feat(ui): support Pandoc style code blocks (#12099)

This resolves https://codeberg.org/forgejo/forgejo/issues/11107.

Codeberg doesn't support [Pandoc style code blocks](https://pandoc.org/MANUAL.html#extension-fenced_code_attributes), so only the two of these 3 will have syntax highlighting.

\`\`\`haskell
qsort []     = []
qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++
               qsort (filter (>= x) xs)
\`\`\`

\`\`\`haskell {.numberLines}
qsort []     = []
qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++
               qsort (filter (>= x) xs)
\`\`\`

\`\`\`{.numberLines .haskell}
qsort []     = []
qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++
               qsort (filter (>= x) xs)
\`\`\`

```haskell
qsort []     = []
qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++
               qsort (filter (>= x) xs)
```

```haskell {.numberLines}
qsort []     = []
qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++
               qsort (filter (>= x) xs)
```

```{.numberLines .haskell}
qsort []     = []
qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++
               qsort (filter (>= x) xs)
```

This PR adds syntax highlighting to the examples with Pandoc style code blocks. It also adds redundant code to explicitly handle the second case with the trailing attribute syntax, which might be unnecessary since it already works, but I think should be fine to leave in.

Reviewed-on: https://codeberg.org/forgejo/forgejo/pulls/12099
Reviewed-by: Ellen Εμίλια Άννα Zscheile <fogti@noreply.codeberg.org>
Reviewed-by: Gusted <gusted@noreply.codeberg.org>

modules/markup/markdown/markdown_test.go

@@ -1489,14 +1489,14 @@ func TestCallout(t *testing.T) {
 </blockquote>`)
 }
 
-func TestCodeblockLanguageStripping(t *testing.T) {
+func TestCodeblockLanguageTransformation(t *testing.T) {
 	test := func(input, expected string) {
 		buffer, err := markdown.RenderString(&markup.RenderContext{Ctx: git.DefaultContext}, input)
 		require.NoError(t, err)
 		assert.Equal(t, strings.TrimSpace(expected), strings.TrimSpace(string(buffer)))
 	}
 
-	// Unstripped
+	// No transformation
 	test(
 		"```rust\n"+
 			"fn main() {}\n"+
@@ -1504,7 +1504,7 @@ func TestCodeblockLanguageStripping(t *testing.T) {
 		`<pre class="code-block"><code class="chroma language-rust display"><span class="k">fn</span> <span class="nf">main</span><span class="p">()</span><span class="w"> </span><span class="p">{}</span><span class="w">
 </span></code></pre>`)
 
-	// Stripped
+	// Comma stripped
 	test(
 		"```rust,ignore\n"+
 			"fn main() {}\n"+
@@ -1512,6 +1512,32 @@ func TestCodeblockLanguageStripping(t *testing.T) {
 		`<pre class="code-block"><code class="chroma language-rust display"><span class="k">fn</span> <span class="nf">main</span><span class="p">()</span><span class="w"> </span><span class="p">{}</span><span class="w">
 </span></code></pre>`)
 
+	// Pandoc stripping
+	// https://pandoc.org/MANUAL.html#extension-fenced_code_attributes
+	test(
+		"```haskell {.numberLines}\n"+
+			"qsort []     = []\n"+
+			"qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++\n"+
+			"               qsort (filter (>= x) xs)\n"+
+			"```",
+		`<pre class="code-block"><code class="chroma language-haskell display"><span class="nf">qsort</span> <span class="kt">[]</span>     <span class="ow">=</span> <span class="kt">[]</span>
+<span class="nf">qsort</span> <span class="p">(</span><span class="n">x</span><span class="kt">:</span><span class="n">xs</span><span class="p">)</span> <span class="ow">=</span> <span class="n">qsort</span> <span class="p">(</span><span class="n">filter</span> <span class="p">(</span><span class="o">&lt;</span> <span class="n">x</span><span class="p">)</span> <span class="n">xs</span><span class="p">)</span> <span class="o">++</span> <span class="p">[</span><span class="n">x</span><span class="p">]</span> <span class="o">++</span>
+               <span class="n">qsort</span> <span class="p">(</span><span class="n">filter</span> <span class="p">(</span><span class="o">&gt;=</span> <span class="n">x</span><span class="p">)</span> <span class="n">xs</span><span class="p">)</span>
+</code></pre>`)
+
+	// Pandoc language extracting
+	// https://pandoc.org/MANUAL.html#extension-fenced_code_attributes
+	test(
+		"```   { #mycode .numberLines .haskell startFrom=\"100\" }   \n"+
+			"qsort []     = []\n"+
+			"qsort (x:xs) = qsort (filter (< x) xs) ++ [x] ++\n"+
+			"               qsort (filter (>= x) xs)\n"+
+			"```",
+		`<pre class="code-block"><code class="chroma language-haskell display"><span class="nf">qsort</span> <span class="kt">[]</span>     <span class="ow">=</span> <span class="kt">[]</span>
+<span class="nf">qsort</span> <span class="p">(</span><span class="n">x</span><span class="kt">:</span><span class="n">xs</span><span class="p">)</span> <span class="ow">=</span> <span class="n">qsort</span> <span class="p">(</span><span class="n">filter</span> <span class="p">(</span><span class="o">&lt;</span> <span class="n">x</span><span class="p">)</span> <span class="n">xs</span><span class="p">)</span> <span class="o">++</span> <span class="p">[</span><span class="n">x</span><span class="p">]</span> <span class="o">++</span>
+               <span class="n">qsort</span> <span class="p">(</span><span class="n">filter</span> <span class="p">(</span><span class="o">&gt;=</span> <span class="n">x</span><span class="p">)</span> <span class="n">xs</span><span class="p">)</span>
+</code></pre>`)
+
 	// No language identifier
 	test(
 		"```\n"+

modules/markup/markdown/transform_codeblock_lang.go

@@ -6,6 +6,7 @@ package markdown
 import (
 	"bytes"
 
+	"github.com/alecthomas/chroma/v2/lexers"
 	"github.com/yuin/goldmark/ast"
 	"github.com/yuin/goldmark/text"
 )
@@ -16,6 +17,32 @@ func (g *ASTTransformer) transformCodeblockLanguage(v *ast.FencedCodeBlock, read
 	}
 	src := reader.Source()
 	info := v.Info.Segment.Value(src)
+
+	// Parse Pandoc style attributes
+	// https://pandoc.org/MANUAL.html#extension-fenced_code_attributes
+	//
+	// For example,
+	// ```{.haskell .numberLines}
+	// ...
+	// ```
+	// Should have a language of "haskell", not "{.haskell .numberLines}"
+	if trimmed := bytes.TrimSpace(info); bytes.HasPrefix(trimmed, []byte{'{'}) && bytes.HasSuffix(trimmed, []byte{'}'}) {
+		attributes := trimmed[1 : len(trimmed)-1]
+		for attribute := range bytes.SplitSeq(attributes, []byte{' '}) {
+			if class, found := bytes.CutPrefix(attribute, []byte{'.'}); found {
+				if lexer := lexers.Get(string(class)); lexer != nil {
+					lang := class
+					langInx := bytes.Index(info, lang)
+					start := v.Info.Segment.Start + langInx
+					end := start + len(lang)
+					v.Info = ast.NewTextSegment(text.NewSegment(start, end))
+					return
+				}
+			}
+		}
+		return
+	}
+
 	// Strip language after commas
 	//
 	// For example,