We'll still use the major-versioned name file for d3js.org, but this is simpler.

Remove d3.geo.albers origin.

Also adds support for previously unsupported object types e.g. Points
and LineStrings.

I never used these. They're fired!

Conflicts:
	src/core/core.js

This makes it easier to override via stylesheet.

It is now possible to reselect elements with scheduled transitions and redefine
associated tweens; this enables "post-selection" to customize the behavior of
reusable components undergoing transitions, such as an axis. This commit also
makes it much easier to sequence transitions.

Previously, a transition's tweens were stored privately by the transition and
could only be accessed through the transition. This made it impossible to modify
transitions created by components: the transition is not accessible externally,
and cannot be reselected from the document. Consider the following snippet:

  g.select(".x.axis")
      .call(xAxis)
    .selectAll("text")
      .attr("dy", null);

If `g` is a selection, then this code alters the appearance of the axis as
expected. However, if `g` is a transition, then transition.selectAll creates a
new concurrent transition, and now multiple tweens compete to set the "dy"
attribute. Oy!

Under the new design, an element's scheduled tweens are stored semi-privately on
the node (in the existing node.__transition__). Transition parameters can thus
be reselected and modified by transitions that share the same id. If you now
reselect a transitioning element, you modify the transition rather creating a
competing transition; this should be less surprising and allow greater control.

As a side-effect of this change, it is no longer possible to schedule concurrent
transitions on the same element, even with the same id: only one transition may
be active on a given element at any time. (Note that you can still schedule
multiple future transitions on the same element, and concurrent transitions on
different elements.) For example, you could previously schedule overlapping
transitions with different easing functions, delays or durations, provided you
were careful to avoid conflict. This seems like a relatively obscure use-case
compared to modifying a transition, so I believe this is a reasonable change.

This commit also changes transition.transition, such that the returned
transition starts at the end of the originating transition, rather than
overlapping. This makes it much easier to schedule sequenced transitions without
the complexity of transition.each("end") and d3.select(this).

Also, transitions are now simply arrays of nodes, consistent with selections!

Better resampling for non-linear distortions.

Timer callbacks are now assigned a unique id, such that if an existing callback
is later replaced, the timing parameters can be updated using a constant-time
lookup rather than a linear scan of the timer queue.

I've reinstated the original conditional for linear longitudinal
interpolation, since the x * x < ε conditional was not tight enough and
hence caused glitches in projections that include the poles e.g.
stereographic.

Back to solving the issue that was primarily noticeable in the Albers
demo: graticules near the poles would not resample correctly. We made an
assumption that a point mid-way along the connecting great arc is going
to be approximately mid-way in pixel-space. This is a reasonable
assumption in most cases, but there is an area of high instability near
the poles, since a small change in distance could result in a large
change in longitude, e.g. imagine a great arc going right over a pole,
so the longitude will change greatly depending on which side of the pole
you're on.

So the new fix is to see if the distance in pixel-space hardly changed
at all, and in this case we attempt to interpolate again, mid-way to the
new spherical coordinates. I think we can probably improve this even
further by borrowing methods from standard root-finding techniques e.g.
Newton-Raphson.

Add extra ring if polygon "surrounds" clip circle.

This can be optimised using the 3D coordinates directly from the
rotation step, rather than performing additional trig. calculations.

The antemeridian interpolator was mistakenly assuming longitudes of
interpolation points are exactly on the antemeridian. An alternative fix
would be to use exact equality tests for the coincident points test, but
it feels safer to use a margin of error.

This was breaking antemeridian clipping in some rare cases where a point
was coincident with the antemeridian.

Polygon LinearRings weren't being closed correctly.

It was breaking obviously in the resampling demo.

Retains a minor optimisation.

Interpolation of longitudes becomes inaccurate near the poles, since
they become mathematically insignificant in relation to a sphere.
However, for map projections, we want to interpolate longitudes even at
the poles themselves, since many projections distort the poles along an
edge.

Conflicts:
	d3.v2.min.js

This reverts commit b6d7c5e4.

Conflicts:
	d3.v2.min.js

Hurrah for tests!