fix(cpp): require word boundary before numeric literals (#4442)

The floating-point literal pattern lacked a `\b` at the start of its
digit-leading branches, so digits embedded in an identifier were matched
as numbers. For example `a1.foo()` highlighted `1.f` as a floating-point
literal instead of leaving `a1` as an identifier.

Add `\b` to the decimal, exponent-only, and hexadecimal digit-leading
branches (mirroring the integer literal pattern). Dot-leading forms such
as `.5` are intentionally left untouched so they keep highlighting.

Closes #4231

Author: xianjianlf2

CHANGES.md

@@ -4,6 +4,7 @@ Core Grammars:
 
 - fix(lisp) preserve highlighting after quoted multiplication expressions [arturict][]
 - fix(rust) recognize `\\` and `\"` char-literal escapes so highlighting doesn't leak, issue #4351 [Sarath Francis][]
+- fix(cpp) require a word boundary before numeric literals so digits inside identifiers aren't highlighted as numbers, issue #4231 [Mark Xian][]
 
 Documentation:
 
@@ -16,6 +17,7 @@ CONTRIBUTORS
 [Dhruv Maniya]: https://github.com/iamdhrv
 [Elastic]: https://github.com/elastic
 [Sarath Francis]: https://github.com/sarathfrancis90
+[Mark Xian]: https://github.com/xianjianlf2
 
 
 ## Version 11.11.3

src/languages/cpp.js

@@ -57,12 +57,12 @@ export default function(hljs) {
         "[+-]?(?:" // Leading sign.
           // Decimal.
           + "(?:"
-            +"[0-9](?:'?[0-9])*\\.(?:[0-9](?:'?[0-9])*)?"
+            + "\\b[0-9](?:'?[0-9])*\\.(?:[0-9](?:'?[0-9])*)?"
             + "|\\.[0-9](?:'?[0-9])*"
           + ")(?:[Ee][+-]?[0-9](?:'?[0-9])*)?"
-          + "|[0-9](?:'?[0-9])*[Ee][+-]?[0-9](?:'?[0-9])*"
+          + "|\\b[0-9](?:'?[0-9])*[Ee][+-]?[0-9](?:'?[0-9])*"
           // Hexadecimal.
-          + "|0[Xx](?:"
+          + "|\\b0[Xx](?:"
             +"[0-9A-Fa-f](?:'?[0-9A-Fa-f])*(?:\\.(?:[0-9A-Fa-f](?:'?[0-9A-Fa-f])*)?)?"
             + "|\\.[0-9A-Fa-f](?:'?[0-9A-Fa-f])*"
           + ")[Pp][+-]?[0-9](?:'?[0-9])*"

test/markup/cpp/number-literals.expect.txt

@@ -53,3 +53,7 @@
 <span class="hljs-number">9&#x27;0UZ</span> <span class="hljs-number">9&#x27;0Uz</span> <span class="hljs-number">9&#x27;0uZ</span> <span class="hljs-number">9&#x27;0uz</span>
 
 <span class="hljs-type">char</span> word[] = { <span class="hljs-string">&#x27;3&#x27;</span>, <span class="hljs-string">&#x27;\0&#x27;</span> }; <span class="hljs-comment">// Make sure digit separators don&#x27;t mess up chars.</span>
+
+<span class="hljs-comment">/* Digits inside identifiers must not be highlighted as numbers. */</span>
+a1.<span class="hljs-built_in">foo</span>()
+<span class="hljs-type">int</span> a1 = b2.<span class="hljs-built_in">size</span>();

test/markup/cpp/number-literals.txt

@@ -53,3 +53,7 @@
 9'0UZ 9'0Uz 9'0uZ 9'0uz
 
 char word[] = { '3', '\0' }; // Make sure digit separators don't mess up chars.
+
+/* Digits inside identifiers must not be highlighted as numbers. */
+a1.foo()
+int a1 = b2.size();