Add Octave language with its own language distinct from MATLAB

.gitmodules

@@ -1401,3 +1401,6 @@
 [submodule "vendor/grammars/zephir-sublime"]
 	path = vendor/grammars/zephir-sublime
 	url = https://github.com/phalcon/zephir-sublime
+[submodule "vendor/grammars/Octave-language-grammar"]
+	path = vendor/grammars/Octave-language-grammar
+	url = https://github.com/gnu-octave/Octave-language-grammar

grammars.yml

@@ -87,6 +87,8 @@ vendor/grammars/ObjectScript.tmBundle:
 - source.objectscript_class
 - source.objectscript_csp
 - source.objectscript_macros
+vendor/grammars/Octave-language-grammar:
+- source.octave
 vendor/grammars/PHP-Twig.tmbundle:
 - text.html.twig
 vendor/grammars/PogoScript.tmbundle:

lib/linguist/heuristics.yml

@@ -417,6 +417,14 @@ disambiguations:
   rules:
   - language: Objective-C
     named_pattern: objectivec
+  - language: Octave
+    pattern:
+    - '^\s*##'
+    - '^\s*endfunction'
+    - '^\s*endclassdef'
+    - '^\s*endif'
+    - '^\s*endfor'
+    - '^\s*endwhile'
   - language: Mercury
     pattern: ':- module'
   - language: MUF

lib/linguist/languages.yml

@@ -943,11 +943,11 @@ Caddyfile:
   type: data
   color: "#22b638"
   aliases:
-    - Caddy
+  - Caddy
   extensions:
-    - ".caddyfile"
+  - ".caddyfile"
   filenames:
-    - Caddyfile
+  - Caddyfile
   ace_mode: text
   tm_scope: source.Caddyfile
   language_id: 615465151
@@ -4020,15 +4020,11 @@ M4Sugar:
 MATLAB:
   type: programming
   color: "#e16737"
-  aliases:
-  - octave
   extensions:
   - ".matlab"
   - ".m"
   tm_scope: source.matlab
   ace_mode: matlab
-  codemirror_mode: octave
-  codemirror_mime_type: text/x-octave
   language_id: 225
 MAXScript:
   type: programming
@@ -4884,6 +4880,14 @@ Objective-J:
   tm_scope: source.js.objj
   ace_mode: text
   language_id: 259
+Octave:
+  type: programming
+  color: "#2299c4"
+  extensions:
+  - ".m"
+  tm_scope: source.octave
+  ace_mode: text
+  language_id: 854759768
 Odin:
   type: programming
   color: "#60AFFE"
@@ -5359,7 +5363,7 @@ Pkl:
   extensions:
   - ".pkl"
   interpreters:
-    - pkl
+  - pkl
   tm_scope: source.pkl
   ace_mode: text
   language_id: 288822799

