Use dictionaries (objects) to count unique lines and poses. Greatly speeds up...

Use dictionaries (objects) to count unique lines and poses. Greatly speeds up computing especially Pit's and Meia's unique line counts. The old code used linear searches which scaled poorly.

js/spniSelect.js

@@ -1223,27 +1223,25 @@ function fetchBehaviour(path) {
  */
 function countLinesImages(xml) {
     // parse all lines of dialogue and all images
-    var lines = [];
-    var poses = [];
+	var numTotalLines = 0;
+	var numUniqueDialogueLines = 0;
+	var numUniqueUsedPoses = 0;
+    var lines = {};
+    var poses = {};
     $(xml).find('state').each(function(idx, data) {
-        lines.push(data.textContent.trim());
-        poses.push(data.getAttribute("img"));
+		numTotalLines++;
+		// count only unique lines of dialogue
+		if (lines[data.textContent.trim()] === undefined) numUniqueDialogueLines++;
+        lines[data.textContent.trim()] = 1;
+		// count unique number of poses used in dialogue
+		// note that this number may differ from actual image count if some images
+		// are never used, or if images that don't exist are used in the dialogue
+		if (poses[data.getAttribute("img")] === undefined) numUniqueUsedPoses++;
+        poses[data.getAttribute("img")] = 1;
     });
 
-    // count only unique lines of dialogue
-    var numUniqueDialogueLines = lines.filter(function(data, idx) {
-        return idx == lines.lastIndexOf(data);
-    }).length;
-
-    // count unique number of poses used in dialogue
-    // note that this number may differ from actual image count if some images
-    // are never used, or if images that don't exist are used in the dialogue
-    var numUniqueUsedPoses = poses.filter(function(data, idx) {
-        return idx == poses.lastIndexOf(data);
-    }).length;
-
     return {
-        numTotalLines : lines.length,
+        numTotalLines : numTotalLines,
         numUniqueLines : numUniqueDialogueLines,
         numPoses : numUniqueUsedPoses
     };