natural 6.3.0 → 6.5.0

package/lib/natural/ngrams/index.d.ts

@@ -22,7 +22,7 @@ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 THE SOFTWARE.
 */
 
-import Tokenizer from '../tokenizers/tokenizer'
+import { Tokenizer } from '../tokenizers'
 
 declare interface NGramsStats {
   ngrams: string[][]

package/lib/natural/tfidf/index.d.ts

@@ -23,7 +23,7 @@ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
 THE SOFTWARE.
 */
 
-import Tokenizer from '../tokenizers/tokenizer'
+import { Tokenizer } from '../tokenizers'
 
 declare type TfIdfCallback = (i: number, measure: number, key?: string) => void
 

package/lib/natural/tokenizers/sentence_tokenizer.js

@@ -34,7 +34,7 @@ util.inherits(SentenceTokenizer, Tokenizer)
 
 SentenceTokenizer.prototype.tokenize = function (text) {
   // break string up in to sentences based on punctation and quotation marks
-  let tokens = text.match(/(?<=\s+|^)["'‘“'"[({⟨]?(.*?[.?!])(\s[.?!])*["'’”'"\])}⟩]?(?=\s+|$)|(?<=\s+|^)\S(.*?[.?!])(\s[.?!])*(?=\s+|$)/g)
+  let tokens = text.match(/(?<=\s+|^)["'‘“'"[({⟨]?(.*?[.?!…]|[^.?!…]+)(\s[.?!…])*["'’”'"\])}⟩]?(?=\s+|$)/g)
 
   DEBUG && console.log('SentenceTokenizer.tokenize: ' + tokens)
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "natural",
   "description": "General natural language (tokenizing, stemming (English, Russian, Spanish), part-of-speech tagging, sentiment analysis, classification, inflection, phonetics, tfidf, WordNet, jaro-winkler, Levenshtein distance, Dice's Coefficient) facilities for node.",
-  "version": "6.3.0",
+  "version": "6.5.0",
   "homepage": "https://github.com/NaturalNode/natural",
   "repository": {
     "type": "git",