ChangeLog

# markup: rd

== Sat April 29 2017
- Enhancements
  * Suppress warnings in Ruby 2.4 by suppressing use of Fixnum.

== Mon January 30 2017
- Enhancements
  * Support for assigning Ruby Complex types to GSL::Matrix::Complex. (@v0dro)
  * Support for 2D interpolation by exposing GSLs interp2d and spline2d. (@v0dro)
  * Support for Ruby 2.4. (@badlamer)
- Fixes
  * Fix NArray superclass mismatch problem. (@v0dro)

== Fri March 4 2016
* Added compatibility for NMatrix interconversion with GSL::Vector and GSL::Matrix
* Added NMatrix compatibility to various to make them compatible with NMatrix data types.
* Gem is now compatible with GSL 2.1.
* Version bump to 2.1.0

== Thu Jul 03 2015
* Ruby/GSL 1.16.0.6
  * rb-gsl and gsl are now the same gem

== Thu Jul 02 2015
* Ruby/GSL 1.16.0.5
  * Optional narray support
  * Drop Ruby 1.8 support
  * Drop support for GSL < 1.15

== Fri Dec 19 2014
* Ruby/GSL 1.16.0.4
  * Optimize and extract multiplication and division operations on Fixnum
    and Float. Pull request #8 by Veselin Vasilev.
  * Fixed division in GSL::Oper for GSL::Vector::Col. Issue #9.

== Tue Oct 21 2014
* Ruby/GSL 1.16.0.3
  * Fixed RDoc issues. Issue #6 by @bigtunacan.
  * Fixed $LOAD_PATH problem. Pull request #7 by Takahiro SATOH.

== Wed Jul  9 2014
* Ruby/GSL 1.16.0.2
  * Fixed linking problem.

== Thu Apr 24 2014
* Ruby/GSL 1.16.0.1
  * Extensive cleanup.
  * Fixed linking problems.
  * Required Ruby version >= 1.8.7.

== Fri Jan 24 2014
* Ruby/GSL 1.16.0
  * GSL-1.16 support.

== Fri Jan 24 2014
* Ruby/GSL 1.15.3.2
  * Windows compatibility.
  * Ruby 2.x compatibility.
  * Ported tests to Test::Unit.
  * Silenced compilation warnings.
  * Switched to rake-compiler for compilation.

== Sat Feb 26 2011
* Ruby/GSL 1.14.7
  * Fix conversion of non-DFLOAT NArray to GSL::Vector and
    GSL::Vector::Complex.

== Thu Feb 24 2011
* Ruby/GSL 1.14.6
  * Add Vector::Complex support to NArray conversions
  * Raise exception when trying to create View from NArray object of
    incompatible type.

== Sun Nov 14 2010
* Ruby/GSL 1.14.5
  * Convert docs from rdtool to rdoc
  * Fix object id handling bugs on 64-bit systems
  * Fix Vector's handling of Range parameters

== Wed Sep 29 2010
* Modification for ruby-1.9.2
  * Replaced STR2CSTR to StringValuePtr

== Mon Mar 22 2010
* GSL-1.14 support
  * Added GSL::Multiset class
  * Added GSL::Function#glfixed(a, b table)

== Sun Aug 16 2009
* Added Non-negative Matrix Factorization (NMF) support (by Roman Shterenzon)

== Sat Aug  8 2009
* Added GSL-1.13 features:
  * Added GSL::Monte::Miser#params_get, GSL::Monte::Vegas#params_get
  * Added GSL::BSpline::greville_abscissa(i)
  * Added GSL::Poly#eval_derivs(x)
  * Added a new minimization algorithm "quad_golden"
  * Added a new multimin algorithm "nmsimplex2rand"

== Tue Jun  9 2009
* Fixed bug in linalg.c (bug [#25842])

== Thu Feb  2 2009
* Fixed a bug in GSL::Histogram.alloc_uniform() (bug [#23040])
* Fixed a bug in GSL::Vector#/ (bug [#23334])

== Thu Nov  6 2008
* Match with the tensor extension version 0.99
  * now not compatible for the versions older than tensor-0.97.

