Commit Graph

109 Commits

Author SHA1 Message Date
Sébastien Crozet
40c14ace39 Make from_array-like methods static.
Fix #45.
2014-11-25 11:00:26 +01:00
Eduard Bopp
733219d728 Have ToHomogeneous use &self in method signature
I find makes the syntax a bit lighter.
2014-11-22 12:01:45 +01:00
Sébastien Crozet
bab38ca6d5 Add conversion from/to arrays for matrices and vectors.
Use `.as_array()`, `.as_array_mut()`, `.from_array_ref()`, `.from_array_mut()`.
Fix #33.
2014-11-21 11:34:46 +01:00
Sébastien Crozet
d8dfedbf99 Update to the last rust-nightly.
Version of rustc: 0.13.0-nightly (336349c93 2014-11-17 20:37:19 +0000).
2014-11-19 12:11:32 +01:00
Sébastien Crozet
77348f668f Use our own traits instead of the deprecated ones.
We might switch to bjz/num-rs eventually (#32), but in the mean time we just re-declare the
deprecrated traits in nalgebra.
2014-11-16 14:42:18 +01:00
Sébastien Crozet
39b9d20717 Quick update to the last rust-nightly.
This is just a quick-fix so that nalgebra compiles.
This does not fix the deprecation warnings!

Version of rustc: 0.13.0-nightly (d91a015ab 2014-11-14 23:37:27 +0000).
2014-11-15 15:47:59 +01:00
Sébastien Crozet
226b115f1e Overload DVec * DVec -> DVec and DVec / Dvec -> DVec.
Fix #31.
2014-11-06 23:17:35 +01:00
Sébastien Crozet
a2848e6e18 Update to the last rust-nightly.
Version of rustc: 0.13.0-nightly (63c4f22f2 2014-11-05 22:31:44 +0000).
2014-11-06 14:52:52 +01:00
Sébastien Crozet
3834913402 Update to the last rust-nightly.
Version of rustc: 0.13.0-nightly (3327ecca4 2014-11-01 22:41:48 +0000).
2014-11-02 22:47:11 +01:00
Sébastien Crozet
eb745df4bc Update to the last rust-nightly.
Version of rustc: 0.13.0-nightly (221fc1e3c 2014-10-31 02:27:15 +0000).
2014-10-31 17:40:47 +01:00
Sébastien Crozet
36d3e88cb0 Do not automatically impl Scalar{Mul,Div,Add,Sub}.
This makes them implementable without using the double dispatch trick.
2014-10-30 09:21:22 +01:00
Sébastien Crozet
93b184815f Always use Cast<f64> instead of Cast<f32>. 2014-10-30 09:21:22 +01:00
Sébastien Crozet
6b4b25acd4 Basis: add a method to compute the i-th element of the canonical basis. 2014-10-30 09:21:20 +01:00
Sébastien Crozet
171576e2a0 Move the eigen_qr function behind the EigenQR trait.
This simplifies generic programming.
2014-10-30 09:21:20 +01:00
Sébastien Crozet
27be1f0651 Add a SquareMat trait for square matrices. 2014-10-30 09:21:20 +01:00
Sébastien Crozet
6a194b2b62 Rename: Any{Vec,Pnt} to Num{Vec,Pnt} + do not impl them automatically.
Instead, implement them manually. This clarifies error messages for the users.
2014-10-30 09:21:19 +01:00
Sébastien Crozet
a0fffe93a9 Move the .shape() method to its own trait: Shape. 2014-10-30 09:21:18 +01:00
Sébastien Crozet
2b23446d54 Matrix Index/IndexMut implementation: return an element instead of a column.
There is no reason why indexing would prefer returning a column instead of a line. Instead, we
return an element, and let the user use the `Col` and `Row` traits istead.
2014-10-30 08:38:05 +01:00
Sébastien Crozet
074c9356bf Rename: PartialOrd -> POrd, PartialOrdering -> POrdering.
This avoids name clashes with the standard library.
2014-10-30 08:38:05 +01:00
Steve Klabnik
426ba307ab fail -> panic 2014-10-30 00:25:47 -04:00
Robin Gloster
4e4eb62ffc upgrade to newer rust 2014-10-22 13:35:17 +02:00
Sébastien Crozet
e2fc8aa764 Deprecate to_pnt, as_pnt and VecAsPnt.
Those are meaningless. Use `na::orig() + the_vector` instead.
2014-10-20 22:35:43 +02:00
Sébastien Crozet
5ba9f27530 Add quaternions.
Fix #24.
2014-10-14 22:20:49 +02:00
Sébastien Crozet
0cf725f5f2 Add 3D orthogonal and perspective projection structures.
Fix #26.
2014-10-12 20:40:46 +02:00
Sébastien Crozet
03eda09084 Some documentation fixes.
Fix #25.
2014-10-12 10:35:56 +02:00
Sébastien Crozet
b6090042ac Add a Axpy that implements y <- y + ax.
This is very useful for accumulations to compute, e.g. centers.
2014-10-12 09:24:16 +02:00
Sébastien Crozet
de2479b70f Do not implement Transform<Vec*> for Vec*.
This is error prone as this is a no-op for vectors and a translation for points.
2014-10-11 00:23:10 +02:00
Sébastien Crozet
70de340498 Remove the Iso * Vec overload and the ability to Transform a Vector with an Iso.
This is too error prone as the user has to remember if he is using a vector or a point.
Istead, to transform a vector with an Isometry, use the internal rotation matrix of the isometry:
`iso.rotation * vector` instead of `iso * vector`.

