utf8: Fix encoding of U+FFFE and U+FFFF

Fixes commonmark#548.
nwellnhof · Apr 23, 2024 · 12f5205 · 12f5205
1 parent 2632fdc
commit 12f5205
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 6 deletions.
diff --git a/src/utf8.c b/src/utf8.c
@@ -199,12 +199,6 @@ void cmark_utf8proc_encode_char(int32_t uc, cmark_strbuf *buf) {
     dst[0] = (uint8_t)(0xC0 + (uc >> 6));
     dst[1] = 0x80 + (uc & 0x3F);
     len = 2;
-  } else if (uc == 0xFFFF) {
-    dst[0] = 0xFF;
-    len = 1;
-  } else if (uc == 0xFFFE) {
-    dst[0] = 0xFE;
-    len = 1;
   } else if (uc < 0x10000) {
     dst[0] = (uint8_t)(0xE0 + (uc >> 12));
     dst[1] = 0x80 + ((uc >> 6) & 0x3F);

diff --git a/test/regression.txt b/test/regression.txt
@@ -275,3 +275,10 @@ Case fold test
 .
 <p><a href="/url">link</a></p>
 ````````````````````````````````
+
+https://github.com/commonmark/cmark/issues/548
+```````````````````````````````` example
+(&#xFFFE;&#xFFFF;)
+.
+<p>()</p>
+````````````````````````````````