html::collect_text: deal in String, not Vec<u8>.

Author: kivikakk

changelog.txt

@@ -1,3 +1,13 @@
+# [v0.42.0] - unreleased
+
+Changed APIs:
+
+* `html::collect_text` now returns a `String`.  `html::collect_text_append` is
+  added if you still want to start with your own (`String`) buffer.
+  * There was no particular reason for this populating a `Vec<u8>` instead of a
+    `String`; it was just old.
+
+
 # [v0.41.1] - 2025-09-14
 
 Bug fixes:

examples/headers.rs

@@ -1,9 +1,6 @@
 // Extract the document title by srching for a level-one header at the root level.
 
-use comrak::{
-    nodes::{AstNode, NodeCode, NodeValue},
-    parse_document, Arena, Options,
-};
+use comrak::{html::collect_text, nodes::NodeValue, parse_document, Arena, Options};
 
 fn main() {
     println!("{:?}", get_document_title("# Hello\n"));
@@ -25,27 +22,8 @@ fn get_document_title(document: &str) -> String {
             continue;
         }
 
-        let mut text = String::new();
-        collect_text(node, &mut text);
-
-        // The input was already known good UTF-8 (document: &str) so comrak
-        // guarantees the output will be too.
-        return text;
+        return collect_text(node);
     }
 
     "Untitled Document".to_string()
 }
-
-fn collect_text<'a>(node: &'a AstNode<'a>, output: &mut String) {
-    match node.data.borrow().value {
-        NodeValue::Text(ref literal) | NodeValue::Code(NodeCode { ref literal, .. }) => {
-            output.push_str(literal)
-        }
-        NodeValue::LineBreak | NodeValue::SoftBreak => output.push(' '),
-        _ => {
-            for n in node.children() {
-                collect_text(n, output);
-            }
-        }
-    }
-}

src/html.rs

@@ -632,11 +632,8 @@ fn render_heading<'a, T>(
                 context.write_all(b">")?;
 
                 if let Some(ref prefix) = context.options.extension.header_ids {
-                    let mut text_content = Vec::with_capacity(20);
-                    collect_text(node, &mut text_content);
-
-                    let mut id = String::from_utf8(text_content).unwrap();
-                    id = context.anchorizer.anchorize(id);
+                    let text_content = collect_text(node);
+                    let id = context.anchorizer.anchorize(text_content);
                     write!(
                         context,
                         "<a href=\"#{}\" aria-hidden=\"true\" class=\"anchor\" id=\"{}{}\"></a>",
@@ -648,12 +645,10 @@ fn render_heading<'a, T>(
             }
         }
         Some(adapter) => {
-            let mut text_content = Vec::with_capacity(20);
-            collect_text(node, &mut text_content);
-            let content = String::from_utf8(text_content).unwrap();
+            let text_content = collect_text(node);
             let heading = HeadingMeta {
                 level: nh.level,
-                content,
+                content: text_content,
             };
 
             if entering {
@@ -1587,22 +1582,30 @@ fn render_wiki_link<'a, T>(
 
 // Helpers
 
+/// Recurses through a node and all of its children in depth-first (document)
+/// order, returning the concatenated literal contents of text, code and math
+/// blocks. Line breaks and soft breaks are represented as a single whitespace
+/// character.
+pub fn collect_text<'a>(node: &'a AstNode<'a>) -> String {
+    let mut text = String::with_capacity(20);
+    collect_text_append(node, &mut text);
+    text
+}
+
 /// Recurses through a node and all of its children in depth-first (document)
 /// order, appending the literal contents of text, code and math blocks to
 /// an output buffer. Line breaks and soft breaks are represented as a single
 /// whitespace character.
-pub fn collect_text<'a>(node: &'a AstNode<'a>, output: &mut Vec<u8>) {
+pub fn collect_text_append<'a>(node: &'a AstNode<'a>, output: &mut String) {
     match node.data.borrow().value {
         NodeValue::Text(ref literal) | NodeValue::Code(NodeCode { ref literal, .. }) => {
-            output.extend_from_slice(literal.as_bytes())
-        }
-        NodeValue::LineBreak | NodeValue::SoftBreak => output.push(b' '),
-        NodeValue::Math(NodeMath { ref literal, .. }) => {
-            output.extend_from_slice(literal.as_bytes())
+            output.push_str(literal)
         }
+        NodeValue::LineBreak | NodeValue::SoftBreak => output.push(' '),
+        NodeValue::Math(NodeMath { ref literal, .. }) => output.push_str(literal),
         _ => {
             for n in node.children() {
-                collect_text(n, output);
+                collect_text_append(n, output);
             }
         }
     }