projects/gsll/gsll.git
2 months agoImproved docstring for #'linear-fit master QL-2013-03
Liam M. Healy [Thu, 3 Jan 2013 23:21:01 +0000] 
Improved docstring for #'linear-fit

2 months agoFix flysheet to indicate dependence on new version of CFFI
Liam M. Healy [Sat, 2 Mar 2013 21:13:56 +0000] 
Fix flysheet to indicate dependence on new version of CFFI

4 months agoImproved docstrings
Liam M. Healy [Wed, 2 Jan 2013 03:40:40 +0000] 
Improved docstrings

4 months agoFix typo in docstring
Liam M. Healy [Mon, 31 Dec 2012 22:58:25 +0000] 
Fix typo in docstring

6 months agoSubstitute for dim0 and dim1 in funcallables when scalarsp=NIL
Liam M. Healy [Sun, 18 Nov 2012 16:27:24 +0000] 
Substitute for dim0 and dim1 in funcallables when scalarsp=NIL

Previously, the funcallable form made by #'make-funcallable-form (for
e.g. ODE solvers) when scalarsp=T involved substituting the symbols
'dim0 and 'dim1, but when scalarsp=NIL, would not do the substitution.
This caused an error when those unevaluated and unbound symbols
appeared in the lambda.  By altering #'faify-form with an additional
'dimension-values argument and calling
(value-from-dimensions argspec dimension-values)
instead of
(parse-callback-argspec argspec 'dimensions),
this problem is fixed.

15 months agoRename dependent system cffi-libffi, fix Darwin path
Liam M. Healy [Sun, 19 Feb 2012 15:42:07 +0000] 
Rename dependent system cffi-libffi, fix Darwin path

16 months agoForeign type name :sizet from CFFI replaces 'cffi:sizet
Liam M. Healy [Fri, 13 Jan 2012 17:12:31 +0000] 
Foreign type name :sizet from CFFI replaces 'cffi:sizet

16 months agoOptional index arrays for sort-smallest-index, sort-largest-index
Liam M. Healy [Tue, 3 Jan 2012 23:05:55 +0000] 
Optional index arrays for sort-smallest-index, sort-largest-index

Arguments simplified in sort-smallest-index, sort-largest-index;
output array is optional and may be specified by size.  Docstrings
made accurate.  Previous versions selected only the first n part of
the unsorted array.  Other functions in this file could benefit from
similar changes, but the changes have not been made because of the
existence of tests.

18 months agoFix special function return sf-result and sf-result-e10 type specifications
Liam M. Healy [Fri, 25 Nov 2011 03:58:02 +0000] 
Fix special function return sf-result and sf-result-e10 type specifications

18 months agoUse (:struct foo) syntax to eliminate style warnings from CFFI
Liam M. Healy [Sun, 30 Oct 2011 14:35:32 +0000] 
Use (:struct foo) syntax to eliminate style warnings from CFFI

18 months agoForeign structure with CFFI's new syntax
Liam M. Healy [Sun, 30 Oct 2011 04:44:08 +0000] 
Foreign structure with CFFI's new syntax

Specify the foreign structures with CFFI's new syntax, (:struct foo),
and fix #'creturn-st so that it will not take :struct (or any other
keyword) as a return variable name.  GSLL now compiles and loads
without error or warning, but it has not been tested.

18 months agoFix sf-result* definitions, eliminate ffexpand, fsbv:object
Liam M. Healy [Sat, 29 Oct 2011 23:35:40 +0000] 
Fix sf-result* definitions, eliminate ffexpand, fsbv:object

Fix sf-result and sf-result-e10 translation methods, and use in
complex-with-error and values-with-errors.  Some usage in gamma.lisp
updated, but there will be a need to replace sf-result with (:struct
sf-result) in all declarations for many special functions.

I removed ffexpand and everything it depended on because there is no
reason to make a side defcfun for call-by-value functions, and if FSBV
is missing CFFI already emits an error.  All usage of fsbv:object and
any thing from the FSBV package is now gone.

19 months agoEliminate fsbv:object for built-in types
Liam M. Healy [Mon, 24 Oct 2011 02:21:19 +0000] 
Eliminate fsbv:object for built-in types

Eliminate fsbv:object for built-in types; started conversion of
special function result (sf-result*) structures but this requires a
modification to CFFI.  Simulated annealing structure called by value.
Use cffi-fsbv's sizet definition; this imposes a dependence on that
system.  No compilation attempted yet.

