Skip to content

fix: handle escaped chars in doc comments #17024

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.

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
merged 2 commits into from
Apr 19, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
6 changes: 5 additions & 1 deletion crates/hir-def/src/attr.rs
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@ pub mod builtin;
#[cfg(test)]
mod tests;

use std::{hash::Hash, ops, slice::Iter as SliceIter};
use std::{borrow::Cow, hash::Hash, ops, slice::Iter as SliceIter};

use base_db::CrateId;
use cfg::{CfgExpr, CfgOptions};
Expand Down Expand Up @@ -573,6 +573,10 @@ impl<'attr> AttrQuery<'attr> {
self.attrs().find_map(|attr| attr.string_value())
}

pub fn string_value_unescape(self) -> Option<Cow<'attr, str>> {
self.attrs().find_map(|attr| attr.string_value_unescape())
}

pub fn exists(self) -> bool {
self.attrs().next().is_some()
}
Expand Down
13 changes: 9 additions & 4 deletions crates/hir-def/src/nameres/collector.rs
Original file line number Diff line number Diff line change
Expand Up @@ -1917,7 +1917,7 @@ impl ModCollector<'_, '_> {
}

fn collect_module(&mut self, module_id: FileItemTreeId<Mod>, attrs: &Attrs) {
let path_attr = attrs.by_key("path").string_value();
let path_attr = attrs.by_key("path").string_value_unescape();
let is_macro_use = attrs.by_key("macro_use").exists();
let module = &self.item_tree[module_id];
match &module.kind {
Expand All @@ -1931,7 +1931,8 @@ impl ModCollector<'_, '_> {
module_id,
);

let Some(mod_dir) = self.mod_dir.descend_into_definition(&module.name, path_attr)
let Some(mod_dir) =
self.mod_dir.descend_into_definition(&module.name, path_attr.as_deref())
else {
return;
};
Expand All @@ -1952,8 +1953,12 @@ impl ModCollector<'_, '_> {
ModKind::Outline => {
let ast_id = AstId::new(self.file_id(), module.ast_id);
let db = self.def_collector.db;
match self.mod_dir.resolve_declaration(db, self.file_id(), &module.name, path_attr)
{
match self.mod_dir.resolve_declaration(
db,
self.file_id(),
&module.name,
path_attr.as_deref(),
) {
Ok((file_id, is_mod_rs, mod_dir)) => {
let item_tree = db.file_item_tree(file_id.into());
let krate = self.def_collector.def_map.krate;
Expand Down
49 changes: 46 additions & 3 deletions crates/hir-expand/src/attrs.rs
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
//! A higher level attributes based on TokenTree, with also some shortcuts.
use std::{fmt, ops};
use std::{borrow::Cow, fmt, ops};

use base_db::CrateId;
use cfg::CfgExpr;
Expand All @@ -8,6 +8,7 @@ use intern::Interned;
use mbe::{syntax_node_to_token_tree, DelimiterKind, Punct};
use smallvec::{smallvec, SmallVec};
use span::{Span, SyntaxContextId};
use syntax::unescape;
use syntax::{ast, format_smolstr, match_ast, AstNode, AstToken, SmolStr, SyntaxNode};
use triomphe::ThinArc;

Expand Down Expand Up @@ -54,8 +55,7 @@ impl RawAttrs {
Attr {
id,
input: Some(Interned::new(AttrInput::Literal(tt::Literal {
// FIXME: Escape quotes from comment content
text: SmolStr::new(format_smolstr!("\"{doc}\"",)),
text: SmolStr::new(format_smolstr!("\"{}\"", Self::escape_chars(doc))),
span,
}))),
path: Interned::new(ModPath::from(crate::name!(doc))),
Expand All @@ -74,6 +74,10 @@ impl RawAttrs {
RawAttrs { entries }
}

fn escape_chars(s: &str) -> String {
s.replace('\\', r#"\\"#).replace('"', r#"\""#)
}

pub fn from_attrs_owner(
db: &dyn ExpandDatabase,
owner: InFile<&dyn ast::HasAttrs>,
Expand Down Expand Up @@ -297,6 +301,18 @@ impl Attr {
}
}

pub fn string_value_unescape(&self) -> Option<Cow<'_, str>> {
match self.input.as_deref()? {
AttrInput::Literal(it) => match it.text.strip_prefix('r') {
Some(it) => {
it.trim_matches('#').strip_prefix('"')?.strip_suffix('"').map(Cow::Borrowed)
}
None => it.text.strip_prefix('"')?.strip_suffix('"').and_then(unescape),
},
_ => None,
}
}

