(enh) lots of small Ruby improvements (#3491)

* (enh) lots of small Ruby improvements
* do not scope () as params
* better camel case regex

CHANGES.md

@@ -11,6 +11,7 @@ New Grammars:
 
 Grammars:
 
+- enh(ruby) lots of small Ruby cleanups/improvements [Josh Goebel][]
 - enh(objectivec) add `type` and `variable.language` scopes [Josh Goebel][]
 - enh(xml) support processing instructions (#3492) [Josh Goebel][]
 - enh(ruby ) better support multi-line IRB prompts

src/highlight.js

@@ -277,8 +277,8 @@ const HLJS = function(hljs) {
      */
     function emitMultiClass(scope, match) {
       let i = 1;
-      // eslint-disable-next-line no-undefined
-      while (match[i] !== undefined) {
+      const max = match.length - 1;
+      while (i <= max) {
         if (!scope._emit[i]) { i++; continue; }
         const klass = language.classNameAliases[scope[i]] || scope[i];
         const text = match[i];

src/languages/ruby.js

@@ -10,15 +10,71 @@ Category: common
 export default function(hljs) {
   const regex = hljs.regex;
   const RUBY_METHOD_RE = '([a-zA-Z_]\\w*[!?=]?|[-+~]@|<<|>>|=~|===?|<=>|[<>]=?|\\*\\*|[-/+%^&*~`|]|\\[\\]=?)';
+  // TODO: move concepts like CAMEL_CASE into `modes.js`
+  const CLASS_NAME_RE = regex.either(
+    /\b([A-Z]+[a-z0-9]+)+/,
+    // ends in caps
+    /\b([A-Z]+[a-z0-9]+)+[A-Z]+/,
+  )
+  ;
+  const CLASS_NAME_WITH_NAMESPACE_RE = regex.concat(CLASS_NAME_RE, /(::\w+)*/)
   const RUBY_KEYWORDS = {
-    keyword:
-      'and then defined module in return redo if BEGIN retry end for self when '
-      + 'next until do begin unless END rescue else break undef not super class case '
-      + 'require yield alias while ensure elsif or include attr_reader attr_writer attr_accessor '
-      + '__FILE__',
-    built_in: 'proc lambda',
-    literal:
-      'true false nil'
+    "variable.constant": [
+      "__FILE__",
+      "__LINE__"
+    ],
+    "variable.language": [
+      "self",
+      "super",
+    ],
+    keyword: [
+      "alias",
+      "and",
+      "attr_accessor",
+      "attr_reader",
+      "attr_writer",
+      "begin",
+      "BEGIN",
+      "break",
+      "case",
+      "class",
+      "defined",
+      "do",
+      "else",
+      "elsif",
+      "end",
+      "END",
+      "ensure",
+      "for",
+      "if",
+      "in",
+      "include",
+      "module",
+      "next",
+      "not",
+      "or",
+      "redo",
+      "require",
+      "rescue",
+      "retry",
+      "return",
+      "then",
+      "undef",
+      "unless",
+      "until",
+      "when",
+      "while",
+      "yield",
+    ],
+    built_in: [
+      "proc",
+      "lambda"
+    ],
+    literal: [
+      "true",
+      "false",
+      "nil"
+    ]
   };
   const YARDOCTAG = {
     className: 'doctag',
@@ -42,7 +98,7 @@ export default function(hljs) {
         relevance: 10
       }
     ),
-    hljs.COMMENT('^__END__', '\\n$')
+    hljs.COMMENT('^__END__', hljs.MATCH_NOTHING_RE)
   ];
   const SUBST = {
     className: 'subst',
@@ -156,49 +212,82 @@ export default function(hljs) {
   };
 
   const PARAMS = {
-    className: 'params',
-    begin: '\\(',
-    end: '\\)',
-    endsParent: true,
+    variants: [
+      {
+        match: /\(\)/,
+      },
+      {
+        className: 'params',
+        begin: /\(/,
+        end: /(?=\))/,
+        excludeBegin: true,
+        endsParent: true,
+        keywords: RUBY_KEYWORDS,
+      }
+    ]
+  };
+
+  const CLASS_DEFINITION = {
+    variants: [
+      {
+        match: [
+          /class\s+/,
+          CLASS_NAME_WITH_NAMESPACE_RE,
+          /\s+<\s+/,
+          CLASS_NAME_WITH_NAMESPACE_RE
+        ]
+      },
+      {
+        match: [
+          /class\s+/,
+          CLASS_NAME_WITH_NAMESPACE_RE
+        ]
+      }
+    ],
+    scope: {
+      2: "title.class",
+      4: "title.class.inherited"
+    },
     keywords: RUBY_KEYWORDS
   };
 