19 months agoEliminate defenumeration
Liam M. Healy [Mon, 24 Oct 2011 00:48:20 +0000] 
Eliminate defenumeration

19 months agoMerge branch 'experimental' into cffi-fsbv
Liam M. Healy [Mon, 24 Oct 2011 00:33:42 +0000] 
Merge branch 'experimental' into cffi-fsbv

Conflicts:
init/funcallable.lisp
linear-algebra/blas2.lisp
linear-algebra/blas3.lisp
solve-minimize-fit/linear-least-squares.lisp
special-functions/coulomb.lisp

19 months agoLibrary path for Darwin + CCL per Yuan MEI
Liam M. Healy [Tue, 18 Oct 2011 03:11:33 +0000] 
Library path for Darwin + CCL per Yuan MEI

19 months agoImprove wording and grammar in docstrings
Liam M. Healy [Wed, 28 Sep 2011 17:36:47 +0000] 
Improve wording and grammar in docstrings

20 months agoUse generic functions for histogram
Liam M. Healy [Mon, 12 Sep 2011 22:03:52 +0000] 
Use generic functions for histogram

Define methods for existing generic functions for histogram:
grid:dimensions (replacing #'bins), set-zero (replacing #'reset).
Also, add a comment that the functions for 2D histogram have not been
defined as a "to be done" reminder.

21 months agoMove asdf-system-connections into :defsystem-depends-on
Liam M. Healy [Fri, 26 Aug 2011 14:37:40 +0000] 
Move asdf-system-connections into :defsystem-depends-on

21 months agoRestore connections in asd file
Liam M. Healy [Fri, 26 Aug 2011 02:37:34 +0000] 
Restore connections in asd file

Apparently asdf-system-connections requires must be loaded _before_
the system is loaded; trying to make it a dependent system seems to
cause it to drop the main system directory name in the path.  To make
this load on quicklisp, put an explicit load of
asdf-system-connections in the asd file.

21 months agoMoved connected system definitions to a new file gsll-connections.lisp
Liam M. Healy [Tue, 23 Aug 2011 13:57:21 +0000] 
Moved connected system definitions to a new file gsll-connections.lisp

21 months agoMerge branch 'master' of git://repo.or.cz/gsll
soemraws [Sat, 20 Aug 2011 13:12:52 +0000] 
Merge branch 'master' of git://repo.or.cz/gsll

21 months agoChange shadowing of 'acceleration
Liam M. Healy [Sat, 20 Aug 2011 03:40:32 +0000] 
Change shadowing of 'acceleration

CCL gets a conflict between the type antik:acceleration and the class
gsll:acceleration, so remove the symbol from the shadowing list of
GSLL and add it to antik::*antik-user-shadow-symbols*.

21 months agoAdded dimensionless physical constants from GSL
soemraws [Sat, 20 Aug 2011 00:38:57 +0000] 
Added dimensionless physical constants from GSL

21 months agoRemove reference to gsll-tests
Liam M. Healy [Fri, 19 Aug 2011 14:16:52 +0000] 
Remove reference to gsll-tests

21 months agoTest ntuple in histogram by writing the data to a temporary file, then reading the...
Liam M. Healy [Fri, 19 Aug 2011 13:51:47 +0000] 
Test ntuple in histogram by writing the data to a temporary file, then reading the file

21 months agoResults of test matrix-product-hermitian as single-floats
Liam M. Healy [Fri, 19 Aug 2011 03:08:56 +0000] 
Results of test matrix-product-hermitian as single-floats

21 months agoRefer to Antik in documentation, delete reference to GSD
Liam M. Healy [Wed, 17 Aug 2011 05:03:13 +0000] 
Refer to Antik in documentation, delete reference to GSD

21 months agoFix test failures due to creation of wrong type of grid from #m macro
Liam M. Healy [Wed, 17 Aug 2011 04:52:46 +0000] 
Fix test failures due to creation of wrong type of grid from #m macro

21 months agoMerge branch 'antik'
Liam M. Healy [Wed, 17 Aug 2011 03:03:28 +0000] 
Merge branch 'antik'

Conflicts:
gsll.asd
ordinary-differential-equations/ode-system.lisp
random/dirichlet.lisp

21 months agoUpdated flysheet analytics code
Liam M. Healy [Wed, 10 Aug 2011 02:40:40 +0000] 
Updated flysheet analytics code

21 months agoDirichlet PDF functions, vector length should be K, not K-1 final-gsd
Liam M. Healy [Fri, 29 Jul 2011 16:23:39 +0000] 
Dirichlet PDF functions, vector length should be K, not K-1

For the Dirichlet PDF functions, the vector length should be K, the
first argument, not K-1.  Reported by Nils Bertschinger.