/// #[path(ident)]
pub fn single_ident_value(&self) -> Option<&tt::Ident> {
match self.input.as_deref()? {
Expand Down Expand Up @@ -346,6 +362,33 @@ impl Attr {
}
}

fn unescape(s: &str) -> Option<Cow<'_, str>> {
let mut buf = String::new();
let mut prev_end = 0;
let mut has_error = false;
unescape::unescape_unicode(s, unescape::Mode::Str, &mut |char_range, unescaped_char| match (
unescaped_char,
buf.capacity() == 0,
) {
(Ok(c), false) => buf.push(c),
(Ok(_), true) if char_range.len() == 1 && char_range.start == prev_end => {
prev_end = char_range.end
}
(Ok(c), true) => {
buf.reserve_exact(s.len());
buf.push_str(&s[..prev_end]);
buf.push(c);
}
(Err(_), _) => has_error = true,
});

match (has_error, buf.capacity() == 0) {
(true, _) => None,
(false, false) => Some(Cow::Owned(buf)),
(false, true) => Some(Cow::Borrowed(s)),
}
}

pub fn collect_attrs(
owner: &dyn ast::HasAttrs,
) -> impl Iterator<Item = (AttrId, Either<ast::Attr, ast::Comment>)> {
Expand Down
13 changes: 7 additions & 6 deletions crates/ide-db/src/documentation.rs
Original file line number Diff line number Diff line change
Expand Up @@ -91,8 +91,10 @@ pub fn docs_with_rangemap(
db: &dyn DefDatabase,
attrs: &AttrsWithOwner,
) -> Option<(Documentation, DocsRangeMap)> {
let docs =
attrs.by_key("doc").attrs().filter_map(|attr| attr.string_value().map(|s| (s, attr.id)));
let docs = attrs
.by_key("doc")
.attrs()
.filter_map(|attr| attr.string_value_unescape().map(|s| (s, attr.id)));
let indent = doc_indent(attrs);
let mut buf = String::new();
let mut mapping = Vec::new();
Expand Down Expand Up @@ -132,7 +134,7 @@ pub fn docs_with_rangemap(
}

pub fn docs_from_attrs(attrs: &hir::Attrs) -> Option<String> {
let docs = attrs.by_key("doc").attrs().filter_map(|attr| attr.string_value());
let docs = attrs.by_key("doc").attrs().filter_map(|attr| attr.string_value_unescape());
let indent = doc_indent(attrs);
let mut buf = String::new();
for doc in docs {
Expand Down Expand Up @@ -270,10 +272,9 @@ fn doc_indent(attrs: &hir::Attrs) -> usize {
attrs
.by_key("doc")
.attrs()
.filter_map(|attr| attr.string_value())
.filter_map(|attr| attr.string_value()) // no need to use unescape version here
.flat_map(|s| s.lines())
.filter(|line| !line.chars().all(|c| c.is_whitespace()))
.map(|line| line.chars().take_while(|c| c.is_whitespace()).count())
.filter_map(|line| line.chars().position(|c| !c.is_whitespace()))
.min()
.unwrap_or(0)
}
1 change: 1 addition & 0 deletions crates/syntax/src/lib.rs
Original file line number Diff line number Diff line change
Expand Up @@ -65,6 +65,7 @@ pub use rowan::{
api::Preorder, Direction, GreenNode, NodeOrToken, SyntaxText, TextRange, TextSize,
TokenAtOffset, WalkEvent,
};
pub use rustc_lexer::unescape;
pub use smol_str::{format_smolstr, SmolStr};

/// `Parse` is the result of the parsing: a syntax tree and a collection of
Expand Down