Skip to content

✨ feat(mq-markdown): convert HTML form elements to Markdown #2201

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
harehare merged 1 commit into from
Aug 14, 2026
Merged

✨ feat(mq-markdown): convert HTML form elements to Markdown #2201

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
111 changes: 101 additions & 10 deletions crates/mq-markdown/src/html_to_markdown/converter.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -605,6 +605,66 @@ fn handle_dl_element(element: &HtmlElement, options: &ConversionOptions) -> miet
}
}

/// Converts a `<select>`'s `<option>`/`<optgroup>` children into a task-list, reusing the
/// same `[x]`/`[ ]` marker convention as checkbox/radio `<input>`s.
fn handle_select_element(element: &HtmlElement) -> miette::Result<String> {
let mut items = Vec::new();
collect_select_options(&element.children, &mut items, 0)?;
Ok(items.join("\n"))
}

fn collect_select_options(nodes: &[HtmlNode], items: &mut Vec<String>, indent_level: usize) -> miette::Result<()> {
let indent = " ".repeat(indent_level);
for node in nodes {
if let HtmlNode::Element(el) = node {
match el.tag_name.as_str() {
"option" => {
let text = convert_children_to_string(&el.children)?;
let text = text.trim();
let label = if !text.is_empty() {
text.to_string()
} else if let Some(Some(label_attr)) = el.attributes.get("label") {
label_attr.clone()
} else {
String::new()
};
if label.is_empty() {
continue;
}
let marker = if el.attributes.contains_key("selected") {
"[x]"
} else {
"[ ]"
};
items.push(format!("{}* {} {}", indent, marker, label));
}
"optgroup" => {
if let Some(Some(group_label)) = el.attributes.get("label")
&& !group_label.is_empty()
{
items.push(format!("{}* **{}**", indent, group_label));
}
collect_select_options(&el.children, items, indent_level + 1)?;
}
_ => {}
}
}
}
Ok(())
}

/// Converts `<textarea>` content to a fenced code block, mirroring `<pre>` since a
/// textarea's value is verbatim, whitespace-significant text.
fn handle_textarea_element(element: &HtmlElement) -> miette::Result<String> {
let text_content = extract_text_from_pre_children(&element.children);
let text_content = text_content.strip_prefix('\n').unwrap_or(&text_content);
let text_content = dedent(text_content.trim_end_matches('\n'));
if text_content.trim().is_empty() {
return Ok(String::new());
}
Ok(format!("```\n{}\n```", text_content))
}