Conflicts:

random/dirichlet.lisp

21 months agoDirichlet PDF functions, vector length should be K, not K-1
Liam M. Healy [Fri, 29 Jul 2011 16:15:58 +0000] 
Dirichlet PDF functions, vector length should be K, not K-1

For the Dirichlet PDF functions, the vector length should be K, the
first argument, not K-1.  Reported by Nils Bertschinger.

2 years agoUse grid:aref instead of grid:gref
Liam M. Healy [Thu, 26 May 2011 21:56:22 +0000] 
Use grid:aref instead of grid:gref

2 years agoClarify with-ode-integration used for scalar dependent variables
Liam M. Healy [Wed, 4 May 2011 14:26:13 +0000] 
Clarify with-ode-integration used for scalar dependent variables

2 years agoNew arglist for grid:ensure-foreign-array; replace #'dot calls with #'grid:inner
Liam M. Healy [Sat, 23 Apr 2011 22:12:47 +0000] 
New arglist for grid:ensure-foreign-array; replace #'dot calls with #'grid:inner

2 years agoExport 'next step for ODE
Liam M. Healy [Fri, 8 Apr 2011 13:20:36 +0000] 
Export 'next step for ODE

2 years agoAdd :description and :long-description to the gsll.asd file
Liam M. Healy [Sun, 27 Feb 2011 01:45:11 +0000] 
Add :description and :long-description to the gsll.asd file

2 years agoMerge branch 'antik' of ssh://repo.or.cz/srv/git/gsll into antik
Liam M. Healy [Tue, 22 Feb 2011 17:06:27 +0000] 
Merge branch 'antik' of ssh://repo.or.cz/srv/git/gsll into antik

Conflicts:
histogram/ntuple-example.dat

removed

2 years agoRemove ntuple-example.dat
Liam M. Healy [Tue, 22 Feb 2011 17:05:28 +0000] 
Remove ntuple-example.dat

2 years agoRenamed grid:ensure-foreign-array from make-foreign-array-or-default
Liam M. Healy [Sat, 19 Feb 2011 18:09:29 +0000] 
Renamed grid:ensure-foreign-array from make-foreign-array-or-default

2 years agoDefine methods for #'grid:inner instead of #'dot
Liam M. Healy [Sat, 12 Feb 2011 22:27:55 +0000] 
Define methods for #'grid:inner instead of #'dot

2 years agoRemove invert-matrix to Antik, optional argument for LU-decomposition
Liam M. Healy [Sun, 30 Jan 2011 16:39:49 +0000] 
Remove invert-matrix to Antik, optional argument for LU-decomposition

The user-friendly ("higher") functions layered on GSLL are now in
Antik's math-high system, so the antik directory has been removed, and
linear-algebra.lisp is enhanced and in Antik. #'LU-invert now takes
'inverse as an optional argument, with the sensible default.

2 years agoDefine gsll-tests as a system connection
Liam M. Healy [Sun, 30 Jan 2011 15:18:21 +0000] 
Define gsll-tests as a system connection

Define gsll-tests as a system connection, rather than a separate ASDF
system.  This means that if lisp-unit and GSLL are loaded (in either
order), then the tests will automatically be available.

2 years agoIn actual-array-class, don't replace real class names
Liam M. Healy [Sun, 30 Jan 2011 02:59:24 +0000] 
In actual-array-class, don't replace real class names

In actual-array-class, if 'category is not one of 'vector, 'matrix or
'both, then assume it's actually a class name and just return it
without trying to make a class name.  This fixes a problem with
e.g. elt+ where the second argument specializer is a class name of the
scalar.

2 years agoRemove dependence on foreign-array-tests, which is obsolete
Liam M. Healy [Tue, 25 Jan 2011 17:12:36 +0000] 
Remove dependence on foreign-array-tests, which is obsolete

2 years agoAntik user package symbol shadowing and package use
Liam M. Healy [Thu, 13 Jan 2011 15:29:18 +0000] 
Antik user package symbol shadowing and package use

Define symbols to be shadowed and packages to be used for convenience
and conflict resolution in the Antik user packages.

