Account for left-hand whitespace in token positions

Fixes GH olivernn#417
hoelzro · Oct 5, 2019 · c00c2b9 · c00c2b9
1 parent 28d4f97
commit c00c2b9
Showing 1 changed file with 4 additions and 2 deletions.
diff --git a/lib/tokenizer.js b/lib/tokenizer.js
@@ -35,8 +35,10 @@ lunr.tokenizer = function (obj, metadata) {
     })
   }
 
-  var str = obj.toString().trim().toLowerCase(),
+  var origStr = obj.toString(),
+      str = origStr.trim().toLowerCase(),
       len = str.length,
+      leftPadLength = origStr.length - len,
       tokens = []
 
   for (var sliceEnd = 0, sliceStart = 0; sliceEnd <= len; sliceEnd++) {
@@ -47,7 +49,7 @@ lunr.tokenizer = function (obj, metadata) {
 
       if (sliceLength > 0) {
         var tokenMetadata = lunr.utils.clone(metadata) || {}
-        tokenMetadata["position"] = [sliceStart, sliceLength]
+        tokenMetadata["position"] = [sliceStart + leftPadLength, sliceLength]
         tokenMetadata["index"] = tokens.length
 
         tokens.push(