== Sat Jul 26 2008
* GSL::Poly.eval now handles complex numbers (GSL-1.11)
* Added new quasi-random generators "halton" and "reversehalton"
* Added a module function GSL::MultiFit::linear_residuals() (GSL-1.11)
* Fixed a bug in ODE solver

== Mon Jun 30 2008
* Fixed GSL::Vector#normalize (#20865)

== Mon May 26 2008
* Fixed a bug in GSL::Matrix::to_v (#19757)
* Fixed a bug in GSL::Vector::Int multiplication (#19924)

== Sat Dec  8 2007
* Version 1.10.3

== Tue Dec  2 2007
* Improved GSL::Vector#get for a Range arg
* Improved GSL::Vector#concat for an Array arg
* Fixed a bug in GSL::Matrix.alloc with column vector arguments

== Thu Dec  2 2007
* Added methods GSL::Vector#pow(p), GSL::Vector#pow!(p)

== Wed Oct 17 2007
* Fixed a bug in poly_source.c

== Sun Oct  7 2007
* Version 1.10.0
* ndlinear extension support

== Sat Sep 22 2007
* Added GSL-1.10 features
  * Added generalized eigensystem methods
  * Added method GSL::Stats::correlation(v1, v2) to compute
    Pearson correlation of two datasets
  * Added methods GSL::Sf::expint_En(n, x), GSL::Sf::expint_En_e(n, x)
    for computing the n-th order exponential integral.
  * Added methods GSL::Vector#isnonneg, GSL::Vector#isnonneg?,
    GSL::Matrix#isnonneg, and GSL::Matrix#isnonneg?
  * Added methods GSL::Matrix#subrow, GSL::Matrix#subcolumn
  * Added methods GSL::Linalg::Chomplex::Cholesky::decomp,
    GSL::Linalg::Chomplex::Cholesky::solve,
    and GSL::Linalg::Chomplex::Cholesky::svx
  * Added method GSL::hypot3(x, y, z)

== Sat Aug 11 2007
* Modified ext/extconf.rb: Check gsl_stats_correlation()

== Sun Aug  5 2007
* Improved GSL::Vector::filescan()
== Tue May 15 2007
* Fixed a bug in Vector#get (Thank to Daigo Moriwaki)

== Sun Feb 18 2007
* Completed GSL-1.8.90 features
  * Change names "unsymm" ---> "nonsymm" for non-symmetric eigen systems
  * Added Mathieu functions
  * Added Vector#ispos, isneg
  * Added knuthran2002 generator
  * Added a new minimization method GSL::MultiMin::FdfMinimizer::BFGS2

== Sun Dec 31 2006
* Remove the module function GSL::equal?, which breaks the Ruby policy.

== Sun Dec 24 2006
* Added Basis splines
* CQP and Bundle extentions support

== Wed Dec 13 2006
* OOL extension support (OOL version 0.2)
  * OOL: Open Optimization Library

== Mon Dec 11 2006
* Add a quasi-random sequence generator hdsobol
  (require the qrngextra extension)

== Tue Sep 26 2006
* Fixed a bug in GSL::Vector#*

== Sun Sep 24 2006
* Fixed a bug in GSL::Matrix.vandermonde()
* Added methods GSL::Vector#join, zip, cumsum, cumprod

== Thu Sep 21 2006
* Added methods to calculate eigenvectors of nonsymmetric matrices
* Added test scripts tests/eigen/unsymm.rb, tests/eigen/unsymmv.rb

== Wed Aug 23 2006
* Added methods to calculate eigenvalues of nonsymmetric matrices

== Fri Aug 18 2006
* Fixed bug in Sf::hypot

== Sat Apr 29 2006
* Applied patch provided by Cameron McBride, to fix a bug in
  randist.c
* Applied patched by zunda to fix some test/ scripts.

