Re: [boost] Egg 2nd request for reviews

From: Daniel Walker (daniel.j.walker_at_[hidden])
Date: 2008-04-12 12:40:05

On Fri, Apr 11, 2008 at 3:48 PM, dan marsden <danmarsden_at_[hidden]> wrote:
> ><gpderetta_at_[hidden]> wrote:
> >> If there is a possibility to extend the review period of another week,
> >> I'll take sometime to try the library, possibly with different
> >> compilers. This may also let other boosters have time write their own
> >> review (even if strangely so far no one else seems to have shown
> >> interest).
> >>
> >> If you think that the review should still end the 13th, I'll wrap up a
> >> review for today.
> >
> >Ditto. I have an interest in the problem domain Egg addresses, and
> >from what I've read, I like Shunsuke's approach. But I'm just now
> >sitting down to take a real look at it. I can wrap up a review today,
> >but if you don't mind extending the review period, I think the
> >discussion could benefit from more time.
> >
> >Thanks,
> >Daniel Walker
>
> I do intend to extend the review period by a week. I appreciate that Egg
> is a big library, and it takes a bit of time to review fully. So no need
> to rush a review, but please do put them in before the end of next week.
> Currently the review count sadly still sits at zero, so your reviews will
> be gladly received thanks!

Unfortunately, I believe I'm going to have to renege on my review.
After spending some time with Egg, I see that its scope goes way
beyond my initial interest. I'm afraid I can't do it justice in a
review at this time. However, I will share some of my first
impressions.

My initial interests in Egg stems from my experience last year looking
into the whole lambda/result_of compatibility issue (as some of you
may remember since you gave me great feedback and advice!). It was
fairly simple to modify lambda to work with TR1 result_of. (I think my
patch may have gotten lost in the SVN migration.) I also tried to
bring lambda completely up to date by making it play nicely with TR1
bind and placeholders as well. But the project died under feature
bloat; bind and placeholders were too much for me, or at least, it got
too tedious to keep my interest. I think it doesn't really matter,
though. Given the most recent draft of C++0x (n2588), there will no
longer be result_of/lambda compatibility problems as std::result_of
will always work by fiat - even with lambda-style functors. Actually,
things are progressing so quickly with C++0x that Boost.Lambda itself
may be made obsolete by the new lambda language extension (n2550),
which apparently was adopted sometime in the last six weeks. I think
that's great news! But I'm not sure what it means for Egg.

Egg seems to have a lot of valuable components that may fill a niche
in the existing paradigm for functional programing in C++, but the
shape of that niche is changing as more classic boost techniques are
transmuted into the standard and newer boost libraries come online. So
there's an unfixed boundary between advancing the cause of functional
programing and reinventing existing primitives. For example, in C++03
(and C++0x) there's a need for more flexible/powerful function
parameters. C++0x variadic functions provides for a variable number of
type-safe parameters, but what if a user wants to change their order
or name them? In other words, in either standard, when users need more
advanced function parameters, how does Egg relate to Boost.Parameter?
Another example: there's a long standing practice of currying
functions with bind, which will be standardized in C++0x. How does
Egg's curring "function adaptor" relate to std::bind? (These are just
some questions that jumped out at me on my first impression. I haven't
spent a lot of time trying to reason through the documentation.)

Still, I think it is a noble ambition to create the equivalent of
Boost.Iterator for functions - i.e. to do for std::unary_function and
std::binary_function what Boost.Iterator did for std::iterator. Surely
Egg represents progress toward that goal. Perhaps if Egg were
repackaged/refocused to become a more comprehensive Boost.Function (a
true counterpart to Boost.Iterator: intuitive and accessible), it
would draw wider interest. I'm not sure though, because by nature
functors may be harder to package/sell than iterators.

FWIW, I did build and run Egg's test suite using gcc 4.3 with both
-std=c++98 and -std=c++0x. The test suite is large and seems to have
good coverage. It ran with only one apparently minor glitch, which is
testimony to the quality of Egg's implementation. Egg is certainly an
accomplishment. Shunsuke, later this weekend I'll try to send you the
test suite output and a few other comments off-list.

Daniel Walker