2 years agoSingle-float tests have explicit single-float numbers (finish)
Liam M. Healy [Wed, 12 Jan 2011 14:15:28 +0000] 
Single-float tests have explicit single-float numbers (finish)

2 years agoDon't import grid:element-type, grid:foreign-array, and grid:matrix
Liam M. Healy [Wed, 12 Jan 2011 05:51:43 +0000] 
Don't import grid:element-type, grid:foreign-array, and grid:matrix

Don't import grid:element-type, grid:foreign-array, and grid:matrix,
but rather refer to them with package prefix explicitly.
Superficially checked but requires more careful check.

2 years agoDon't import 'grid:dimensions, use alexandria:make-keyword
Liam M. Healy [Wed, 12 Jan 2011 00:13:03 +0000] 
Don't import 'grid:dimensions, use alexandria:make-keyword

2 years agoExplicit package reference to grid:dimensions
Liam M. Healy [Mon, 10 Jan 2011 23:34:03 +0000] 
Explicit package reference to grid:dimensions

Explicit package reference to grid:dimensions but there's an error
because if the symbol isn't imported many more tests fail.

2 years agoExplicit package home in grid:foreign-pointer
Liam M. Healy [Mon, 10 Jan 2011 23:01:48 +0000] 
Explicit package home in grid:foreign-pointer

2 years agoSingle-float tests have explicit single-float numbers (continued)
Liam M. Healy [Mon, 10 Jan 2011 23:01:17 +0000] 
Single-float tests have explicit single-float numbers (continued)

2 years agoShadow symbols from antik
Liam M. Healy [Mon, 10 Jan 2011 15:35:47 +0000] 
Shadow symbols from antik

With the export of unit and dimension names, there are conflicts with
gsl:psi, gsl:acceleration, gsl:knots.  There is no actual conflict in
the usage because Antik physical quantities only needs to use these
symbols as symbols, there is no binding.  So they are now
shadowing-imported into gsll.

Added explicit "d0" to numbers in polynomial to insure they are read
as double-floats.

2 years agoSingle-float tests have explicit single float numbers (start)
Liam M. Healy [Fri, 7 Jan 2011 23:12:59 +0000] 
Single-float tests have explicit single float numbers (start)

Change literal floats in tests from an unspecified precision to
explicit single floats, so that if *read-default-float-format* is
something other than 'single-float the tests will still pass.  About
half the failures now fixed.  Function #'cdot is returning a pointer
instead of a numerical value.

2 years agoImport symbols from gsll into antik-user
Liam M. Healy [Sat, 1 Jan 2011 21:17:41 +0000] 
Import symbols from gsll into antik-user

Import symbols from the gsll package into the antik-user package,
managing conflicts.

2 years agoMove invert-matrix to antik/linear-algebra.lisp
Liam M. Healy [Sat, 1 Jan 2011 16:49:35 +0000] 
Move invert-matrix to antik/linear-algebra.lisp

2 years agoMethods for arithmetic functions on foreign-arrays
Liam M. Healy [Sat, 1 Jan 2011 16:31:40 +0000] 
Methods for arithmetic functions on foreign-arrays

Depend on Antik instead of GSD.  Add new methods for Antik functions
on foreign-arrays.  This permits algebra-style computation on arrays,
e.g.,
(in-package :antik-user)
(setf grid:*default-grid-type* 'grid:foreign-array)
(expt (* (rotate-3d :z #_15_deg) (rotate-3d :z #_45_deg)) -1)
#m((0.5 0.8660254037844386 0.0) (-0.8660254037844385 0.5000000000000001 -0.0)
   (0.0 0.0 1.0))
(rotate-3d :z #_-60_deg)
#m((0.5000000000000001 0.8660254037844386 0.0)
   (-0.8660254037844386 0.5000000000000001 0.0) (0.0 0.0 1.0))

2 years agoUse fsbv:object instead of cffi:mem-aref
Liam Healy [Sun, 19 Dec 2010 21:14:09 +0000] 
Use fsbv:object instead of cffi:mem-aref

2 years agoEliminate scref and use of grid:dcref
Liam Healy [Sun, 19 Dec 2010 20:58:25 +0000] 
Eliminate scref and use of grid:dcref

2 years agoUse new macros from FSBV
Liam Healy [Sun, 19 Dec 2010 16:43:39 +0000] 
Use new macros from FSBV

Use new macros define-equivalent-type, defconvert, and defenumeration
from FSBV.