== Thu Apr 13 2006
* Version 1.8.0
  * Added cumulative distribution functions under the module GSL::Cdf:
    binomial, poisson, geometric, negative binomial, pascal and hypergeometric.
  * Added module functions GSL::Cdf::beta_Pinv, beta_Qinv, fdist_Pinv, and
    fdist_Qinv.
  * Added the multilinear fit estimator function GSL::MultiFit::linear_est.
  * Added new random variate generators,
    GSL::Ran::gaussian_ziggurat and gamma_mt.
  * Added Debye functions for n=5 and n=6, as
    GSL::Sf::debye_5 and debye_6.
  * Added GSL::Spline#min_size
  * Added the Debye unit to physical constants.

== Fri Dec  2 2005
* Singleton method "new" ---> "alloc"

== Sun Nov 27 2005
* Apply patch by zunda to improve multifit.
* Fix bug in dirac.c

== Tue Apr  3 JST 2005
* Added Vector#collect!, Matrix#collect!
* Added lib/rbgsl.rb, in which the GSL module is not included by default

== Wed Apr 27 2005
* The following Linalg methods now can handle NArray objects:
  * GSL::Linalg::
    * LU.decomp(m)
    * LU.solve(lu, b)
    * LU.svx(lu, bx)
    * LU.det(lu, sign)
    * LU.lndet(lu)
    * LU.invert(lu, perm)
    * QR.decomp(m)
    * QR.solve(qr, tau, b)
    * QR.svx(qr, tau, bx)
    * SV.decomp(m)
    * SV.solve(u, v, s, b)
    * SV.svx(u, v, s, bx)
    * Cholesky.decomp(m)
    * Cholesky.solve(u, v, s, b)
    * Cholesky.svx(u, v, s, bx)
    * HH.solve(m, b)
    * HH.svx(m, bx)

== Mon Apr 18 2005
* Improved gsl_narray.c (thanks to T. Horinouchi)
* Modified extconf.rb
  * Automatic finding of NArray
* Applied a patch by zunda

== Thu Apr  7 2005
* Modified extconf.rb

== Mon Mar 21 2005
* Version 1.6.2

== Sat Mar 19 2005
* Fixed Blas.dgemm, .zgemm
* Fixed method re-definitions

== Tue Mar 15 2005
* Added class GSL::Block::Byte, GSL::Block::Index
* Added NArray-like methods, Vector#eq, ne, gt, where, any? etc.
* Added methods Vector#floor, Vector#ceil,
  and Vector#round (if the system has)

== Sat Mar 12 2005
* Fixed Vector#get

== Sat Feb 26 2005
* Ruby 1.9.0 support

== Sat Feb  5 2005
* Version 1.6.1

== Wed Jan 26 2005
* Added in-place FFT methods
  * Fixed FFT document

== Thu Jan 20 2005
* Added Vector#decimate, Vector#diff
* Added Matrix#rot90
* Fixed Matrix::Complex

== Thu Jan 06 2005
* Some methods which receive GSL::Vector now can also manipulate
  NArray objects.
  * Special functions
  * Probability distributions
  * Statistics
  * FFTs
  * Wavelet transforms
  * Interpolations
  * Serices accelerations
* Added method Vector#decimate
* Sf::exp now can receive Complex or Vector::Complex

== Tue Jan 04
* Version 1.6.0

== Thu Dec 30 2004 Yoshiki Tsunesada
* Update for GSL-1.5.90
  * Added Sf::psi_1
  * Added ODE stepper rk2simp
  * Added LQ, PTLQ decompositions
  * Added methods Cdf::exppow_P, Cdf::exppow_Q

== Mon Dec 27 2004 Yoshiki Tsunesada
* All Ruby/GSL classes are defined as derivatives of the base class
  GSL::Object.
* Defined the method GSL::Object#info
* Added Vector#histogram
* Added Vector.filescan
* Modified the methods Histogram#increment, accumulate
* Added methods Poly.fit, Poly.wfit, MultiFit.polyfit
* Added methods Vector::Complex#abs2, abs, arg
* Modified FFT methods: now they return FFTed data as a new vector,
  and do not destroy the input vector.

