Wrote spothelpbrowser.m script to take matlab published examples from…

… spotdoc and format them for the help browser

doc/guide_circulant.m → doc/docguide_circulant.m

@@ -1,4 +1,4 @@
-%% Building a Circulant Operator
+%% Building a circulant operator
 % In our first example we will use Spot to create an implicit
 % circulant matrix and that can be used as a fast operator.  Circulant
 % matrices are fully specified by their first column, and each
@@ -29,7 +29,8 @@
 c = (1:n)';  % make sure c is a column vector
 
 %%
-% Our very first Spot command will create the required DFT operator.
+% Our very first Spot command will create the required
+% <http://www.cs.ubc.ca/labs/scl/spot/operators.html#opDFT DFT operator>.
 % We omit the semicolon so that Matlab will display details of the
 % resulting operator:
 
@@ -60,8 +61,8 @@
 C = real( F'*opDiag(d)*F )
 
 %%
-% We could have left out the |real| modifier, and used the simpler
-% command
+% We could have left out the <matlab:doc('opSpot/real') real>
+% modifier, and used the simpler command
 %
 %   C = F'*opDiag(d)*F
 %
@@ -111,3 +112,4 @@
 % simultaneously to a collection of vectors:
 
 C*eye(n)
+

doc/guide_containers.m → doc/docguide_containers.m

@@ -1,4 +1,4 @@
-%% Using Spot Containers
+%% Using Spot containers
 % Spot provides a set of "wrappers" that can be used to integrate
 % external opertors with Spot utilities.
 
@@ -15,7 +15,7 @@
 
 %%
 % In order to make this function available as a Spot operator, we wrap it
-% using <matlab:doc('opFunction') opFunction>:
+% using <http://www.cs.ubc.ca/labs/scl/spot/operators.html\#opFunction/ |opFunction|>:
 
 n = 5; m = 5;
 A = opFunction(n,m,@heaviside);
@@ -43,9 +43,10 @@
 
 %%
 % Though in this special case, we might have as well used the Spot operator
-% <matlab:doc('opDiag') opDiag> instead of the more general (and in this
-% case, cumbersome) <matlab:doc('opMatrix') opMatrix>(diag(1:m)).
+% <http://www.cs.ubc.ca/labs/scl/spot/operators.html\#opDiag/ |opDiag|> instead of the more general (and in this
+% case, cumbersome) <http://www.cs.ubc.ca/labs/scl/spot/operators.html\#opMatrix/ |opMatrix|>|(diag(1:m))|.
 
 %% Class containers
-% Finally, the <matlab:doc('opClass') opClass> container can be used to encapsulate
+% Finally, the <http://www.cs.ubc.ca/labs/scl/spot/operators.html\#opClass/ |opClass|> container can be used to encapsulate
 % objects defined by external toolboxes.
+

doc/guide_convolution.m → doc/docguide_convolution.m

@@ -1,4 +1,4 @@
-%% Blurring and Deblurring an Image with opConvolve
+%% Blurring and deblurring with opConvolve
 % The convolution operator is often used in signal processing as a
 % technique for filtering and noise removal. In image processing, for
 % example, convolution can be used to blur images or remove noise. First, a
@@ -21,7 +21,7 @@
 %
 % |opConvolve(M,N,KERNEL,OFFSET,MODE)|
 %
-% This example will demonstrate how to blur an image using opConvolve and
+% This example will demonstrate how to blur an image using |opConvolve| and
 % two different types of kernel. We will then deblur it using the same operator.
 
 %% Using a Uniform Kernel
@@ -75,7 +75,7 @@
 A = opConvolve(m, n, K, center, 'truncated');
 
 %%
-% The third argument to this psfGaussian function determines the standard deviation of
+% The third argument to this |psfGaussian| function determines the standard deviation of
 % the Gaussian distribution, or how quickly the values diminish to zero.
 % Therefore, using a larger standard deviation will make a "larger" kernel,
 % and cause more blur. Compare the distribution we've used (first image) to
@@ -177,4 +177,4 @@
 %%
 % This means that when you create your convolution operator with |A =
 % Finv*S*F|, where |S| is some scaling operator, the same efficient |F'| operation
-% will be used whether you write |inv(F)*S*F|, |F'*S*F|, or |F\S*F|.
+% will be used whether you write |inv(F)*S*F|, |F'*S*F|, or |F\S*F|.

