formatting.  I poked around with this
a bit before, but wasn't able to get a clean solution, should probably try
again before complaining...





On Fri, Jul 11, 2014 at 9:27 AM, H. S. Teoh via Digitalmars-d <
digitalmars-d at puremagic.com> wrote:

> On Fri, Jul 11, 2014 at 02:59:43PM +0000, Dicebot via Digitalmars-d wrote:
> > On Friday, 11 July 2014 at 14:39:09 UTC, David Nadlinger wrote:
> > >On Friday, 11 July 2014 at 14:36:34 UTC, Dicebot wrote:
> > >>Round of a formal review before proceeding to voting. Subject for
> > >>Phobos inclusion : http://wiki.dlang.org/Review/std.logger authored
> > >>by Robert Schadek.
> > >
> > >Is this for std.* or std.experimental.*?
> > >
> > >David
> >
> > Deciding this is subject of this review/voting iteration too - it is
> > mostly matter of API stability, how much of a trust reviewers are
> > ready to put into existing API.
> >
> > Personally I believe that for something like logging library
> > stabilization period of one release cycle in std.experimental is
> > desirable because wider usage is very likely to result in breaking
> > change suggestions.
>
> I vote for std.experimental. We keep talking about it, but never do
> anything in that direction. Let's start. If it works out poorly, we can
> always scrap the idea later. But we'll never know if we never do it.
>
> (In contrast, putting it directly in std risks the necessity of breaking
> changes later, which is a Bad Thing. Putting it in std.experimental now
> does no harm whatsoever -- the worst that can happen is that it's
> delayed entering std. The best is that breaking changes will not annoy
> users. So we have nothing to lose.)
>
>
> T
>
> --
> "Computer Science is no more about computers than astronomy is about
> telescopes." -- E.W. Dijkstra
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.puremagic.com/pipermail/digitalmars-d/attachments/20140711/7ac0facb/attachment.html>