== Sat Dec 11 2004 Yoshiki Tsunesada
* Modified Histogram#integral

== Fri Dec 10 2004 Yoshiki Tsunesada
* Added methods Histogram#reverse, Histogram2d#integrate,
  Histogram2d#normalize

== Mon Dec  6 2004 Yoshiki Tsunesada
* Version 1.5.4

== Fri Dec  3 2004 Yoshiki Tsunesada
* Added Histogram#rebin

== Fri Nov 26 2004 Yoshiki Tsunesada
* Added method Vector#to_m
* Added methods Matrix#reverse_rows, Matrix#reverse_columns

== Tue Nov 16 2004 Yoshiki Tsunesada
* The method Matrix#** is now obsolete, use "*"
  for matrix multiplication instead.
* Improved Matrix#* and Vetor#*
* Added methods Vector#to_m_diagonal, Vector#to_m_circulant,
  Vector#collect
* Added methods Matrix.hilbert, Matrix.invhilbert, Matrix.pascal,
  Matrix.vandermonde, Matrix.toeplitz, Matrix.circulant
* Added methods Vector.indgen, Vector#indgen,
  Matrix.indgen, Matrix#indgen (mimics NArray)
* Added class GSL::Histogram3d
* Checked GSL lower version compatibilities
  * gsl-1.0 ~ gsl-1.5: OK
  * gsl-0.9.4: OK
  * gsl-0.9.3 or older: not tested
  * gsl-1.6 (development version, CVS Nov2004): OK

== Sun Nov  7 2004 Yoshiki Tsunesada
* Support the gsl_tensor package
  * http://sources.redhat.com/ml/gsl-discuss/2004-q4/msg00053.html
  * Added class GSL::Tensor, GSL::Tensor::Int
* The methods Vector#get, Vector#[] now receive multiple
  arguments, or an Array, or a Range object.

== Thu Nov  4 2004 Yoshiki Tsunesada
* Added class GSL::Histogram::Integral, which represents the
  cumulative distribution.
* Added method GSL::Histogram#integrate, which creates a
  GSL::Histogram::Integral object.
* Added methods GSL::Histogram2d#xproject, yproject, normalize, increment2.

== Sun Oct 31 2004 Yoshiki Tsunesada
* Fix the document for Polynomials
* Added methods Poly::hermite(n), Poly::cheb(n), and Poly::cheb_II(n) etc..
* Added method Poly#compan, which returns the companion matrix.
* The method Matrix#/ is improved: if a Vector is given, this method
  solves the linear system.

== Mon Oct 25 2004 Yoshiki Tsunesada
* Split the source file "randist.c" into "randist.c" and "cdf.c"
* Split the source file "diff.c" into "diff.c" and "deriv.c"
* Split the source file "interp.c" into "interp.c" and "spline.c"
* Added methods Vector#abs, sqrt, square, normalize
* Added sample scripts odeiv/frei1.rb, fft/frei2.rb
  and eigen/qhoscillator.rb, which deal with quantum mechanical
  calculations. These are translated from the C++ code in the
  Harald Wiedemann's textbook "Numerische Physik"
  (Springer 2004, ISBN 3-540-40774-X).

== Sat Oct 23 2004 Yoshiki Tsunesada
* Added a method GSL::Vector#logspace2. The following two are
  equivalent:
     v1 = Vector.logspace(1, 3, 5)
     v2 = Vector.logspace2(10, 1000, 5)

== Tue Oct 19 2004 Yoshiki Tsunesada
* Support the rngextra package
  * http://www.network-theory.co.uk/download/rngextra/
  * Added two random number generators "rngextra_rng1" and "rngextra_rng2"
  * Fixed a bug in the method Vector#subvector_with_stride

== Mon Oct 18 2004 Yoshiki Tsunesada
* Added a sample script dht.rb

== Fri Oct  8 2004 Yoshiki Tsunesada
* Version 1.5.3

== Thu Oct  7 2004 Yoshiki Tsunesada
* The methods Vector.graph and GSL::graph improved.
* Added sample scripts of special functions and probability distribution
  functions.