doc/guide_quick.m → doc/docguide_quick.m

@@ -1,4 +1,4 @@
-%% A Quick Guide to Spot
+%% A quick guide to Spot 
 % Using explicit matrices is not practical for some very large problems.
 % Instead, we can use Spot operators. A Spot operator represents a matrix,
 % and can be treated in a similar way, but it doesn't rely on the matrix
@@ -15,8 +15,8 @@
 %%
 % We see that Spot displays some information about A when we leave off the
 % semicolon, such as its construction, the number of rows and columns, and
-% whether it is complex. We can also use the <matlab:doc('opSpot/double') double> method to construct the
-% underlying matrix:
+% whether it is complex. We can also use the <matlab:doc('opSpot/double') double>
+% method to construct the underlying matrix:
 
 double(A)
 
@@ -31,8 +31,10 @@
 %%
 % If you add a matrix and an operator, Spot will first automatically
 % convert the matrix into a Spot operator using opMatrix. We can discover
-% other information about the operator C by using the methods <matlab:doc('opSpot/size') size>,
-% <matlab:doc('opSpot/disp') disp>, and <matlab:doc('opSpot/whos') whos>:
+% other information about the operator C by using the methods
+% <matlab:doc('opSpot/size') size>,
+% <matlab:doc('opSpot/disp') disp>,
+% and <matlab:doc('opSpot/whos') whos>:
 
 whos C
 
@@ -80,9 +82,9 @@
 
 %% Creating More Complex Operators
 % Spot operators can be combined into more complex operators using methods
-% such as <matlab:doc('opSpot/blkdiag') blkdiag> and <matlab:doc('opSpot/kron') kron>.
-% Whenever a matrix is passed to one of these
-% methods, it is automatically converted to a Spot operator. blkdiag takes
+% such as blkdiag and kron. Whenever a matrix is passed to one of these
+% methods, it is automatically converted to a Spot operator.
+% <matlab:doc('opSpot/blkdiag') blkdiag> takes
 % a list of operators and matrices and creates a block diagonal operator:
 
 A = opOnes(2,2);
@@ -93,18 +95,17 @@
 
 %%
 % We can also have the blocks overlap and create anti-diagonal operators
-% (see <usingmethods.html "Using methods on operators">). Operators can be horizontally or
-% vertically concatenated using <matlab:doc('opDictionary') opDictionary>
-% or <matlab:doc('opStack') opStack>, or simply by passing them as elements
-% in a matrix:
+% (see <usingmethods.html "Using the Methods">).
+% Operators can be horizontally or vertically concatenated using
+% opDictionary or opStack, or simply by passing them as elements in a matrix:
 
 E = 4*opOnes(2,3);
 F = [A E];
 double(F)
 
 %%
-% The <matlab:doc('opSpot/kron') kron> method computes the Kronecker product of an arbitrary number of
-% operators:
+% The <matlab:doc('opSpot/kron') kron> method
+% computes the Kronecker product of an arbitrary number of operators:
 
 G = opMatrix([2 1;3 0]);
 H = opMatrix([1 2]);
@@ -113,6 +114,9 @@
 
 %% 
 % For more information on how to work with Spot operators, see
-% <usingmethods.html "Using methods on operators">. For a list of the Spot operator classes and
-% what they do, including fast transformations, random ensembles, and
-% convolution, see the <spotoperators.html "Operators">.
+% <usingmethods.html "Using the Methods">.
+% For a list of methods, see the <http://www.cs.ubc.ca/labs/scl/spot/methods.html "Index of Methods">.
+% For a list of the Spot operator classes and what they do, including fast
+% transformations, random ensembles, and convolution, see the
+% <http://www.cs.ubc.ca/labs/scl/spot/operators.html "Index of Operators">.
+

doc/guide_sparserecovery.m → doc/docguide_sparserecovery.m

@@ -71,7 +71,8 @@
 m = 120;
 
 %%
-% Instead of an explicit matrix, we can use a Spot opGaussian operator.
+% Instead of an explicit matrix, we can use a Spot
+% <http://www.cs.ubc.ca/labs/scl/spot/operators.html#opGaussian |opGuassian|> operator.
 % Using mode 3 of opGaussian creates an implicit Gaussian matrix with
 % unit-norm columns (EXPLAIN). This
 % means that the columns of the matrix are generated as the operator is
@@ -119,4 +120,4 @@
 