samples/Octave/ClusterCriterion.m

@@ -0,0 +1,245 @@
+## Copyright (C) 2021 Stefano Guidoni <[email protected]>
+## Copyright (C) 2024 Andreas Bertsatos <[email protected]>
+##
+## This file is part of the statistics package for GNU Octave.
+##
+## This program is free software; you can redistribute it and/or modify it under
+## the terms of the GNU General Public License as published by the Free Software
+## Foundation; either version 3 of the License, or (at your option) any later
+## version.
+##
+## This program is distributed in the hope that it will be useful, but WITHOUT
+## ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+## FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more
+## details.
+##
+## You should have received a copy of the GNU General Public License along with
+## this program; if not, see <http://www.gnu.org/licenses/>.
+
+classdef ClusterCriterion < handle
+  ## -*- texinfo -*-
+  ## @deftypefn  {statistics} {@var{obj} =} ClusterCriterion (@var{x}, @var{clust}, @var{criterion})
+  ##
+  ## A clustering evaluation object as created by @code{evalclusters}.
+  ##
+  ## @code{ClusterCriterion} is a superclass for clustering evaluation objects
+  ## as created by @code{evalclusters}.
+  ##
+  ## List of public properties:
+  ## @table @code
+  ## @item @qcode{ClusteringFunction}
+  ## a valid clustering funtion name or function handle.  It can be empty if
+  ## the clustering solutions are passed as an input matric.
+  ##
+  ## @item @qcode{CriterionName}
+  ## a valid criterion name to evaluate the clustering solutions.
+  ##
+  ## @item @qcode{CriterionValues}
+  ## a vector of values as generated by the evaluation criterion for each
+  ## clustering solution.
+  ##
+  ## @item @qcode{InspectedK}
+  ## the list of proposed cluster numbers.
+  ##
+  ## @item @qcode{Missing}
+  ## a logical vector of missing observations.  When there are @code{NaN}
+  ## values in the data matrix, the corresponding observation is excluded.
+  ##
+  ## @item @qcode{NumObservations}
+  ## the number of non-missing observations in the data matrix.
+  ##
+  ## @item @qcode{OptimalK}
+  ## the optimal number of clusters.
+  ##
+  ## @item @qcode{OptimalY}
+  ## the clustering solution corresponding to @code{OptimalK}.
+  ##
+  ## @item @qcode{X}
+  ## the data matrix.
+  ##
+  ## @end table
+  ##
+  ## List of public methods:
+  ## @table @code
+  ## @item @qcode{addK}
+  ## add a list of numbers of clusters to evaluate.
+  ##
+  ## @item @qcode{compact}
+  ## return a compact clustering evaluation object. Not implemented
+  ##
+  ## @item @qcode{plot}
+  ## plot the clustering evaluation values against the corresponding number of
+  ## clusters.
+  ##
+  ## @end table
+  ##
+  ## @seealso{evalclusters, CalinskiHarabaszEvaluation, DaviesBouldinEvaluation,
+  ## GapEvaluation, SilhouetteEvaluation}
+  ## @end deftypefn
+
+  properties (Access = public)
+    ## public properties
+  endproperties
+
+  properties (GetAccess = public, SetAccess = protected)
+    ClusteringFunction = "";
+    CriterionName = "";
+    CriterionValues = [];
+    InspectedK = [];
+    Missing = [];
+    NumObservations = 0;
+    OptimalK = 0;
+    OptimalY = [];
+    X = [];
+  endproperties
+
+  properties (Access = protected)
+    N = 0; # number of observations
+    P = 0; # number of variables
+    ClusteringSolutions = []; #
+    OptimalIndex = 0; # index of the optimal K
+  endproperties
+
+  methods (Access = public)
+
+    ## constructor
+    function this = ClusterCriterion (x, clust, KList)
+      ## parsing input data
+      if ((! ismatrix (x)) || (! isnumeric (x)))
+        error ("ClusterCriterion: 'x' must be a numeric matrix");
+      endif
+      this.X = x;
+      this.N = rows (this.X);
+      this.P = columns (this.X);
+      ## look for missing values
+      for iter = 1 : this.N
+        if (any (find (x(iter, :) == NaN)))
+          this.Missing(iter) = true;
+        else
+          this.Missing(iter) = false;
+        endif
+      endfor
+      ## number of usable observations
+      this.NumObservations = sum (this.Missing == false);
+
+      ## parsing the clustering algorithm
+      if (ischar (clust))
+        if (any (strcmpi (clust, {"kmeans", "linkage", "gmdistribution"})))
+          this.ClusteringFunction = lower (clust);
