1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
|
//! An experimental implementation of [Rust RFC#2256 libsyntax2.0][rfc#2256].
//!
//! The intent is to be an IDE-ready parser, i.e. one that offers
//!
//! - easy and fast incremental re-parsing,
//! - graceful handling of errors, and
//! - maintains all information in the source file.
//!
//! For more information, see [the RFC][rfc#2265], or [the working draft][RFC.md].
//!
//! [rfc#2256]: <https://github.com/rust-lang/rfcs/pull/2256>
//! [RFC.md]: <https://github.com/matklad/libsyntax2/blob/master/docs/RFC.md>
#![forbid(
missing_debug_implementations,
unconditional_recursion,
future_incompatible
)]
#![deny(bad_style, missing_docs)]
#![allow(missing_docs)]
//#![warn(unreachable_pub)] // rust-lang/rust#47816
extern crate itertools;
extern crate unicode_xid;
extern crate drop_bomb;
extern crate parking_lot;
extern crate smol_str;
extern crate text_unit;
pub mod algo;
pub mod ast;
mod lexer;
#[macro_use]
mod parser_api;
mod grammar;
mod parser_impl;
mod syntax_kinds;
mod yellow;
/// Utilities for simple uses of the parser.
pub mod utils;
pub mod text_utils;
pub use {
text_unit::{TextRange, TextUnit},
smol_str::SmolStr,
ast::AstNode,
lexer::{tokenize, Token},
syntax_kinds::SyntaxKind,
yellow::{SyntaxNode, SyntaxNodeRef, OwnedRoot, RefRoot, TreeRoot, SyntaxError},
};
use {
SyntaxKind::*,
yellow::{GreenNode, SyntaxRoot},
parser_api::Parser,
};
#[derive(Clone, Debug)]
pub struct File {
root: SyntaxNode
}
impl File {
fn new(green: GreenNode, errors: Vec<SyntaxError>) -> File {
let root = SyntaxRoot::new(green, errors);
let root = SyntaxNode::new_owned(root);
validate_block_structure(root.borrowed());
File { root }
}
pub fn parse(text: &str) -> File {
let tokens = tokenize(&text);
let (green, errors) = parser_impl::parse_with::<yellow::GreenBuilder>(
text, &tokens, grammar::root,
);
File::new(green, errors)
}
pub fn reparse(&self, edit: &AtomEdit) -> File {
self.incremental_reparse(edit).unwrap_or_else(|| self.full_reparse(edit))
}
pub fn incremental_reparse(&self, edit: &AtomEdit) -> Option<File> {
let (node, reparser) = find_reparsable_node(self.syntax(), edit.delete)?;
let text = replace_range(
node.text(),
edit.delete - node.range().start(),
&edit.insert,
);
let tokens = tokenize(&text);
if !is_balanced(&tokens) {
return None;
}
let (green, new_errors) = parser_impl::parse_with::<yellow::GreenBuilder>(
&text, &tokens, reparser,
);
let green_root = node.replace_with(green);
let errors = merge_errors(self.errors(), new_errors, node, edit);
Some(File::new(green_root, errors))
}
fn full_reparse(&self, edit: &AtomEdit) -> File {
let text = replace_range(self.syntax().text(), edit.delete, &edit.insert);
File::parse(&text)
}
pub fn ast(&self) -> ast::Root {
ast::Root::cast(self.syntax()).unwrap()
}
pub fn syntax(&self) -> SyntaxNodeRef {
self.root.borrowed()
}
pub fn errors(&self) -> Vec<SyntaxError> {
self.syntax().root.syntax_root().errors.clone()
}
}
#[cfg(not(debug_assertions))]
fn validate_block_structure(_: SyntaxNodeRef) {}
#[cfg(debug_assertions)]
fn validate_block_structure(root: SyntaxNodeRef) {
let mut stack = Vec::new();
for node in algo::walk::preorder(root) {
match node.kind() {
SyntaxKind::L_CURLY => {
stack.push(node)
}
SyntaxKind::R_CURLY => {
if let Some(pair) = stack.pop() {
assert_eq!(
node.parent(),
pair.parent(),
"unpaired curleys:\n{}",
utils::dump_tree(root),
);
assert!(
node.next_sibling().is_none() && pair.prev_sibling().is_none(),
"floating curlys at {:?}\nfile:\n{}\nerror:\n{}\n",
node,
root.text(),
node.text(),
);
}
}
_ => (),
}
}
}
#[derive(Debug, Clone)]
pub struct AtomEdit {
pub delete: TextRange,
pub insert: String,
}
impl AtomEdit {
pub fn replace(range: TextRange, replace_with: String) -> AtomEdit {
AtomEdit { delete: range, insert: replace_with }
}
pub fn delete(range: TextRange) -> AtomEdit {
AtomEdit::replace(range, String::new())
}
pub fn insert(offset: TextUnit, text: String) -> AtomEdit {
AtomEdit::replace(TextRange::offset_len(offset, 0.into()), text)
}
}
fn find_reparsable_node(node: SyntaxNodeRef, range: TextRange) -> Option<(SyntaxNodeRef, fn(&mut Parser))> {
let node = algo::find_covering_node(node, range);
return algo::ancestors(node)
.filter_map(|node| reparser(node).map(|r| (node, r)))
.next();
fn reparser(node: SyntaxNodeRef) -> Option<fn(&mut Parser)> {
let res = match node.kind() {
BLOCK => grammar::block,
NAMED_FIELD_DEF_LIST => grammar::named_field_def_list,
_ => return None,
};
Some(res)
}
}
pub /*(meh)*/ fn replace_range(mut text: String, range: TextRange, replace_with: &str) -> String {
let start = u32::from(range.start()) as usize;
let end = u32::from(range.end()) as usize;
text.replace_range(start..end, replace_with);
text
}
fn is_balanced(tokens: &[Token]) -> bool {
if tokens.len() == 0
|| tokens.first().unwrap().kind != L_CURLY
|| tokens.last().unwrap().kind != R_CURLY {
return false
}
let mut balance = 0usize;
for t in tokens.iter() {
match t.kind {
L_CURLY => balance += 1,
R_CURLY => balance = match balance.checked_sub(1) {
Some(b) => b,
None => return false,
},
_ => (),
}
}
balance == 0
}
fn merge_errors(
old_errors: Vec<SyntaxError>,
new_errors: Vec<SyntaxError>,
old_node: SyntaxNodeRef,
edit: &AtomEdit,
) -> Vec<SyntaxError> {
let mut res = Vec::new();
for e in old_errors {
if e.offset < old_node.range().start() {
res.push(e)
} else if e.offset > old_node.range().end() {
res.push(SyntaxError {
msg: e.msg,
offset: e.offset + TextUnit::of_str(&edit.insert) - edit.delete.len(),
})
}
}
for e in new_errors {
res.push(SyntaxError {
msg: e.msg,
offset: e.offset + old_node.range().start(),
})
}
res
}
|