2 years agoUse fsbv:object to replace fsbv:convert-from-pointer
Liam M. Healy [Thu, 16 Dec 2010 17:24:59 +0000] 
Use fsbv:object to replace fsbv:convert-from-pointer

2 years agoDefine cl-convert-form in terms of fsbv:convert-from-pointer
Liam Healy [Sun, 12 Dec 2010 05:27:34 +0000] 
Define cl-convert-form in terms of fsbv:convert-from-pointer

2 years agoUse fsbv:defconvert to define sf-result and sf-result-e10
Liam Healy [Sat, 4 Dec 2010 22:47:22 +0000] 
Use fsbv:defconvert to define sf-result and sf-result-e10

Use fsbv:defconvert to define sf-result and sf-result-e10, and use
#'fsbv:object instead of #'val, #'err, #'e10, which have been
eliminated.

2 years agoIntroduce values-unless-singleton for return values in defmfun expansion
Liam Healy [Sat, 4 Dec 2010 15:03:36 +0000] 
Introduce values-unless-singleton for return values in defmfun expansion

New function values-unless-singleton will wrap the forms in 'values if
there are more than one, otherwise it just returns the form.  This is
used in the return value(s) in the defmfun expansion (by way of
body-expand) so that if the returned form returns multiple values,
they all come back from the defmfunned function (note that (values
(values ...)) removes only returns the first value.

2 years agoEliminate calls to grid:complex-to-cl
Liam Healy [Sun, 28 Nov 2010 03:53:34 +0000] 
Eliminate calls to grid:complex-to-cl

Eliminate calls to grid:complex-to-cl because it has been eliminated;
rely on fsbv:object to do the conversion.  Use new function
#'fsbv:converter-defined-p.

2 years agoUse fsbv:object for conversion if available; fix results of dot, cdot tests
Liam Healy [Sat, 27 Nov 2010 18:59:27 +0000] 
Use fsbv:object for conversion if available; fix results of dot, cdot tests

In cl-convert-form, use fsbv:object on foreign structs if FSBV has
been loaded.  Previously, the conversion of (complex single-float) was
wrong because it assumed numbers were double-float; the test results
for dot and cdot have been fixed.

2 years agoAdded comments for maref
Liam Healy [Sat, 27 Nov 2010 17:16:58 +0000] 
Added comments for maref

2 years agoNew macros #'maref and #'(setf maref)
Liam Healy [Thu, 25 Nov 2010 05:11:20 +0000] 
New macros #'maref and #'(setf maref)

New macros #'maref and #'(setf maref) get and set elements of the GSL
array directly from the mpointer.  These are not used by users, but in
the make-funcallable-form expansion of callbacks when scalarsp = T is
specified.  This fixes an error in the argument order in the (setf
get-value) generic function that this replaces.

2 years agoNew macro access-value
Liam M. Healy [Wed, 24 Nov 2010 20:27:01 +0000] 
New macro access-value

New macro access-value to call gsl_*_set and gsl_*_get, as yet unused.
It will eventually replace get-value and (setf get-value) as used in
reference-foreign-element, but has the arguments in the right order
and because it's a macro, expands immediately to a
cffi:foreign-funcall instead of doing a generic function dispatch on
an eql spcializer of the class name.

2 years agoRevert the histogram change
Liam Healy [Sun, 14 Nov 2010 22:59:54 +0000] 
Revert the histogram change

Revert the last histogram change because GSD has been changed so that
the gref compiler macro does not take effect except for foreign-array
types.

2 years agoDefinition of grid:gref* for histogram
Liam Healy [Sun, 14 Nov 2010 00:33:44 +0000] 
Definition of grid:gref* for histogram

Define method grid:gref* for histograms; it was a definition of
grid:gref, but with the compiler macro that turns that into
grid:gref*, we need it to be a definition of grid:gref* instead.
Removed the definition of +foreign-array-pointer+ and
+foreign-array-type+ to the grid package.