 %%
 % The signal and reconstruction overlap almost completely, so our
-% reconstruction is accurate.
+% reconstruction is accurate.

doc/examples.m

@@ -5,4 +5,4 @@
 % * <guide_circulant.html Building a circulant operator>
 % * <guide_containers.html Using Spot containers>
 % * <guide_convolution.html Blurring and deblurring with opConvolve>
-% * <guide_sparserecovery.html Sparse Recovery>
+% * <guide_sparserecovery.html Sparse recovery>

doc/helptoc.xml

@@ -3,7 +3,7 @@
 <toc version="1.0">
 
 <tocitem target="html/spot_main_page.html">Spot Linear Operator Toolbox
-    <tocitem target="html/guide_quick.html"
+    <tocitem target="html/docguide_quick.html"
              image="HelpIcon.GETTING_STARTED">
              A quick guide to Spot
     </tocitem>
@@ -39,16 +39,16 @@
     <tocitem target="html/examples.html"
              image="HelpIcon.EXAMPLES">
              Examples
-             <tocitem target="html/guide_circulant.html">
+             <tocitem target="html/docguide_circulant.html">
                       Building a circulant operator
              </tocitem>
-             <tocitem target="html/guide_containers.html">
+             <tocitem target="html/docguide_containers.html">
                       Using Spot containers
              </tocitem>
-             <tocitem target="html/guide_convolution.html">
+             <tocitem target="html/docguide_convolution.html">
                       Blurring and deblurring with opConvolve
              </tocitem>
-             <tocitem target="html/guide_sparserecovery.html">
+             <tocitem target="html/docguide_sparserecovery.html">
                       Sparse Recovery
              </tocitem>
     </tocitem>

doc/html/guide_circulant.html → doc/html/docguide_circulant.html

@@ -6,7 +6,7 @@
    <!--
 This HTML was auto-generated from MATLAB code.
 To make changes, update the MATLAB code and republish this document.
-      --><title>Building a Circulant Operator</title><meta name="generator" content="MATLAB 7.12"><link rel="schema.DC" href="http://purl.org/dc/elements/1.1/"><meta name="DC.date" content="2012-07-10"><meta name="DC.source" content="guide_circulant.m"><style type="text/css">
+      --><title>Building a circulant operator</title><meta name="generator" content="MATLAB 7.12"><link rel="schema.DC" href="http://purl.org/dc/elements/1.1/"><meta name="DC.date" content="2012-07-12"><meta name="DC.source" content="docguide_circulant.m"><style type="text/css">
 
