fix: keep a backslash in a code span from hiding emphasis

src/Lexer.ts

@@ -397,7 +397,17 @@ export class _Lexer<ParserOutput = string, RendererOutput = string> {
     // Every mask must keep the length it replaces: emStrong and del line
     // maskedSrc up with src by slicing from the end. `anyPunctuation` matches
     // unicode punctuation, so an escaped astral character is 3 code units.
-    maskedSrc = maskedSrc.replace(this.tokenizer.rules.inline.anyPunctuation, match0 => '+'.repeat(match0.length));
+    // A backslash inside a code span is literal, not an escape. Mask the
+    // backticks inside the span instead, so blockSkip still skips all of it.
+    maskedSrc = maskedSrc.replace(this.tokenizer.rules.inline.anyPunctuationOutsideCode, (match0, code, punct) => {
+      if (punct) {
+        return '+'.repeat(match0.length);
+      }
+      if (code) {
+        return code + match0.slice(code.length, -code.length).replace(/`/g, '+') + code;
+      }
+      return match0;
+    });
 
     // Mask out other blocks
     maskedSrc = maskedSrc.replace(this.tokenizer.rules.inline.blockSkip, (match0, _link, context) => {

src/rules.ts

@@ -431,6 +431,14 @@ const anyPunctuation = edit(/\\(punct)/, 'gu')
   .replace(/punct/g, _punctuation)
   .getRegex();
 
+// `anyPunctuation` for the em/strong mask. A backslash is literal inside a code
+// span, so a code span (matched like `inline.code`), or a backtick run that
+// opens none, is matched whole and kept; only the escapes outside code spans
+// are masked.
+const anyPunctuationOutsideCode = edit(/(`+)(?:[^`]|[^`][\s\S]*?[^`])\1(?!`)|`+|\\(punct)/, 'gu')
+  .replace(/punct/g, _punctuation)
+  .getRegex();
+
 const autolink = edit(/^<(scheme:[^\s\x00-\x1f<>]*|email)>/)
   .replace('scheme', /[a-zA-Z][a-zA-Z0-9+.-]{1,31}/)
   .replace('email', /[a-zA-Z0-9.!#$%&'*+/=?^_`{|}~-]+(@)[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)+(?![-_])/)
@@ -513,6 +521,7 @@ const _extendedEmailProtocol = edit(/(?:mailto:email|xmpp:email(?:\/[A-Za-z0-9@.
 const inlineNormal = {
   _backpedal: noopTest, // only used for GFM url
   anyPunctuation,
+  anyPunctuationOutsideCode,
   autolink,
   blockSkip,
   br,

test/specs/new/codespan_backslash_emphasis.html

@@ -0,0 +1,13 @@
+<p>Note that <code>\</code> or <code>/</code> <em>will</em> be interpreted as path separators on Windows, and will match against <code>/</code> in glob expressions.</p>
+
+<p>Use <code>C:\Users\</code> for <strong>home</strong> and <code>D:\</code> for data.</p>
+
+<p><code>a\</code> <em>b</em> <code>c</code></p>
+
+<p>`<em>escaped tick</em>` <code>code</code></p>
+
+<p><code>\</code> <code>\</code> <strong>after double backticks</strong></p>
+
+<p>_<code>\`_</code></p>
+
+<p>_<code>a`_</code></p>

test/specs/new/codespan_backslash_emphasis.md

@@ -0,0 +1,13 @@
+Note that `\` or `/` _will_ be interpreted as path separators on Windows, and will match against `/` in glob expressions.
+
+Use `C:\Users\` for **home** and `D:\` for data.
+
+`a\` *b* `c`
+
+\`*escaped tick*\` `code`
+
+`\` ``\`` **after double backticks**
+
+_``\`_``
+
+_`` a`_ ``