Mercurial > octave-nkf

changeset 7312:1ebbf0924217

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
[project @ 2007-12-12 22:19:54 by jwe]
author jwe
date Wed, 12 Dec 2007 22:22:10 +0000
parents 1b535aed87e2
children 7fd7abd37113
files scripts/ChangeLog scripts/linear-algebra/condest.m
diffstat 2 files changed, 30 insertions(+), 18 deletions(-) [+]
line wrap: on
line diff
--- a/scripts/ChangeLog	Wed Dec 12 22:13:44 2007 +0000
+++ b/scripts/ChangeLog	Wed Dec 12 22:22:10 2007 +0000
@@ -1,8 +1,8 @@
 2007-12-12  John W. Eaton  <jwe@octave.org>
 
 	* linear-algebra/condest.m: Use nargin instead of size(varargin,2).	
-	Condense argument processing logic. Allow 6 arguments.
-	Use issquare.
+	Condense argument processing logic.  Allow 6 arguments.
+	Use issquare.  Avoid calling "exist".
 
 	* plot/__go_draw_axes__.m: Handle the axes layer property.
 
--- a/scripts/linear-algebra/condest.m	Wed Dec 12 22:13:44 2007 +0000
+++ b/scripts/linear-algebra/condest.m	Wed Dec 12 22:22:10 2007 +0000
@@ -111,57 +111,69 @@
 
   default_t = 5;
 
+  have_A = false;
+  have_t = false;
+  have_solve = false;
+
   if (ismatrix (varargin{1}))
     A = varargin{1};
     n = issquare (A);
     if (! n)
       error ("condest: matrix must be square.");
     endif
+    have_A = true;
 
-    if (nargin > 1 && isscalar (varargin{2}))
-      t = varargin{2};
-    elseif (nargin > 2)
-      solve = varargin{2};
-      solve_t = varargin{3};
-      if (nargin > 3)
-	t = varargin{4};
+    if (nargin > 1)
+      if (isscalar (varargin{2}))
+	t = varargin{2};
+	have_t = true;
+      elseif (nargin > 2)
+	solve = varargin{2};
+	solve_t = varargin{3};
+	have_solve = true;
+	if (nargin > 3)
+	  t = varargin{4};
+	  have_t = true;
+	endif
+      else
+	error ("condest: must supply both solve and solve_t.");
       endif
-    else
-      error ("condest: must supply both solve and solve_t.");
     endif
   elseif (nargin > 4)
     apply = varargin{1};
     apply_t = varargin{2};
     solve = varargin{3};
     solve_t = varargin{4};
+    have_solve = true;
     n = varargin{5};
     if (! isscalar (n))
       error ("condest: dimension argument of implicit form must be scalar.");
     endif
     if (nargin > 5)
       t = varargin{6};
+      have_t = true;
     endif
   else
     error ("condest: implicit form of condest requires at least 5 arguments.");
   endif
 
-  if (! exist ("t", "var"))
+  if (! have_t)
     t = min (n, default_t);
   endif
 
-  if (! exist ("solve", "var"))
+  if (! have_solve)
     if (issparse (A))
       [L, U, P, Pc] = splu (A);
-      solve = @(x) Pc' * (U\ (L\ (P*x)));
-      solve_t = @(x) P'*(L'\ (U'\ (Pc*x)));
+      solve = @(x) Pc' * (U \ (L \ (P * x)));
+      solve_t = @(x) P' * (L' \ (U' \ (Pc * x)));
     else
       [L, U, P] = lu (A);
-      solve = @(x) U\ (L\ (P*x));
-      solve_t = @(x) P' * (L'\ (U'\x));
+      solve = @(x) U \ (L \ (P*x));
+      solve_t = @(x) P' * (L' \ (U' \ x));
     endif
   endif
 
-  if (exist ("A", "var"))
+  if (have_A)
     Anorm = norm (A, 1);
   else
     Anorm = onenormest (apply, apply_t, n, t);