2 years agoInstructions to use quicklisp and how to send patches
Liam M. Healy [Tue, 12 Oct 2010 21:48:37 +0000] 
Instructions to use quicklisp and how to send patches

2 years agoAdd tests for rank-1-update
Liam Healy [Sun, 3 Oct 2010 22:15:44 +0000] 
Add tests for rank-1-update

2 years agoMerge branch 'master' of ssh://repo.or.cz/srv/git/gsll
Liam Healy [Sun, 3 Oct 2010 21:40:54 +0000] 
Merge branch 'master' of ssh://repo.or.cz/srv/git/gsll

2 years agoFixed EBADLEN error in `rank-1-update' and `conjugate-rank-1-update'
James Wright [Wed, 29 Sep 2010 23:51:11 +0000] 
Fixed EBADLEN error in `rank-1-update' and `conjugate-rank-1-update'

2 years agoAdded non-radix-2 tests for radix-2-sized vectors
Sumant Oemrawsingh [Fri, 1 Oct 2010 20:13:01 +0000] 
Added non-radix-2 tests for radix-2-sized vectors

Effectively, non-radix-2 algorithm tests on power-of-2 vector sizes are added.
The result is, that some tests fail, and the cause at this point is not yet
clear. The tests that fail, are all related to non-radix-2 algorithms working
on power-of-2 vector sizes.

Only tests that are missing, are the so-called bitreverse tests.

2 years agoChanged off-stride checks to be only one assertion
Sumant Oemrawsingh [Wed, 15 Sep 2010 21:18:57 +0000] 
Changed off-stride checks to be only one assertion

Previously, each element that was checked off-stride would count as an
assertion.

2 years agoFixed offset for vector initialisation to better match GSL's tests.
Sumant Oemrawsingh [Wed, 15 Sep 2010 20:54:49 +0000] 
Fixed offset for vector initialisation to better match GSL's tests.

2 years agoFixed FFT tests, both on- and off-stride.
Sumant Oemrawsingh [Wed, 15 Sep 2010 20:48:55 +0000] 
Fixed FFT tests, both on- and off-stride.

For stride > 1 elements, the results are compared only on-stride, while
(for complex only) it is verified that off-stride elements are left alone.

Tests are still disabled due to time it takes to complete tests.

Tests are not yet complete wrt GSL's FFT tests.

2 years agoAdded function #'off-stride to make duplicate array with only the off-stride elements
Liam Healy [Sat, 4 Sep 2010 21:09:10 +0000] 
Added function #'off-stride to make duplicate array with only the off-stride elements

2 years agoUse vector/length for all versions of GSL in testing FFT
Liam Healy [Sat, 4 Sep 2010 12:54:05 +0000] 
Use vector/length for all versions of GSL in testing FFT

Since the function vector/length no longer relies on GSL for doing the
division, it works for all versions of GSL and whether or not FSBV is
loaded.  This fixes a failure for those versions where
:initial-contents was getting an array and not a list.

2 years agoExpected number of assertions fail consistently in fast-fourier-transform.
Sumant Oemrawsingh [Thu, 2 Sep 2010 22:07:35 +0000] 
Expected number of assertions fail consistently in fast-fourier-transform.

There was one case left where even out-of-stride elements were scaled as well.
Tests are still not enabled.

2 years agoFix for random errors in tests
Sumant Oemrawsingh [Wed, 1 Sep 2010 21:11:21 +0000] 
Fix for random errors in tests

The test functions did not properly initialize newly created vectors to
contain unique values. This caused comparison of elements that are out of the
stride to randomly succeed. The new test functions properly initialize the
vectors just like is done in the GSL test functions. The result is that
comparison of elements that are out of stride should fail consistently.

This automatically means that the comparison should only check elements which
are on stride, just like is done in GSL. This is not fixed yet, so the
fast-fourier-transform unit tests are still left disabled for now.

2 years agoFix FFT to actually do something, but comment out because of failures
Liam Healy [Mon, 23 Aug 2010 02:18:03 +0000] 
Fix FFT to actually do something, but comment out because of failures

The FFT tests were testing nothing of consequence, just that NIL was
equal to NIL, because it was looking at the result of (reset-urand)
and then three non-existent multiple values.  This is now fixed, so
that fft-*-result-check actually binds the results of the
calculations.  Also, single-float tests are referenced to
+sgl-epsilon+ and not +dbl-epsilon+.  However, for
(all-fft-test-forms 9 3 (64 99)), many tests fail:
FAST-FOURIER-TRANSFORM: 235 assertions passed, 55 failed.
so the tests are commented out.

