Support raw strings in lexer

This commit is contained in:
Aleksey Kladov 2018-08-14 15:03:27 +03:00
parent 199e3b73c7
commit 109658332a
4 changed files with 52 additions and 34 deletions

View file

@ -20,7 +20,6 @@ use std::{
atomic::{AtomicUsize, Ordering::SeqCst}, atomic::{AtomicUsize, Ordering::SeqCst},
}, },
collections::hash_map::HashMap, collections::hash_map::HashMap,
path::{PathBuf, Path},
time::Instant, time::Instant,
}; };
@ -38,6 +37,7 @@ pub type Result<T> = ::std::result::Result<T, ::failure::Error>;
const INDEXING_THRESHOLD: usize = 128; const INDEXING_THRESHOLD: usize = 128;
pub struct WorldState { pub struct WorldState {
next_file_id: u32,
data: Arc<WorldData> data: Arc<WorldData>
} }
@ -46,9 +46,13 @@ pub struct World {
data: Arc<WorldData>, data: Arc<WorldData>,
} }
#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord, Hash)]
pub struct FileId(u32);
impl WorldState { impl WorldState {
pub fn new() -> WorldState { pub fn new() -> WorldState {
WorldState { WorldState {
next_file_id: 0,
data: Arc::new(WorldData::default()) data: Arc::new(WorldData::default())
} }
} }
@ -57,21 +61,27 @@ impl WorldState {
World { data: self.data.clone() } World { data: self.data.clone() }
} }
pub fn change_file(&mut self, path: PathBuf, text: Option<String>) { pub fn new_file_id(&mut self) -> FileId {
self.change_files(::std::iter::once((path, text))); let id = FileId(self.next_file_id);
self.next_file_id += 1;
id
} }
pub fn change_files(&mut self, changes: impl Iterator<Item=(PathBuf, Option<String>)>) { pub fn change_file(&mut self, file_id: FileId, text: Option<String>) {
self.change_files(::std::iter::once((file_id, text)));
}
pub fn change_files(&mut self, changes: impl Iterator<Item=(FileId, Option<String>)>) {
let data = self.data_mut(); let data = self.data_mut();
let mut cnt = 0; let mut cnt = 0;
for (path, text) in changes { for (id, text) in changes {
cnt += 1; cnt += 1;
data.file_map.remove(&path); data.file_map.remove(&id);
if let Some(text) = text { if let Some(text) = text {
let file_data = FileData::new(text); let file_data = FileData::new(text);
data.file_map.insert(path, Arc::new(file_data)); data.file_map.insert(id, Arc::new(file_data));
} else { } else {
data.file_map.remove(&path); data.file_map.remove(&id);
} }
} }
*data.unindexed.get_mut() += cnt; *data.unindexed.get_mut() += cnt;
@ -92,37 +102,33 @@ impl WorldState {
impl World { impl World {
pub fn file_syntax(&self, path: &Path) -> Result<ast::File> { pub fn file_syntax(&self, file_id: FileId) -> Result<ast::File> {
let data = self.file_data(path)?; let data = self.file_data(file_id)?;
Ok(data.syntax(path).clone()) Ok(data.syntax().clone())
} }
pub fn file_line_index(&self, path: &Path) -> Result<LineIndex> { pub fn file_line_index(&self, id: FileId) -> Result<LineIndex> {
let data = self.file_data(path)?; let data = self.file_data(id)?;
let index = data.lines let index = data.lines
.get_or_init(|| { .get_or_init(|| LineIndex::new(&data.text));
trace!("calc line index: {}", path.display());
LineIndex::new(&data.text)
});
Ok(index.clone()) Ok(index.clone())
} }
pub fn world_symbols<'a>(&'a self, mut query: Query) -> impl Iterator<Item=(&'a Path, &'a FileSymbol)> + 'a { pub fn world_symbols<'a>(&'a self, mut query: Query) -> impl Iterator<Item=(FileId, &'a FileSymbol)> + 'a {
self.reindex(); self.reindex();
self.data.file_map.iter() self.data.file_map.iter()
.flat_map(move |(path, data)| { .flat_map(move |(id, data)| {
let symbols = data.symbols(); let symbols = data.symbols();
let path: &'a Path = path.as_path(); query.process(symbols).into_iter().map(move |s| (*id, s))
query.process(symbols).into_iter().map(move |s| (path, s))
}) })
} }
pub fn approximately_resolve_symbol<'a>( pub fn approximately_resolve_symbol<'a>(
&'a self, &'a self,
path: &Path, id: FileId,
offset: TextUnit, offset: TextUnit,
) -> Result<Vec<(&'a Path, &'a FileSymbol)>> { ) -> Result<Vec<(FileId, &'a FileSymbol)>> {
let file = self.file_syntax(path)?; let file = self.file_syntax(id)?;
let syntax = file.syntax(); let syntax = file.syntax();
let syntax = syntax.as_ref(); let syntax = syntax.as_ref();
let name_ref = let name_ref =
@ -160,10 +166,10 @@ impl World {
} }
} }
fn file_data(&self, path: &Path) -> Result<Arc<FileData>> { fn file_data(&self, file_id: FileId) -> Result<Arc<FileData>> {
match self.data.file_map.get(path) { match self.data.file_map.get(&file_id) {
Some(data) => Ok(data.clone()), Some(data) => Ok(data.clone()),
None => bail!("unknown file: {}", path.display()), None => bail!("unknown file: {:?}", file_id),
} }
} }
} }
@ -182,7 +188,7 @@ pub const BREAK: SearchResult = Err(Break);
#[derive(Default, Debug)] #[derive(Default, Debug)]
struct WorldData { struct WorldData {
unindexed: AtomicUsize, unindexed: AtomicUsize,
file_map: HashMap<PathBuf, Arc<FileData>>, file_map: HashMap<FileId, Arc<FileData>>,
} }
#[derive(Debug)] #[derive(Debug)]
@ -203,12 +209,9 @@ impl FileData {
} }
} }
fn syntax(&self, path: &Path) -> &ast::File { fn syntax(&self) -> &ast::File {
self.syntax self.syntax
.get_or_init(|| { .get_or_init(|| ast::File::parse(&self.text))
trace!("parsing: {}", path.display());
ast::File::parse(&self.text)
})
} }
fn syntax_transient(&self) -> ast::File { fn syntax_transient(&self) -> ast::File {

View file

@ -55,6 +55,11 @@ pub(crate) fn scan_string(ptr: &mut Ptr) {
} }
pub(crate) fn scan_raw_string(ptr: &mut Ptr) { pub(crate) fn scan_raw_string(ptr: &mut Ptr) {
let mut hashes = 0;
while ptr.next_is('#') {
hashes += 1;
ptr.bump();
}
if !ptr.next_is('"') { if !ptr.next_is('"') {
return; return;
} }
@ -62,7 +67,14 @@ pub(crate) fn scan_raw_string(ptr: &mut Ptr) {
while let Some(c) = ptr.bump() { while let Some(c) = ptr.bump() {
if c == '"' { if c == '"' {
return; let mut hashes_left = hashes;
while ptr.next_is('#') && hashes_left > 0{
hashes_left -= 1;
ptr.bump();
}
if hashes_left == 0 {
return;
}
} }
} }
} }

View file

@ -0,0 +1 @@
r###"this is a r##"raw"## string"###

View file

@ -0,0 +1,2 @@
RAW_STRING 36 "r###\"this is a r##\"raw\"## string\"###"
WHITESPACE 1 "\n"