 body {
   background-color: white;
@@ -62,9 +62,9 @@
   color: gray;
 }
 
-  </style></head><body><div class="content"><h1>Building a Circulant Operator</h1><!--introduction--><p>In our first example we will use Spot to create an implicit circulant matrix and that can be used as a fast operator.  Circulant matrices are fully specified by their first column, and each remaining column is a cyclic permutation of the first:</p><p><img src="guide_circulant_eq08112.png" alt="$$&#xA;C =  \pmatrix{&#xA;       c_1   &amp; c_{n}      &amp; \ldots &amp; c_2&#xA;   \cr c_2   &amp; c_1        &amp; \ldots &amp; \vdots&#xA;   \cr \vdots&amp; \ddots     &amp; \ddots &amp; c_n&#xA;   \cr c_n   &amp; \ldots     &amp; c_2    &amp; c_1&#xA;    }.&#xA;$$"></p><p>These matrices have the property that they can be diagonalized by the discrete Fourier transform (DFT), so that</p><pre> C = F'diag(d)F,</pre><p>where d = Fc. An important implication is that <tt>C</tt> can be treated as a fast operator.</p><!--/introduction--><h2>Contents</h2><div><ul><li><a href="#1">Instantiating a simple Spot operator</a></li><li><a href="#6">Building more complex operators</a></li></ul></div><h2>Instantiating a simple Spot operator<a name="1"></a></h2><p>In our toy example, we create a circulant matrix whose first column is simply the sequence 1,...,5:</p><pre class="codeinput">n = 5;
+  </style></head><body><div class="content"><h1>Building a circulant operator</h1><!--introduction--><p>In our first example we will use Spot to create an implicit circulant matrix and that can be used as a fast operator.  Circulant matrices are fully specified by their first column, and each remaining column is a cyclic permutation of the first:</p><p><img src="docguide_circulant_eq08112.png" alt="$$&#xA;C =  \pmatrix{&#xA;       c_1   &amp; c_{n}      &amp; \ldots &amp; c_2&#xA;   \cr c_2   &amp; c_1        &amp; \ldots &amp; \vdots&#xA;   \cr \vdots&amp; \ddots     &amp; \ddots &amp; c_n&#xA;   \cr c_n   &amp; \ldots     &amp; c_2    &amp; c_1&#xA;    }.&#xA;$$"></p><p>These matrices have the property that they can be diagonalized by the discrete Fourier transform (DFT), so that</p><pre> C = F'diag(d)F,</pre><p>where d = Fc. An important implication is that <tt>C</tt> can be treated as a fast operator.</p><!--/introduction--><h2>Contents</h2><div><ul><li><a href="#1">Instantiating a simple Spot operator</a></li><li><a href="#6">Building more complex operators</a></li></ul></div><h2>Instantiating a simple Spot operator<a name="1"></a></h2><p>In our toy example, we create a circulant matrix whose first column is simply the sequence 1,...,5:</p><pre class="codeinput">n = 5;
 c = (1:n)';  <span class="comment">% make sure c is a column vector</span>
-</pre><p>Our very first Spot command will create the required DFT operator. We omit the semicolon so that Matlab will display details of the resulting operator:</p><pre class="codeinput">F = opDFT(n)
+</pre><p>Our very first Spot command will create the required <a href="http://www.cs.ubc.ca/labs/scl/spot/operators.html#opDFT">DFT operator</a>. We omit the semicolon so that Matlab will display details of the resulting operator:</p><pre class="codeinput">F = opDFT(n)
 </pre><pre class="codeoutput">F = 
   Spot operator: DFT(5,5)
     rows:      5    complex: yes       
@@ -94,7 +94,7 @@
   Spot operator: Real(DFT(5,5)' * Diag(5,5) * DFT(5,5))
     rows:      5    complex: no        
     cols:      5    type:    real      
-</pre><p>We could have left out the <tt>real</tt> modifier, and used the simpler command</p><pre> C = F'*opDiag(d)*F</pre><p>However, we need to safeguard against numerical errors that might allow complex values to seep in. Hence, we finish off our construction with the <tt>real</tt> operator.</p><p>Just as the DFT operator <tt>F</tt> could be applied to vectors, the same is true of the compound operator <tt>C</tt>. Here we apply it to the first unit vector:</p><pre class="codeinput">e = eye(5,1);
+</pre><p>We could have left out the <a href="matlab:doc('opSpot/real')">real</a> modifier, and used the simpler command</p><pre> C = F'*opDiag(d)*F</pre><p>However, we need to safeguard against numerical errors that might allow complex values to seep in. Hence, we finish off our construction with the <tt>real</tt> operator.</p><p>Just as the DFT operator <tt>F</tt> could be applied to vectors, the same is true of the compound operator <tt>C</tt>. Here we apply it to the first unit vector:</p><pre class="codeinput">e = eye(5,1);
 C*e
 </pre><pre class="codeoutput">
 ans =
@@ -140,7 +140,7 @@
 </pre><p class="footer"><br>
       Published with MATLAB&reg; 7.12<br></p></div><!--
 ##### SOURCE BEGIN #####
-%% Building a Circulant Operator
+%% Building a circulant operator
 % In our first example we will use Spot to create an implicit
 % circulant matrix and that can be used as a fast operator.  Circulant
 % matrices are fully specified by their first column, and each
@@ -171,7 +171,8 @@
 c = (1:n)';  % make sure c is a column vector
 
 %%
-% Our very first Spot command will create the required DFT operator.
+% Our very first Spot command will create the required
+% <http://www.cs.ubc.ca/labs/scl/spot/operators.html#opDFT DFT operator>.
 % We omit the semicolon so that Matlab will display details of the
 % resulting operator:
 
@@ -202,8 +203,8 @@
 C = real( F'*opDiag(d)*F )
 
 %%
-% We could have left out the |real| modifier, and used the simpler
-% command
+% We could have left out the <matlab:doc('opSpot/real') real>
+% modifier, and used the simpler command
 %
 %   C = F'*opDiag(d)*F
 %
@@ -254,5 +255,6 @@
 
 C*eye(n)
 
+
 ##### SOURCE END #####
 --></body></html>