+        else
+          error ("ClusterCriterion: unknown clustering algorithm '%s'", clust);
+        endif
+      elseif (isa (clust, "function_handle"))
+        this.ClusteringFunction = clust;
+      elseif (ismatrix (clust))
+        if (isnumeric (clust)  && (length (size (clust)) == 2) && ...
+            (rows (clust) == this.N))
+          this.ClusteringFunction = "";
+          this.ClusteringSolutions = clust(find (this.Missing == false), :);
+        else
+          error ("ClusterCriterion: invalid matrix of clustering solutions");
+        endif
+      else
+        error ("ClusterCriterion: invalid argument");
+      endif
+
+      ## parsing the list of cluster sizes to inspect
+      this.InspectedK = parseKList (this, KList);
+    endfunction
+
+    ## -*- texinfo -*-
+    ## @deftypefn {ClusterCriterion} {@var{obj} =} addK (@var{obj}, @var{K})
+    ##
+    ## Add a new cluster array to inspect the ClusterCriterion object.
+    ##
+    ## @end deftypefn
+    function this = addK (this, k)
+
+      ## if there is not a clustering function, then we are using a predefined
+      ## set of clustering solutions, hence we cannot redefine the number of
+      ## solutions
+      if (isempty (this.ClusteringFunction))
+        warning (["ClusterCriterion.addK: cannot redefine the list of cluster"...
+                  "numbers to evaluate when there is not a clustering function"]);
+        return;
+      endif
+
+      ## otherwise go on
+      newList = this.parseKList ([this.InspectedK k]);
+
+      ## check if the list has changed
+      if (length (newList) == length (this.InspectedK))
+        warning ("ClusterCriterion.addK: the list has not changed");
+      else
+        ## update ClusteringSolutions and CriterionValues
+        ClusteringSolutions_tmp = zeros (this.NumObservations, ...
+                                    length (newList));
+        CriterionValues_tmp = zeros (length (newList), 1);
+        for iter = 1 : length (this.InspectedK)
+          idx = find (newList == this.InspectedK(iter));
+
+          if (! isempty (idx))
+            ClusteringSolutions_tmp(:, idx) = this.ClusteringSolutions(:, iter);
+            CriterionValues_tmp(idx) = this.CriterionValues(iter);
+          endif
+        endfor
+        this.ClusteringSolutions = ClusteringSolutions_tmp;
+        this.CriterionValues = CriterionValues_tmp;
+
+        ## reset the old results
+        this.OptimalK = 0;
+        this.OptimalY = [];
+        this.OptimalIndex = 0;
+
+        ## update the list of cluster numbers to evaluate
+        this.InspectedK = newList;
+      endif
+    endfunction
+
+    ## -*- texinfo -*-
+    ## @deftypefn  {ClusterCriterion} {} plot (@var{obj})
+    ## @deftypefnx {ClusterCriterion} {@var{h} =} plot (@var{obj})
+    ##
+    ## Plot the evaluation results.
+    ##
+    ## Plot the CriterionValues against InspectedK from the ClusterCriterion,
+    ## @var{obj}, to the current plot. It can also return a handle to the
+    ## current plot.
+    ##
+    ## @end deftypefn
+    function h = plot (this)
+      yLabel = sprintf ("%s value", this.CriterionName);
+      h = gca ();
+      hold on;
+      plot (this.InspectedK, this.CriterionValues, "bo-");
+      plot (this.OptimalK, this.CriterionValues(this.OptimalIndex), "b*");
+      xlabel ("number of clusters");
+      ylabel (yLabel);
+      hold off;
+    endfunction
+
+    ## -*- texinfo -*-
+    ## @deftypefn {ClusterCriterion} {@var{obj} =} compact (@var{obj})
+    ##
+    ## Return a compact ClusterCriterion object (not implemented yet).
+    ##
+    ## @end deftypefn
+    function this = compact (this)
+      warning ("ClusterCriterion.compact: this method is not yet implemented.");
+    endfunction
+
+  endmethods
+
+  methods (Access = private)
+    ## check if a list of cluster sizes is correct
+    function retList = parseKList (this, KList)
+      if (isnumeric (KList) && isvector (KList) && all (find (KList > 0)) && ...
+          all (floor (KList) == KList))
+        retList = unique (KList);
+      else
+        error (["ClusterCriterion: the list of cluster sizes must be an " ...
+                "array of positive integer numbers"]);
+      endif
+    endfunction
+  endmethods
+endclassdef