Preserve non-ASCII whitespace in test suite

src/firstpass.rs

@@ -1237,12 +1237,12 @@ fn scan_paragraph_interrupt(bytes: &[u8]) -> bool {
 
 /// Assumes `text_bytes` is preceded by `<`.
 fn get_html_end_tag(text_bytes: &[u8]) -> Option<&'static str> {
-    static BEGIN_TAGS: &[&[u8]; 3] = &[b"pre", b"style", b"script"];
+    static BEGIN_TAGS: &[&[u8]; 4] = &[b"pre", b"style", b"script", b"textarea"];
     static ST_BEGIN_TAGS: &[&[u8]; 3] = &[b"!--", b"?", b"![CDATA["];
 
     for (beg_tag, end_tag) in BEGIN_TAGS
         .iter()
-        .zip(["</pre>", "</style>", "</script>"].iter())
+        .zip(["</pre>", "</style>", "</script>", "</textarea>"].iter())
     {
         let tag_len = beg_tag.len();
 

tests/suite/spec.rs

@@ -319,7 +319,7 @@ fn spec_test_25() {
 ¾ ℋ ⅆ
 ∲ ≧̸
 "##;
-    let expected = r##"<p>  &amp; © Æ Ď
+    let expected = r##"<p>  &amp; © Æ Ď
 ¾ ℋ ⅆ
 ∲ ≧̸</p>
 "##;
@@ -5070,20 +5070,20 @@ fn spec_test_332() {
 
 #[test]
 fn spec_test_333() {
-    let original = r##"` b `
+    let original = r##"` b `
 "##;
-    let expected = r##"<p><code> b </code></p>
+    let expected = r##"<p><code> b </code></p>
 "##;
 
     test_markdown_html(original, expected, false);
 }
 
 #[test]
 fn spec_test_334() {
-    let original = r##"` `
+    let original = r##"` `
 `  `
 "##;
-    let expected = r##"<p><code> </code>
+    let expected = r##"<p><code> </code>
 <code>  </code></p>
 "##;
 
@@ -5279,9 +5279,9 @@ fn spec_test_352() {
 
 #[test]
 fn spec_test_353() {
-    let original = r##"* a *
+    let original = r##"* a *
 "##;
-    let expected = r##"<p>* a *</p>
+    let expected = r##"<p>* a *</p>
 "##;
 
     test_markdown_html(original, expected, false);
@@ -6839,7 +6839,7 @@ fn spec_test_505() {
 
 #[test]
 fn spec_test_506() {
-    let original = r##"[link](/url "title")
+    let original = r##"[link](/url "title")
 "##;
     let expected = r##"<p><a href="/url%C2%A0%22title%22">link</a></p>
 "##;

third_party/CommonMark/spec.txt

@@ -644,7 +644,7 @@ references and their corresponding code points.
 ¾ ℋ ⅆ
 ∲ ≧̸
 .
-<p>  &amp; © Æ Ď
+<p>  &amp; © Æ Ď
 ¾ ℋ ⅆ
 ∲ ≧̸</p>
 ````````````````````````````````
@@ -5927,18 +5927,18 @@ Only [spaces], and not [unicode whitespace] in general, are
 stripped in this way:
 
 ```````````````````````````````` example
-` b `
+` b `
 .
-<p><code> b </code></p>
+<p><code> b </code></p>
 ````````````````````````````````
 
 No stripping occurs if the code span contains only spaces:
 
 ```````````````````````````````` example
-` `
+` `
 `  `
 .
-<p><code> </code>
+<p><code> </code>
 <code>  </code></p>
 ````````````````````````````````
 
@@ -6332,9 +6332,9 @@ a*"foo"*
 Unicode nonbreaking spaces count as whitespace, too:
 
 ```````````````````````````````` example
-* a *
+* a *
 .
-<p>* a *</p>
+<p>* a *</p>
 ````````````````````````````````
 
 
@@ -7755,7 +7755,7 @@ ending.
 Other [Unicode whitespace] like non-breaking space doesn't work.
 
 ```````````````````````````````` example
-[link](/url "title")
+[link](/url "title")
 .
 <p><a href="/url%C2%A0%22title%22">link</a></p>
 ````````````````````````````````