Files
oak-gpui/crates/markdown_preview/src/markdown_elements.rs
T
Remco SmitsandBennet Bo Fenner f318bb5fd7 markdown: Add support for HTML href elements (#42265)
This PR adds support for `HTML` href elements. It also refactored the
way we stored the regions, this was done because otherwise I had to add
2 extra arguments to each `HTML` parser method. It's now also more
inline with how we have done it for the highlights.

**Small note**: the markdown parser only supports HTML href tags inside
a paragraph tag. So adding them as a root node will result in just
showing the inner text. This is a limitation of the markdown parser we
use itself.

**Before**
<img width="935" height="174" alt="Screenshot 2025-11-08 at 15 40 28"
src="https://github.com/user-attachments/assets/42172222-ed49-4a4b-8957-a46330e54c69"
/>

**After**
<img width="1026" height="180" alt="Screenshot 2025-11-08 at 15 29 55"
src="https://github.com/user-attachments/assets/9e139c2d-d43a-4952-8d1f-15eb92966241"
/>

**Example code**
```markdown
<p>asd <a href="https://example.com">Link Text</a> more text</p>
<p><a href="https://example.com">Link Text</a></p>

[Duck Duck Go](https://duckduckgo.com)
```

**TODO**:
- [x] Add tests

cc @bennetbo

Release Notes:

- Markdown Preview: Add support for `HTML` href elements.