The `Rotation` trait can also be used.
2014-10-10 22:21:05 +02:00
Sébastien Crozet
6fb6d35980 Documentation fixes. 2014-10-10 20:56:40 +02:00
Sébastien Crozet
e852eb4783 Fix the implementation of ToHomogeneous for Iso*. 2014-10-10 20:56:24 +02:00
Sébastien Crozet
7e9bb98823 Add to_pnt and to_vec methods. 2014-10-10 12:19:37 +02:00
Sébastien Crozet
924d8269d8 Add points.
This adds the Pnt{1,2,3,4,5,6} structures.
This adds the traits:
  − AnyPnt
  − FloatPnt
  − PntExt
  − FloatPntExt
  − Orig (to return the zero point)
  − PntAsVec
  − VecAsPnt
This adds operator overloading:
  − Pnt + Vec
  − Pnt - Vec
  − Pnt * Scalar
  − Pnt / Scalar
  − Pnt + Scalar
  − Pnt - Scalar
  − Iso * Pnt
  − Rot * Pnt
  − Pnt * Iso
  − Pnt * Rot
This changes some behavior:
  − Iso multiplication with a Vec does not translate the vector any more.
  − ToHomogeneous adds a 0.0 at the end of a Vec and a 1.0 at the end of a Pnt.
  − FromHomogeneous performs w-normalization on a Pnt, but not on a Vec.
  − The Translate<Vec> trait is never implemented (i-e. a Vec is not to be translated).

cc #25
2014-10-10 11:45:20 +02:00
Sébastien Crozet
8f89ac421c Update to the last rust.
Version of rustc: rustc 0.12.0-nightly (d64b4103d 2014-09-26 21:47:47 +0000).
2014-09-27 09:54:03 +02:00
Sébastien Crozet
086088a0ed Fix the implementation of Row for fixed-size matrices.
Matrices are column-major.
2014-09-19 23:51:27 +02:00
Sébastien Crozet
1945cb09f2 Merge pull request #20 from bjz/warnings
Fix warnings
2014-09-19 23:43:46 +02:00
Brendan Zabarauskas
d3abbfcd4c Fix warnings 2014-09-18 15:20:04 +10:00
Brendan Zabarauskas
b6959ef913 Implement index operators
Closes #18
2014-09-18 12:23:52 +10:00
Ben Foppa
6c8f14966c update for rustc 21d1f4d7c 2014-09-15 21:30:36 -04:00
Sébastien Crozet
560dff3891 Update to the last rust-nightly.
Version of rustc: 0.12.0-pre-nightly (2e92c67dc 2014-08-28 23:56:20 +0000).
2014-08-29 21:35:42 +02:00
Sébastien Crozet
844092205e Do not compute the matrix inverse if the determinant is near-zero. 2014-08-17 22:41:55 +02:00
Sébastien Crozet
663f8b3ccb Add a Diag to build, get and set a matrix diagonal. 2014-08-16 13:22:53 +02:00
Sébastien Crozet
40c9915870 Add the types: DVec1, .., DVec6.
They are stack-allocated, dynamically sized vectors with a maximum size.
This is useful for slicing small matrices, without allocation.
2014-08-16 12:16:26 +02:00
Sébastien Crozet
7d6884c3df Add a Det trait to compute the determinant + implement it for Mat{1,2,3}. 2014-08-11 19:54:13 +02:00
Sébastien Crozet
2f0234e7ac Avoid modulo on DMat::from_fn. 2014-07-27 09:13:13 +02:00
Nathan Stien
ebe1ed1f16 Fix coordinate calculation in DMat::from_fn().
The builder function `f` was receiving incorrect coordinates, often
involving a uint underflow.

Added a test case to verify the new behavior.
2014-07-26 19:03:37 -05:00
Sébastien Crozet
e3d1bf6f92 Update to the last rust-nightly.
Version of rustc: 0.11.0-nightly (5ad7286dc37839b7234ff51aab172e48935869d8 2014-07-12 00:31:46 +0000)
2014-07-12 09:30:49 +02:00
Sébastien Crozet
7b9a3f2bb9 Update to the last rust-nightly.
Version of rustc: 0.11.0-nightly (459f155f81291c46633e86a480628b50304ffb1c 2014-07-04 23:46:44 +0000).
2014-07-05 10:33:57 +02:00
Sébastien Crozet
efc2f46019 Update to the last rust-nightly.
Version of rustc: 0.11.0-pre-nightly (e55f64f 2014-06-09 01:11:58 -0700).
2014-06-09 21:03:36 +02:00
Sébastien Crozet
5066d6b9de Update to the last rust-nightly.
Version of rustc: 0.11.0-pre-nightly (918dbfe 2014-06-02 20:51:30 -0700).
2014-06-03 22:37:46 +02:00
Sébastien Crozet
f6c597f102 Update to the last rust-nightly.
Version of rustc: 0.11.0-pre-nightly (faa7ba7 2014-05-31 01:06:40 -0700).

Main changes:
        * `cmp::Ord` -> `cmp::PartialOrd`
        * `cmp::Eq`  ->  `cmp::PartialEq`

Note that `na::PartialOrd` is not the same as `cmp::PartialOrd`
(which lacks a lot of partial ordering operators).
2014-06-01 15:22:11 +02:00