/usr/share/doc/libwfmath-1.0-dev/TODO is in libwfmath-1.0-dev 1.0.2+dfsg1-0.4.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 | Inline in the foo.h any method in foo_funcs.h that does not require another
header, and has no loop
Re-write operators so that for a given operator O, operatorO is implemented
using operatorO= rather than the other way around which is inefficient.
Add throw(ColinearVectors) to the appropriate vector and matrix
rotation functions
Get rid of templates as template args, use ValueType for standard
containers instead
Intersect functions unimplemented for dim != 2,3 (extremely low priority):
Intersect(Polygon<>, AxisBox<>, bool)
Intersect(Polygon<>, RotBox<>, bool)
Intersect(RotBox<>, AxisBox<>, bool)
Intersect(RotBox<>, RotBox<>, bool)
These will be postponed until I can think of a way to write them,
or someone needs them, whichever comes first
Add shape tests from COAL
Write rotateInverse() functions
Look into whether miniball code can handle using float instead of double;
if so, replace Wrapped_array with Point
Requests from Jesse: NAN, INFINITY, RADIANS_TO_DEGREES, DEGREES_TO_RADIANS
Write an API reference for the library. Find some inline documentation tool?
Deal with difficulties in making certain functions friends of the
appropriate classes
Rearrange the functions in foo_funcs.h to match the order in the class
in foo.h
Schedule:
0.3.0 will come out once I get around to testing the Polygon intersect functions
and make sure they work. At that point the old Intersect()/IntersectProper()
interface will be removed, as will support for deprecated network
formats in atlas_conv.h. This is also an appropriate time to make changes
that break binary compatibility.
alriddoch's TODO 2008-11-10
Clean up const.h removing all includes that it itself does not use.
<cmath>, <cassert> and <iosfwd> should go. Clean up the rest of the headers
that turn out to need this, but completely get rid of asserts from the
interface headers. This should reduce the cost of including these headers,
and stop them from polluting the namespace.
|