Merge pull request #158 from esdoppio/fix-multiline-regexp

Recognize multi-line regular expressions

Author: Wliu

grammars/ruby.cson

@@ -486,16 +486,16 @@
   }
   {
     'begin': '''(?x)
-      (?<![\\w)])((/))(?![*+?])
+      (?<![\\w)])((/))(?![?*+])
       (?=
-        (?:\\\\/|[^/])*/[eimnosux]*\\s*
-        (
-          [\\]#).,?:}]
-          |
-          $|\\|\\||&&|<=>|=>|==|=~|!~|!=|;
-          |
+        (?:\\\\/|[^/])*+          # Do NOT change the order
+        /[eimnosux]*\\s*
+        (?:
+          [)\\]}#.,?:]|\\|\\||&&|<=>|=>|==|=~|!~|!=|;|$|
           if|else|elsif|then|do|end|unless|while|until|or|and
         )
+        |
+        $
       )
     '''
     'captures':
@@ -505,7 +505,7 @@
         'name': 'punctuation.section.regexp.ruby'
     'comment': 'regular expression literal with interpolation'
     'contentName': 'string.regexp.interpolated.ruby'
-    'end': '((/[eimnosux]*))(?=[^eimnosux])'
+    'end': '((/[eimnosux]*))'
     'patterns': [
       {
         'include': '#regex_sub'

spec/ruby-spec.coffee

@@ -480,6 +480,22 @@ describe "Ruby grammar", ->
     expect(tokens[6]).toEqual value: ' ', scopes: ['source.ruby']
     expect(tokens[7]).toEqual value: '1', scopes: ['source.ruby', 'constant.numeric.ruby']
 
+  it "tokenizes multiline regular expressions", ->
+    tokens = grammar.tokenizeLines '''
+      regexp = /
+        foo|
+        bar
+      /ix
+    '''
+
+    expect(tokens[0][0]).toEqual value: 'regexp ', scopes: ['source.ruby']
+    expect(tokens[0][1]).toEqual value: '=', scopes: ['source.ruby', 'keyword.operator.assignment.ruby']
+    expect(tokens[0][2]).toEqual value: ' ', scopes: ['source.ruby']
+    expect(tokens[0][3]).toEqual value: '/', scopes: ['source.ruby', 'string.regexp.interpolated.ruby', 'punctuation.section.regexp.ruby']
+    expect(tokens[1][0]).toEqual value: '  foo|', scopes: ['source.ruby', 'string.regexp.interpolated.ruby']
+    expect(tokens[2][0]).toEqual value: '  bar', scopes: ['source.ruby', 'string.regexp.interpolated.ruby']
+    expect(tokens[3][0]).toEqual value: '/ix', scopes: ['source.ruby', 'string.regexp.interpolated.ruby', 'punctuation.section.regexp.ruby']
+
   it "tokenizes the / arithmetic operator", ->
     {tokens} = grammar.tokenizeLine('call/me/maybe')
     expect(tokens[0]).toEqual value: 'call', scopes: ['source.ruby']