---------

Co-authored-by: Bennet Bo Fenner <bennet@zed.dev>
2025-11-13 15:12:17 +01:00

359 lines
9.8 KiB
Rust

use gpui::{
DefiniteLength, FontStyle, FontWeight, HighlightStyle, SharedString, StrikethroughStyle,
UnderlineStyle, px,
};
use language::HighlightId;
use std::{fmt::Display, ops::Range, path::PathBuf};
use urlencoding;
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub enum ParsedMarkdownElement {
Heading(ParsedMarkdownHeading),
ListItem(ParsedMarkdownListItem),
Table(ParsedMarkdownTable),
BlockQuote(ParsedMarkdownBlockQuote),
CodeBlock(ParsedMarkdownCodeBlock),
/// A paragraph of text and other inline elements.
Paragraph(MarkdownParagraph),
HorizontalRule(Range<usize>),
Image(Image),
}
impl ParsedMarkdownElement {
pub fn source_range(&self) -> Option<Range<usize>> {
Some(match self {
Self::Heading(heading) => heading.source_range.clone(),
Self::ListItem(list_item) => list_item.source_range.clone(),
Self::Table(table) => table.source_range.clone(),
Self::BlockQuote(block_quote) => block_quote.source_range.clone(),
Self::CodeBlock(code_block) => code_block.source_range.clone(),
Self::Paragraph(text) => match text.get(0)? {
MarkdownParagraphChunk::Text(t) => t.source_range.clone(),
MarkdownParagraphChunk::Image(image) => image.source_range.clone(),
},
Self::HorizontalRule(range) => range.clone(),
Self::Image(image) => image.source_range.clone(),
})
}
pub fn is_list_item(&self) -> bool {
matches!(self, Self::ListItem(_))
}
}
pub type MarkdownParagraph = Vec<MarkdownParagraphChunk>;
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub enum MarkdownParagraphChunk {
Text(ParsedMarkdownText),
Image(Image),
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedMarkdown {
pub children: Vec<ParsedMarkdownElement>,
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedMarkdownListItem {
pub source_range: Range<usize>,
/// How many indentations deep this item is.
pub depth: u16,
pub item_type: ParsedMarkdownListItemType,
pub content: Vec<ParsedMarkdownElement>,
/// Whether we can expect nested list items inside of this items `content`.
pub nested: bool,
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub enum ParsedMarkdownListItemType {
Ordered(u64),
Task(bool, Range<usize>),
Unordered,
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedMarkdownCodeBlock {
pub source_range: Range<usize>,
pub language: Option<String>,
pub contents: SharedString,
pub highlights: Option<Vec<(Range<usize>, HighlightId)>>,
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedMarkdownHeading {
pub source_range: Range<usize>,
pub level: HeadingLevel,
pub contents: MarkdownParagraph,
}
#[derive(Debug, PartialEq)]
pub enum HeadingLevel {
H1,
H2,
H3,
H4,
H5,
H6,
}
#[derive(Debug)]
pub struct ParsedMarkdownTable {
pub source_range: Range<usize>,
pub header: Vec<ParsedMarkdownTableRow>,
pub body: Vec<ParsedMarkdownTableRow>,
pub caption: Option<MarkdownParagraph>,
}
#[derive(Debug, Clone, Copy, Default)]
#[cfg_attr(test, derive(PartialEq))]
pub enum ParsedMarkdownTableAlignment {
#[default]
None,
Left,
Center,
Right,
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedMarkdownTableColumn {
pub col_span: usize,
pub row_span: usize,
pub is_header: bool,
pub children: MarkdownParagraph,
pub alignment: ParsedMarkdownTableAlignment,
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedMarkdownTableRow {
pub columns: Vec<ParsedMarkdownTableColumn>,
}
impl Default for ParsedMarkdownTableRow {
fn default() -> Self {
Self::new()
}
}
impl ParsedMarkdownTableRow {
pub fn new() -> Self {
Self {
columns: Vec::new(),
}
}
pub fn with_columns(columns: Vec<ParsedMarkdownTableColumn>) -> Self {
Self { columns }
}
}
#[derive(Debug)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedMarkdownBlockQuote {
pub source_range: Range<usize>,
pub children: Vec<ParsedMarkdownElement>,
}
#[derive(Debug, Clone)]
pub struct ParsedMarkdownText {
/// Where the text is located in the source Markdown document.
pub source_range: Range<usize>,
/// The text content stripped of any formatting symbols.
pub contents: SharedString,
/// The list of highlights contained in the Markdown document.
pub highlights: Vec<(Range<usize>, MarkdownHighlight)>,
/// The regions of the Markdown document.
pub regions: Vec<(Range<usize>, ParsedRegion)>,
}
/// A run of highlighted Markdown text.
#[derive(Debug, Clone, PartialEq, Eq)]
pub enum MarkdownHighlight {
/// A styled Markdown highlight.
Style(MarkdownHighlightStyle),
/// A highlighted code block.
Code(HighlightId),
}
impl MarkdownHighlight {
/// Converts this [`MarkdownHighlight`] to a [`HighlightStyle`].
pub fn to_highlight_style(&self, theme: &theme::SyntaxTheme) -> Option<HighlightStyle> {
match self {
MarkdownHighlight::Style(style) => {
let mut highlight = HighlightStyle::default();
if style.italic {
highlight.font_style = Some(FontStyle::Italic);
}
if style.underline {
highlight.underline = Some(UnderlineStyle {
thickness: px(1.),
..Default::default()
});
}
if style.strikethrough {
highlight.strikethrough = Some(StrikethroughStyle {
thickness: px(1.),
..Default::default()
});
}
if style.weight != FontWeight::default() {
highlight.font_weight = Some(style.weight);
}
if style.link {
highlight.underline = Some(UnderlineStyle {
thickness: px(1.),
..Default::default()
});
}
if style.oblique {
highlight.font_style = Some(FontStyle::Oblique)
}
Some(highlight)
}
MarkdownHighlight::Code(id) => id.style(theme),
}
}
}
/// The style for a Markdown highlight.
#[derive(Debug, Clone, Default, PartialEq, Eq)]
pub struct MarkdownHighlightStyle {
/// Whether the text should be italicized.
pub italic: bool,
/// Whether the text should be underlined.
pub underline: bool,
/// Whether the text should be struck through.
pub strikethrough: bool,
/// The weight of the text.
pub weight: FontWeight,
/// Whether the text should be stylized as link.
pub link: bool,
// Whether the text should be obliqued.
pub oblique: bool,
}
/// A parsed region in a Markdown document.
#[derive(Debug, Clone)]
#[cfg_attr(test, derive(PartialEq))]
pub struct ParsedRegion {
/// Whether the region is a code block.
pub code: bool,
/// The link contained in this region, if it has one.
pub link: Option<Link>,
}
/// A Markdown link.
#[derive(Debug, Clone)]
#[cfg_attr(test, derive(PartialEq))]
pub enum Link {
/// A link to a webpage.
Web {
/// The URL of the webpage.
url: String,
},
/// A link to a path on the filesystem.
Path {
/// The path as provided in the Markdown document.
display_path: PathBuf,
/// The absolute path to the item.
path: PathBuf,
},
}
impl Link {
pub fn identify(file_location_directory: Option<PathBuf>, text: String) -> Option<Link> {
if text.starts_with("http") {
return Some(Link::Web { url: text });
}
// URL decode the text to handle spaces and other special characters
let decoded_text = urlencoding::decode(&text)
.map(|s| s.into_owned())
.unwrap_or(text);
let path = PathBuf::from(&decoded_text);
if path.is_absolute() && path.exists() {
return Some(Link::Path {
display_path: path.clone(),
path,
});
}
if let Some(file_location_directory) = file_location_directory {
let display_path = path;
let path = file_location_directory.join(decoded_text);
if path.exists() {
return Some(Link::Path { display_path, path });
}
}
None
}
}
impl Display for Link {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
Link::Web { url } => write!(f, "{}", url),
Link::Path { display_path, .. } => write!(f, "{}", display_path.display()),
}
}
}
/// A Markdown Image
#[derive(Debug, Clone)]
#[cfg_attr(test, derive(PartialEq))]
pub struct Image {
pub link: Link,
pub source_range: Range<usize>,
pub alt_text: Option<SharedString>,
pub width: Option<DefiniteLength>,
pub height: Option<DefiniteLength>,
}
impl Image {
pub fn identify(
text: String,
source_range: Range<usize>,
file_location_directory: Option<PathBuf>,
) -> Option<Self> {
let link = Link::identify(file_location_directory, text)?;
Some(Self {
source_range,
link,
alt_text: None,
width: None,
height: None,
})
}
pub fn set_alt_text(&mut self, alt_text: SharedString) {
self.alt_text = Some(alt_text);
}
pub fn set_width(&mut self, width: DefiniteLength) {
self.width = Some(width);
}
pub fn set_height(&mut self, height: DefiniteLength) {
self.height = Some(height);
}
}