enh(toml)(ini) Improve parsing of complex keys (#2595)

Fixes: #2594

Author: aduh95

CHANGES.md

@@ -48,6 +48,7 @@ Language Improvements:
 - enh(erlang) Add underscore separators to numeric literals (#2554) [Sergey Prokhorov][]
 - enh(handlebars) Support for sub-expressions, path-expressions, hashes, block-parameters and literals (#2344) [Nils Knappmeier][]
 - enh(protobuf) Support multiline comments (#2597) [Pavel Evstigneev][]
+- fix(toml) Improve key parsing (#2595) [Antoine du Hamel][]
 
 [Josh Goebel]: https://github.com/yyyc514
 [Peter Plantinga]: https://github.com/pplantinga
@@ -62,6 +63,7 @@ Language Improvements:
 [Jim Mason]: https://github.com/RocketMan
 [lioshi]: https://github.com/lioshi
 [Pavel Evstigneev]: https://github.com/Paxa
+[Antoine du Hamel]: https://github.com/aduh95
 
 
 ## Version 10.0.2

src/languages/ini.js

@@ -1,3 +1,5 @@
+import * as regex from '../lib/regex';
+
 /*
 Language: TOML, also INI
 Description: TOML aims to be a minimal configuration file format that's easy to read due to obvious semantics.
@@ -54,6 +56,17 @@ export default function(hljs) {
     relevance:0
   };
 
+  var BARE_KEY = /[A-Za-z0-9_-]+/;
+  var QUOTED_KEY_DOUBLE_QUOTE = /"(\\"|[^"])*"/;
+  var QUOTED_KEY_SINGLE_QUOTE = /'[^']*'/;
+  var ANY_KEY = regex.either(
+    BARE_KEY, QUOTED_KEY_DOUBLE_QUOTE, QUOTED_KEY_SINGLE_QUOTE
+  );
+  var DOTTED_KEY = regex.concat(
+    ANY_KEY, '(\\s*\\.\\s*', ANY_KEY, ')*',
+    regex.lookahead(/\s*=\s*[^#\s]/)
+  );
+
   return {
     name: 'TOML, also INI',
     aliases: ['toml'],
@@ -66,7 +79,7 @@ export default function(hljs) {
         begin: /\[+/, end: /\]+/
       },
       {
-        begin: /^[a-z0-9\[\]_\.-]+(?=\s*=\s*)/,
+        begin: DOTTED_KEY,
         className: 'attr',
         starts: {
           end: /$/,

test/markup/ini/keys.expect.txt

@@ -0,0 +1,32 @@
+<span class="hljs-comment"># Bare keys:</span>
+<span class="hljs-attr">key</span> = <span class="hljs-string">&quot;value&quot;</span>
+<span class="hljs-attr">bare_key</span> = <span class="hljs-string">&quot;value&quot;</span>
+<span class="hljs-attr">bare-key</span> = <span class="hljs-string">&quot;value&quot;</span>
+<span class="hljs-attr">1234</span> = <span class="hljs-string">&quot;value&quot;</span>
+
+<span class="hljs-comment"># Quoted keys:</span>
+<span class="hljs-attr">&quot;127.0.0.1&quot;</span> = <span class="hljs-string">&quot;value&quot;</span>
+<span class="hljs-attr">&quot;character encoding&quot;</span> = <span class="hljs-string">&quot;value&quot;</span>
+<span class="hljs-attr">&quot;ʎǝʞ&quot;</span> = <span class="hljs-string">&quot;value&quot;</span>
+<span class="hljs-attr">&#x27;key2&#x27;</span> = <span class="hljs-string">&quot;value&quot;</span>
+<span class="hljs-attr">&#x27;quoted &quot;value&quot;&#x27;</span> = <span class="hljs-string">&quot;value&quot;</span>
+
+<span class="hljs-attr">&quot;key \&quot;containing\&quot; backslash&quot;</span> = <span class="hljs-number">6</span>
+<span class="hljs-attr">&#x27;key \&quot;containing&quot; backslash\&#x27;</span> = <span class="hljs-number">6</span>
+
+<span class="hljs-comment"># empty quoted key is allowed</span>
+<span class="hljs-attr">&quot;&quot;</span> = <span class="hljs-string">&quot;blank&quot;</span>     <span class="hljs-comment"># VALID but discouraged</span>
+<span class="hljs-attr">&#x27;&#x27;</span> = <span class="hljs-string">&#x27;blank&#x27;</span>     <span class="hljs-comment"># VALID but discouraged</span>
+
+<span class="hljs-comment"># Dotted keys:</span>
+<span class="hljs-attr">name</span> = <span class="hljs-string">&quot;Orange&quot;</span>
+<span class="hljs-attr">physical.color</span> = <span class="hljs-string">&quot;orange&quot;</span>
+<span class="hljs-attr">physical.shape</span> = <span class="hljs-string">&quot;round&quot;</span>
+<span class="hljs-attr">site.&quot;google.com&quot;</span> = <span class="hljs-literal">true</span>
+<span class="hljs-attr">3.14159</span> = <span class="hljs-string">&quot;pi&quot;</span>
+
+<span class="hljs-comment"># Whitespace around dot-separated parts is ignored:</span>
+<span class="hljs-attr">hello  .  world</span>  =  <span class="hljs-string">&quot;!&quot;</span>
+
+<span class="hljs-comment"># Whitespace is ignored around key names and values</span>
+  <span class="hljs-attr">hello</span> = <span class="hljs-string">&quot;World!&quot;</span>

test/markup/ini/keys.txt

@@ -0,0 +1,32 @@
+# Bare keys:
+key = "value"
+bare_key = "value"
+bare-key = "value"
+1234 = "value"
+
+# Quoted keys:
+"127.0.0.1" = "value"
+"character encoding" = "value"
+"ʎǝʞ" = "value"
+'key2' = "value"
+'quoted "value"' = "value"
+
+"key \"containing\" backslash" = 6
+'key \"containing" backslash\' = 6
+
+# empty quoted key is allowed
+"" = "blank"     # VALID but discouraged
+'' = 'blank'     # VALID but discouraged
+
+# Dotted keys:
+name = "Orange"
+physical.color = "orange"
+physical.shape = "round"
+site."google.com" = true
+3.14159 = "pi"
+
+# Whitespace around dot-separated parts is ignored:
+hello  .  world  =  "!"
+
+# Whitespace is ignored around key names and values
+  hello = "World!"