== Sat Oct  2 2004 Yoshiki Tsunesada
* Discrete wavelet transform (experimental)
* Presently this is only for CVS trackers,
  since DWT is not supported in the latest version GSL-1.5.
* Added methods Vector.connect, Vector#connect

== Thu Sep 30 2004 Yoshiki Tsunesada
* The methods Vector#[] and Vector#[]= accept an index of
  negative integer. For example, v[-1] returns the last element
  of the vector v, v[-2] does the second last, and so on.

== Fri Sep 24 2004 Yoshiki Tsunesada
* Fixed bug in Sf::legendre_Ql

== Wed Sep  1 2004 Yoshiki Tsunesada
* Version 1.5.2

== Fri Aug 27 2004 Yoshiki Tsunesada
* Experimental implementation of GNU plotutils graph and GNUPLOT interfaces
* Added unary minus operators for GSL::Vector, GSL::Matrix

== Thu Aug 19 2004 Yoshiki Tsunesada
* Added class GSL::Matrix::Int
* Correct descriptions of GSL::Matrix#get_row, get_col
* Changed behaviour of methods NArray#to_gv, NArray#to_gm,
  added methods NArray#to_gv_view, NArray#to_gm_view

== Wed Aug 18 2004
* Added method GSL::set_error_handler

== Tue Aug 17 2004
* Version 1.5.1

== Sat Aug 14 2004 Yoshiki Tsunesada
* Changed behaviour of the methods Odeiv::Evolve#apply, Odeiv::Solver#apply
* Fixed bug in GSL::Histogram#get
* Fixed buf in GSL::Vector#matrix_view_with_tda
* Added class GSL::Vector::Int
* Added test suite

== Wed Aug 11 2004 Yoshiki Tsunesada
* Fixed bug in GSL::Sf::legendle_Ql_e
* Fixed bug in GSL::MultiFit.linear

== Tue Aug 10 2004 Yoshiki Tsunesada
* Fixed bugs in GSL::Cdf::gumbel2_P, gumbel2_Q
* Version 1.5.0

== Fri Aug  6 2004 Yoshiki Tsunesada
* Fixed bug in Histogram2d.new_uniform

== 4.Aug.2004
* Correct source code errors in interp.c, multifit.c

== 2.Aug.2004
* Added methods GSL::Vector#dot (inner product)
* Added classes GSL::Vector::Col < GSL::Vector,
  GSL::Vector::Col::View < GSL::Vector::Col

== 1.Aug.2004
* Version 0.9.9

== 30.Jul.2004
* Added methods GSL::Rng.default_seed, GSL::Rng.default_seed=(seed)

== 26.Jul.2004
* Cumulative distribution functions
* Changed module path GSL::Ran:: ---> GSL::Cdf
* Constants GSL::VERSION, GSL::RUBY_GSL_VERSION defined

== 24.Jul.2004
* Experimental implementation of discrete wavelet transforms (tests/wavelet)

== 22.Jul.2004
* Elementary functions as GSL::sin now can take argument of Integer, Float,
  Array, Vector or Matrix.
* Strict type check

== 20.Jul.2004
* Added methods GSL::heapsort, Vector#heapsort, Vector::Complex#heapsort
* Added methods GSL::heapsort_index, Vector#heapsort_index,
  Vector::Complex#heapsort_index
* version 0.9.4

== 18.Jul.2004
* Added missing functions GSL::Sf::legendre_Ql
* Added missing functions GSL::Sf::psi_e
* Added missing functions GSL::Sf::zetam1, zetam1_int