fn handle_script_element(element: &HtmlElement, options: &ConversionOptions) -> miette::Result<Option<String>> {
if options.extract_scripts_as_code_blocks {
if element.attributes.get("src").and_then(|opt| opt.as_ref()).is_none() {
Expand Down Expand Up @@ -893,23 +953,35 @@ fn convert_children_to_string_impl(nodes: &[HtmlNode], escape_text: bool) -> mie
// Skip inputs used as CSS toggle triggers (not actual form inputs)
let is_ui_toggle = element.attributes.get("role").and_then(|v| v.as_deref()) == Some("button")
|| element.attributes.get("aria-haspopup").and_then(|v| v.as_deref()) == Some("true");
if is_ui_toggle {
// nothing
} else if let Some(Some(type_attr)) = element.attributes.get("type") {
match type_attr.to_lowercase().as_str() {
if !is_ui_toggle {
// No `type` attribute defaults to "text" per the HTML spec.
let type_attr = element
.attributes
.get("type")
.and_then(|opt| opt.as_deref())
.unwrap_or("text")
.to_lowercase();
match type_attr.as_str() {
"checkbox" | "radio" => {
if element.attributes.contains_key("checked") {
parts.push("[x] ".to_string());
} else {
parts.push("[ ] ".to_string());
}
}
"text" | "number" | "button" | "url" | "email"
if element.attributes.contains_key("value") =>
{
parts.push(element.attributes.get("value").cloned().unwrap().unwrap_or_default());
// No meaningful textual content to surface.
"hidden" | "file" | "image" => {}
_ => {
if let Some(Some(value)) = element.attributes.get("value")
&& !value.is_empty()
{
parts.push(value.clone());
} else if let Some(Some(placeholder)) = element.attributes.get("placeholder")
&& !placeholder.is_empty()
{
parts.push(format!("_{}_", placeholder));
}
}
_ => {}
}
}
}
Expand Down Expand Up @@ -1114,7 +1186,7 @@ pub fn convert_nodes_to_markdown(nodes: &[HtmlNode], options: &ConversionOptions
// Skip navigational/sidebar/noscript noise entirely
}
"html" | "head" | "header" | "footer" | "body" | "div" | "main" | "article" | "section"
| "hgroup" | "details" | "figure" => {
| "hgroup" | "details" | "figure" | "form" | "fieldset" => {
if is_css_dropdown_widget(element) {
// Skip CSS-only dropdown widgets (language switchers, nav menus, etc.)
} else if is_heading_with_aux_siblings(element) {
Expand Down Expand Up @@ -1153,6 +1225,25 @@ pub fn convert_nodes_to_markdown(nodes: &[HtmlNode], options: &ConversionOptions
markdown_blocks.push((dl_md, false));
}
}
"select" => {
let select_md = handle_select_element(element)?;
if !select_md.is_empty() {
markdown_blocks.push((select_md, false));
}
}
"textarea" => {
let textarea_md = handle_textarea_element(element)?;
if !textarea_md.is_empty() {
markdown_blocks.push((textarea_md, false));
}
}
"legend" => {
let legend_text = convert_children_to_string(&element.children)?;
let legend_text = legend_text.trim();
if !legend_text.is_empty() {
markdown_blocks.push((format!("**{}**", legend_text), false));
}
}
"summary" => {
let summary_text = convert_children_to_string(&element.children)?;
if !summary_text.is_empty() {
Expand Down
37 changes: 37 additions & 0 deletions crates/mq-markdown/tests/html_to_markdown_tests.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -609,6 +609,43 @@ fn assert_conversion_with_options(html: &str, expected_markdown: &str, options:
"* [x] Done"
)]
#[case::input_other_type_ignored("<input type=\"text\" value=\"Hello\">", ConversionOptions::default(), "Hello")]
#[case::input_no_type_defaults_to_text("<input value=\"Alice\">", ConversionOptions::default(), "Alice")]
#[case::input_placeholder_fallback("<input placeholder=\"Your name\">", ConversionOptions::default(), "_Your name_")]
#[case::input_value_takes_precedence_over_placeholder(
"<input value=\"Alice\" placeholder=\"Your name\">",
ConversionOptions::default(),
"Alice"
)]
#[case::input_hidden_ignored("<input type=\"hidden\" value=\"secret\">", ConversionOptions::default(), "")]
#[case::input_file_ignored("<input type=\"file\">", ConversionOptions::default(), "")]
#[case::input_range_with_value("<input type=\"range\" value=\"5\">", ConversionOptions::default(), "5")]
#[case::select_with_options(
"<select><option>Red</option><option selected>Green</option><option>Blue</option></select>",
ConversionOptions::default(),
"* [ ] Red\n* [x] Green\n* [ ] Blue"
)]
#[case::select_with_optgroup(
"<select><optgroup label=\"Fruits\"><option>Apple</option><option selected>Banana</option></optgroup></select>",
ConversionOptions::default(),
"* **Fruits**\n * [ ] Apple\n * [x] Banana"
)]
#[case::select_empty("<select></select>", ConversionOptions::default(), "")]
#[case::textarea_with_content(
"<textarea>line1\nline2</textarea>",
ConversionOptions::default(),
"```\nline1\nline2\n```"
)]
#[case::textarea_empty("<textarea></textarea>", ConversionOptions::default(), "")]
#[case::fieldset_legend(
"<fieldset><legend>Contact</legend><p>Name</p></fieldset>",
ConversionOptions::default(),
"**Contact**\n\nName"
)]
#[case::form_with_label_and_input(
"<form><label>Name: <input type=\"text\" value=\"World\"></label></form>",
ConversionOptions::default(),
"Name: World"
)]
#[case::dl_simple(
"<dl><dt>Term 1</dt><dd>Definition 1</dd></dl>",
ConversionOptions::default(),
Expand Down