This release includes a refactoring of the optimised read paths for RGBA data, optimisations for some of the python bindings to Imath, improvements to the cmake build environment as well as additional documentation describing deep data in more detail.
inlines. Bump PKGREVISION, as this modifies the installed header
as used by other packages.
Partially fixes building graphics/openexr under Ubuntu 10.10.
* Added explicit copy constructors to Imath::Matrix33<T> and
ImathMatrix44<T> to make conversions between float and double
matrices more convenient.
* Added slerpShortestArc() and euclideanInnerProduct() functions
to Imath::Quat<T>.
* Added 4D vector class template Imath::Vec4<T>.
* Copy constructors and assignment operators for Matrix33<T>
and Matrix44<T> are up to 25% faster. Added matrix constructors
that do not initialize the matrix (this is faster in cases where
the initial value of the matrix is immediately overwritten anyway).
* Rewrote function closestPointOnBox(point,box). Shortened
the code, improved numerical accuracy, fixed a bug where
closestPointOnBox(box.center(),box) would return the center
of the +Z side of the box, even if the +/-X or +/-Y sides
were closer.
* Rewrote function findEntryAndExitPoints() in ImathBoxAlgo.h.
Results are now consistent with those from intersect(), also
in ImathBoxAlgo.h.
* Made Vec2<T>::length() and Vec3<T>::length() more accurate for
vectors whose length is less than sqrt(limits<T>::smallest());
* Made Quat<T>::angle() more accurate for small angles.
This changes the buildlink3.mk files to use an include guard for the
recursive include. The use of BUILDLINK_DEPTH, BUILDLINK_DEPENDS,
BUILDLINK_PACKAGES and BUILDLINK_ORDER is handled by a single new
variable BUILDLINK_TREE. Each buildlink3.mk file adds a pair of
enter/exit marker, which can be used to reconstruct the tree and
to determine first level includes. Avoiding := for large variables
(BUILDLINK_ORDER) speeds up parse time as += has linear complexity.
The include guard reduces system time by avoiding reading files over and
over again. For complex packages this reduces both %user and %sys time to
half of the former time.
Half is a class that encapsulates our 16-bit floating-point format.
IlmThread is a thread abstraction library for use with OpenEXR
and other software packages. It currently supports pthreads and
Windows threads.
Imath implements 2D and 3D vectors, 3x3 and 4x4 matrices, quaternions
and other useful 2D and 3D math functions.
Iex is an exception-handling library.