+  const UPPER_CASE_CONSTANT = {
+    relevance: 0,
+    match: /\b[A-Z][A-Z_0-9]+\b/,
+    className: "variable.constant"
+  };
+
+  const METHOD_DEFINITION = {
+    match: [
+      /def/, /\s+/,
+      RUBY_METHOD_RE
+    ],
+    scope: {
+      1: "keyword",
+      3: "title.function"
+    },
+    contains: [
+      PARAMS
+    ]
+  };
+
+  const OBJECT_CREATION = {
+    relevance: 0,
+    match: [
+      CLASS_NAME_WITH_NAMESPACE_RE,
+      /\.new[ (]/
+    ],
+    scope: {
+      1: "title.class"
+    }
+  };
+
   const RUBY_DEFAULT_CONTAINS = [
     STRING,
-    {
-      className: 'class',
-      beginKeywords: 'class module',
-      end: '$|;',
-      illegal: /=/,
-      contains: [
-        hljs.inherit(hljs.TITLE_MODE, { begin: '[A-Za-z_]\\w*(::\\w+)*(\\?|!)?' }),
-        {
-          begin: '<\\s*',
-          contains: [
-            {
-              begin: '(' + hljs.IDENT_RE + '::)?' + hljs.IDENT_RE,
-              // we already get points for <, we don't need poitns
-              // for the name also
-              relevance: 0
-            }
-          ]
-        }
-      ].concat(COMMENT_MODES)
-    },
-    {
-      className: 'function',
-      // def method_name(
-      // def method_name;
-      // def method_name (end of line)
-      begin: regex.concat(/def\s+/, regex.lookahead(RUBY_METHOD_RE + "\\s*(\\(|;|$)")),
-      relevance: 0, // relevance comes from kewords
-      keywords: "def",
-      end: '$|;',
-      contains: [
-        hljs.inherit(hljs.TITLE_MODE, { begin: RUBY_METHOD_RE }),
-        PARAMS
-      ].concat(COMMENT_MODES)
-    },
+    CLASS_DEFINITION,
+    OBJECT_CREATION,
+    UPPER_CASE_CONSTANT,
+    METHOD_DEFINITION,
     {
       // swallow namespace qualifiers before symbols
       begin: hljs.IDENT_RE + '::' },
@@ -227,6 +316,8 @@ export default function(hljs) {
       className: 'params',
       begin: /\|/,
       end: /\|/,
+      excludeBegin: true,
+      excludeEnd: true,
       relevance: 0, // this could be a lot of things (in other languages) other than params
       keywords: RUBY_KEYWORDS
     },

test/markup/erb/default.expect.txt

@@ -1,6 +1,6 @@
 <span class="hljs-comment">&lt;%# this is a comment %&gt;</span><span class="language-xml">
 
-&lt;%</span><span class="language-ruby"> <span class="hljs-variable">@posts</span>.each <span class="hljs-keyword">do</span> <span class="hljs-params">|post|</span> </span><span class="language-xml">%&gt;
+&lt;%</span><span class="language-ruby"> <span class="hljs-variable">@posts</span>.each <span class="hljs-keyword">do</span> |<span class="hljs-params">post</span>| </span><span class="language-xml">%&gt;
   <span class="hljs-tag">&lt;<span class="hljs-name">p</span>&gt;</span>&lt;%=</span><span class="language-ruby"> link_to post.title, post </span><span class="language-xml">%&gt;<span class="hljs-tag">&lt;/<span class="hljs-name">p</span>&gt;</span>
 &lt;%</span><span class="language-ruby"> <span class="hljs-keyword">end</span> </span><span class="language-xml">%&gt;
 