== 17.Jul.2004
* Added GSL::Eigen::Symmv::sort, GSL::Eigen::Hermv::sort
* Update GSL::Monte

== 15.Jul.2004
* Version 0.9.2
* Added methods Matrix#each_row, Matrix#each_col
* Added methods Matrix::Complex#each_row, Matrix::Complex#each_col
* Fixed bug in Complex::polar
* Fixed bug in Complex#abs, arg, abs2, logabs

== 13.Jul.2004
* Added Matrix#trace, Matrix::Complex#trace

== 11.Jul.2004
* version 0.9.0
* Householder Transformations
* LU decomposition for complex matrices
* All the GSL functions implemented.

== 10.Jul.2004
* Blas level 3
* Fixed GSL::Vector::sort_index

== 8.Jul.2004
* Correct document on random distributions

== 7.Jul.2004
* Fixed bug in the method GSL::Sf::coulomb_wave_sphF_array

== 6.Jul.2004
* GSL 1.5 support
* Fixed a bug in the method GSL::Sf::bessel_sequence_Jnu_e

== 5.Jul.2004
* Compile test with -Wall option:
  * remove unused variables, check missing prototype declarations
* Fixed a bug in the method GSL::Multimin::test_size (wrong binding)

== 30.Jun.2004
* ver 0.4.7

== 28.Jun.2004
* Switch rb_f_lambda() and rb_block_proc(), for proper ruby version.

== 20.Jun.2004
* Added Linalg::balance_comlumns(A, D) and Matrix#balance_columns(D)
* BLAS support
  * Level 1
  * Level 2

== 19.Jun.2004
* Support gsl-1.4.90
  * Added Stefan-Boltzmann constant, Thomson scattering cross section
  * Module Deriv

== 13.Jun.2004
* Fixed a bug in the sample script integration/friedmann.rb
* Improved interfaces of polynomial equations
* Correct the document on GSL::Complex

== 10.Jun.2004
* ver 0.4.2
* The module GSL::Math removed
* The module name GSL::Const is changed to GSL::CONST

== 09.Jun.2004
* ver 0.4.1
* Add random number generators  borosh13, coveyou, fishman18,
  fishman20, fishman2x, knuthran, knuthran2, lecuyer21, waterman14.
* Correct the documentation on random number generators
* Add functions expint_E1_scaled,expint_E2_scaled, and others
* Add methods Rng::fwrite, fread
* Add function binomial_tpe

== 05.Jun.2004
* 0.4.0

== 02.Jun.2004
* rb_gc_mark
* Implemented simulated annealing, Ntuples

== 25.Apr.2004
* Vector, Matrix, Matrix::Complex coerce

== 24.Apr.2004 version 0.3.11
* GSL::Complex, coerce

== 22.Apr.2004  version 0.3.10
* polish sources of special functions

== 18.Apr.2004, version 0.3.9
* bug fix Eigensystems, Interpolation

== 16.Apr.2004
* bug fix Vector, Matrix, Linalg

== 14.Apr.2004, version 0.3.8
* N-tuples
* Improvement in NArray object conversions (thanks to S. Nishizawa)

== 13.Apr.2004
* Poly divided-difference reporesentation, Taylor expansion
* Apply patch by S. Nishizawa, for type checks in C code,
  and some modifications in sample scripts

== 12.Apr.2004, version 0.3.7
* Add mathematical functions ldexp, frexp
* Add gsl_sf_hazard function
* QRng bug fix
* GSL::Stats module

== version 0.3.6
* Monte-Carlo integration

== version 0.3.5
* Multidimensional minimization

== version 0.3.4
* Combinations, One dimensional minimization, Multidimensional root-finding

== version 0.3.3
* Nonlinear least-squared fitting

== versoin 0.3.2
* Least-squared fitting

== version 0.3.1
* One dimensional root-finding

== version 0.3.0
* FFTs, Ordinary differential equations

== 2002 version 0.2.5