Am 01.01.2011 23:22, schrieb Robert Clipsham:
Having seen a post by Peter Alexander (in Re: D for game development),
mentioning some of the issues he's hit I thought I'd post this. I've
been in his shoes (every other time I use D it seems), and feel I should
ask - who here uses D, and to what extent?
I'm mostly interested in those of you with 1000 Line plus projects in D,
as that's when I've found I start hitting issues.
Just to clarify, for those D purists among you... I'm not trolling, just
curious (I wouldn't normally have asked, but now I know I'm not paranoid
and the only one having problems, I thought I'd ask).
My main project is abount 100.000 lines of code. Most of the time now, I
stay away from new features - including pure and the new concurrency
model. However, I use the const/immutable system since the beginning.
Also quite some string-mixin stuff, templates and CTFE. (*)
At times I'm hitting multiple bugs a day, but by far the most
devastating bugs are those freakin' unexpected OPTLINK terminations
(e.g. http://d.puremagic.com/issues/show_bug.cgi?id=4808). With that
code base it is practically impossible to make repro cases or
workarounds for those bugs and they slip up every now and then (right
now I have one of them). Unfortunately those are often bugs that lie
there for years. I remember someone said there should be no known new
regressions in the compiler - the reality seems to be quite different
here and I quit D programming for multiple periods of months because of
such beasts. (Fortunately most of the time there is one method of
compilation or operating system that successfully builds).
The other kind of bug that I find really frustrating because it is hard
to discover and takes a lot of time to track down and work around is
that kind with corruped data/wrong code. Todays examples are
http://d.puremagic.com/issues/show_bug.cgi?id=3863 and some postblit
stuff that I have not yet been able to track down.
At least in times when it is possible to program without hitting those
issues, D somehow is able to close the gap again with its nice and
efficient language constructs. But I think the priority for fixing bugs
really has to be changed because that is what is driving people away
(for good reason):
Blockers, Regressions and maybe criticals should be taken more serious,
as well as the top votes in bugzilla should be handled somehow.
Sönke
(*) I think the fact that only very few people use new features and the
rest is mostly just doing smaller tests with them is a real problem in
the language/compiler development. All of those people think "I will use
it when it's ready", but it will never or really slowly reach that stage
beause of missing input. This also means first use feedback for new
features should be taken more serious - I've often seen important
observations vanish in time and meanwhile the underlying problem was
consolidated in the language.