test/markup/haml/default.expect.txt

@@ -5,7 +5,7 @@
 <span class="hljs-comment">    /html comment</span>
 <span class="hljs-comment">    -# ignore this line</span>
 <span class="hljs-tag">    %<span class="hljs-selector-tag">ul</span>(<span class="hljs-attr">style</span>=<span class="hljs-string">&#x27;margin: 0&#x27;</span>)</span>
-    -<span class="language-ruby">items.each <span class="hljs-keyword">do</span> <span class="hljs-params">|i|</span></span>
+    -<span class="language-ruby">items.each <span class="hljs-keyword">do</span> |<span class="hljs-params">i</span>|</span>
 <span class="hljs-tag">      %<span class="hljs-selector-tag">i</span></span>= i
     =<span class="language-ruby"> variable</span>
     =<span class="language-ruby">variable2</span>

test/markup/ruby/heredoc.expect.txt

@@ -13,15 +13,15 @@ message = <span class="hljs-string">&lt;&lt;-MESSAGE.chomp
   This looks good
 MESSAGE</span>
 
-<span class="hljs-function"><span class="hljs-keyword">def</span> <span class="hljs-title">foo</span><span class="hljs-params">()</span></span>
+<span class="hljs-keyword">def</span> <span class="hljs-title function_">foo</span>()
   msg = <span class="hljs-string">&lt;&lt;-HTML
   &lt;div&gt;
     &lt;h4&gt;<span class="hljs-subst">#{bar}</span>&lt;/h4&gt;
   &lt;/div&gt;
   HTML</span>
 <span class="hljs-keyword">end</span>
 
-<span class="hljs-function"><span class="hljs-keyword">def</span> <span class="hljs-title">baz</span><span class="hljs-params">()</span></span>
+<span class="hljs-keyword">def</span> <span class="hljs-title function_">baz</span>()
   msg = <span class="hljs-string">&lt;&lt;~FOO
   &lt;div&gt;
     &lt;h4&gt;<span class="hljs-subst">#{bar}</span>&lt;/h4&gt;

test/markup/ruby/prompt.expect.txt

@@ -7,7 +7,7 @@
 
 <span class="hljs-meta prompt_">jruby-1.7.16 :001 &gt;</span> <span class="hljs-string">&quot;RVM-Format&quot;</span>
 
-<span class="hljs-meta prompt_">&gt;&gt;</span> obj = OpenStruct.new <span class="hljs-symbol">:integer</span> =&gt; <span class="hljs-number">987</span>, <span class="hljs-symbol">:symbol</span> =&gt; <span class="hljs-symbol">:so_great</span>
+<span class="hljs-meta prompt_">&gt;&gt;</span> obj = <span class="hljs-title class_">OpenStruct</span>.new <span class="hljs-symbol">:integer</span> =&gt; <span class="hljs-number">987</span>, <span class="hljs-symbol">:symbol</span> =&gt; <span class="hljs-symbol">:so_great</span>
 =&gt; #&lt;OpenStruct integer=987, symbol=:so_great&gt;
 <span class="hljs-meta prompt_">&gt;&gt;</span> [obj,obj,obj]
 =&gt; [#&lt;OpenStruct integer=987, symbol=:so_great&gt;, #&lt;OpenStruct integer=987, symbol=:so_great&gt;, #&lt;OpenStruct integer=987, symbol=:so_great&gt;]
@@ -22,8 +22,8 @@
 
 <span class="hljs-meta prompt_">irb(main):002:0&gt;</span> test = <span class="hljs-number">1</span>
 
-<span class="hljs-meta prompt_">irb(main):001:1*</span> <span class="hljs-class"><span class="hljs-keyword">class</span> <span class="hljs-title">Secret</span></span>
-<span class="hljs-meta prompt_">irb(main):002:2*</span>   <span class="hljs-function"><span class="hljs-keyword">def</span> <span class="hljs-title">[]</span><span class="hljs-params">(x)</span></span>
+<span class="hljs-meta prompt_">irb(main):001:1*</span> <span class="hljs-keyword">class</span> <span class="hljs-title class_">Secret</span>
+<span class="hljs-meta prompt_">irb(main):002:2*</span>   <span class="hljs-keyword">def</span> <span class="hljs-title function_">[]</span>(<span class="hljs-params">x</span>)
 <span class="hljs-meta prompt_">irb(main):003:2*</span>     <span class="hljs-string">&quot;TREASURE&quot;</span> <span class="hljs-keyword">if</span> x==<span class="hljs-number">42</span>
 <span class="hljs-meta prompt_">irb(main):004:1*</span>   <span class="hljs-keyword">end</span>
 <span class="hljs-meta prompt_">irb(main):005:0&gt;</span> <span class="hljs-keyword">end</span>