internal: replace TreeSink with a data structure

The general theme of this is to make parser a better independent
library.

The specific thing we do here is replacing callback based TreeSink with
a data structure. That is, rather than calling user-provided tree
construction methods, the parser now spits out a very bare-bones tree,
effectively a log of a DFS traversal.

This makes the parser usable without any *specifc* tree sink, and allows
us to, eg, move tests into this crate.

Now, it's also true that this is a distinction without a difference, as
the old and the new interface are equivalent in expressiveness. Still,
this new thing seems somewhat simpler. But yeah, I admit I don't have a
suuper strong motivation here, just a hunch that this is better.
This commit is contained in:
Aleksey Kladov 2021-12-19 17:36:23 +03:00
parent 2f63558dc5
commit d0d05075ed
10 changed files with 172 additions and 110 deletions

View file

@ -10,7 +10,7 @@ use parser::Reparser;
use text_edit::Indel;
use crate::{
parsing::text_tree_sink::TextTreeSink,
parsing::text_tree_sink::build_tree,
syntax_node::{GreenNode, GreenToken, NodeOrToken, SyntaxElement, SyntaxNode},
SyntaxError,
SyntaxKind::*,
@ -94,11 +94,9 @@ fn reparse_block(
return None;
}
let mut tree_sink = TextTreeSink::new(lexed);
let tree_traversal = reparser.parse(&parser_tokens);
reparser.parse(&parser_tokens, &mut tree_sink);
let (green, new_parser_errors) = tree_sink.finish();
let (green, new_parser_errors, _eof) = build_tree(lexed, tree_traversal, false);
Some((node.replace_with(green), new_parser_errors, node.text_range()))
}

View file

@ -2,7 +2,7 @@
use std::mem;
use parser::{LexedStr, ParseError, TreeSink};
use parser::{LexedStr, TreeTraversal};
use crate::{
ast,
@ -12,6 +12,36 @@ use crate::{
SyntaxTreeBuilder, TextRange,
};
pub(crate) fn build_tree(
lexed: LexedStr<'_>,
tree_traversal: TreeTraversal,
synthetic_root: bool,
) -> (GreenNode, Vec<SyntaxError>, bool) {
let mut builder = TextTreeSink::new(lexed);
if synthetic_root {
builder.start_node(SyntaxKind::SOURCE_FILE);
}
for event in tree_traversal.iter() {
match event {
parser::TraversalStep::Token { kind, n_raw_tokens } => {
builder.token(kind, n_raw_tokens)
}
parser::TraversalStep::EnterNode { kind } => builder.start_node(kind),
parser::TraversalStep::LeaveNode => builder.finish_node(),
parser::TraversalStep::Error { msg } => {
let text_pos = builder.lexed.text_start(builder.pos).try_into().unwrap();
builder.inner.error(msg.to_string(), text_pos);
}
}
}
if synthetic_root {
builder.finish_node()
}
builder.finish_eof()
}
/// Bridges the parser with our specific syntax tree representation.
///
/// `TextTreeSink` also handles attachment of trivia (whitespace) to nodes.
@ -28,7 +58,7 @@ enum State {
PendingFinish,
}
impl<'a> TreeSink for TextTreeSink<'a> {
impl<'a> TextTreeSink<'a> {
fn token(&mut self, kind: SyntaxKind, n_tokens: u8) {
match mem::replace(&mut self.state, State::Normal) {
State::PendingStart => unreachable!(),
@ -70,11 +100,6 @@ impl<'a> TreeSink for TextTreeSink<'a> {
State::Normal => (),
}
}
fn error(&mut self, error: ParseError) {
let text_pos = self.lexed.text_start(self.pos).try_into().unwrap();
self.inner.error(error, text_pos);
}
}
impl<'a> TextTreeSink<'a> {
@ -106,11 +131,6 @@ impl<'a> TextTreeSink<'a> {
(node, errors, is_eof)
}
pub(super) fn finish(self) -> (GreenNode, Vec<SyntaxError>) {
let (node, errors, _eof) = self.finish_eof();
(node, errors)
}
fn eat_trivias(&mut self) {
while self.pos < self.lexed.len() {
let kind = self.lexed.kind(self.pos);