2 years agoNew size-vector-scalar replaces size-vector-real
Liam Healy [Sat, 21 Aug 2010 21:32:45 +0000] 
New size-vector-scalar replaces size-vector-real

New function size-vector-scalar replaces size-vector-real, and gives
the size as a double float if the array is real, and as the element
type if it is complex, in order to be acceptable to the elt* interface
to GSL's _scale routines.

2 years agoFix simulated annealing for new grid:copy
Liam M. Healy [Thu, 19 Aug 2010 20:43:59 +0000] 
Fix simulated annealing for new grid:copy

2 years agoImproved finding of libraries for OS X, by Leo
Liam Healy [Tue, 17 Aug 2010 03:11:18 +0000] 
Improved finding of libraries for OS X, by Leo

2 years agoAdd FFT tests
Liam Healy [Sat, 14 Aug 2010 22:35:21 +0000] 
Add FFT tests

Add FFT tests; these are a subset of the GSL tests.  For some reason
the tests take a lot longer in GSLL, so instead of testing all sizes
1...99, we test only 1...9 and then 64 and 99 for stride 1 only.

2 years agoPut :unix at the end of load-foreign-library list
Liam M. Healy [Fri, 13 Aug 2010 17:56:16 +0000] 
Put :unix at the end of load-foreign-library list

Put :unix at the end of load-foreign-library list so that more
specific unices like :darwin get seen first.

2 years agoGrovel file depends on init
Liam M. Healy [Fri, 13 Aug 2010 15:40:05 +0000] 
Grovel file depends on init

2 years agoInvert the imaginary part in create-complex-matrix
Liam Healy [Fri, 13 Aug 2010 02:52:43 +0000] 
Invert the imaginary part in create-complex-matrix

The imaginary part of the elements in create-complex-matrix should be
inverted to match GSL's create_complex_matrix.  Now all LU tests
pass.

2 years agoWorkaround CLISP bug
Liam M. Healy [Wed, 11 Aug 2010 14:05:34 +0000] 
Workaround CLISP bug

CLISP does not recognize that an uninterned symbol in different
parts of a progn is the same;
http://sourceforge.net/tracker/?func=detail&aid=836838&group_id=1355&atid=101355
Since this bug was "long-standing" the better part of a decade ago, we
will assume it now has tenure and change the progn to let.

2 years agoAdd dependencies on mathematical
Liam Healy [Tue, 10 Aug 2010 02:27:04 +0000] 
Add dependencies on mathematical

2 years agoReplace 'pi with 'dpi, the double-float version
Liam Healy [Sun, 8 Aug 2010 01:50:45 +0000] 
Replace 'pi with 'dpi, the double-float version

The standard declares that 'pi is an implementation-dependent
long-float, but GSL requires double float, so the constant 'dpi has
been defined as (coerce pi 'double-float).  This should make CLISP at
the least happier.

2 years agoUpdated user information
Liam M. Healy [Thu, 22 Jul 2010 21:47:38 +0000] 
Updated user information

2 years agoUpdate status.text, last element of list
Liam M. Healy [Wed, 21 Jul 2010 17:35:37 +0000] 
Update status.text, last element of list

Update status.text mostly be removing obsolete information without
replacing it with current information.  Use alexandria:lastcar for the
last element of list so I do not use private definitions (non-loaded
source).

2 years agoName conditions generic-failure-* differently
Liam M. Healy [Tue, 20 Jul 2010 22:02:16 +0000] 
Name conditions generic-failure-* differently

Name conditions generic-failure-1 and generic-failure-2 because CCL
doesn't like the duplicate differently.  Withdraw the
conditionalization away of +nan+, +positive-infinity+, and
+negative-infinity+ for CCL because the latest version
"Version 1.6-dev-r13994M-trunk  (LinuxX8664)"
can handle them.

2 years agoCCL don't define +nan+, +positive-infinity+, +negative-infinity+
Liam M. Healy [Mon, 19 Jul 2010 16:38:28 +0000] 
CCL don't define +nan+, +positive-infinity+, +negative-infinity+

In CCL, +nan+, +positive-infinity+, +negative-infinity+ are not
defined because of a problem with CCL loading these definitions.

2 years agoMerge branch 'convert'
Liam Healy [Mon, 19 Jul 2010 02:54:19 +0000] 
Merge branch 'convert'

Trivial conflicts:

data/array-structs.lisp
init/init.lisp