wdl_format/token/post.rs
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
//! Postprocessed tokens.
//!
//! Generally speaking, unless you are working with the internals of code
//! formatting, you're not going to be working with these.
use std::fmt::Display;
use wdl_ast::SyntaxKind;
use crate::Comment;
use crate::LineSpacingPolicy;
use crate::NEWLINE;
use crate::PreToken;
use crate::SPACE;
use crate::Token;
use crate::TokenStream;
use crate::Trivia;
use crate::config::Indent;
/// A postprocessed token.
#[derive(Eq, PartialEq)]
pub enum PostToken {
/// A space.
Space,
/// A newline.
Newline,
/// One indentation.
Indent,
/// A string literal.
Literal(String),
}
impl std::fmt::Debug for PostToken {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
Self::Space => write!(f, "<SPACE>"),
Self::Newline => write!(f, "<NEWLINE>"),
Self::Indent => write!(f, "<INDENT>"),
Self::Literal(value) => write!(f, "<LITERAL> {value}"),
}
}
}
impl Token for PostToken {
/// Returns a displayable version of the token.
fn display<'a>(&'a self, config: &'a crate::Config) -> impl Display + 'a {
/// A displayable version of a [`PostToken`].
struct Display<'a> {
/// The token to display.
token: &'a PostToken,
/// The configuration to use.
config: &'a crate::Config,
}
impl std::fmt::Display for Display<'_> {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self.token {
PostToken::Space => write!(f, "{SPACE}"),
PostToken::Newline => write!(f, "{NEWLINE}"),
PostToken::Indent => {
let (c, n) = match self.config.indent() {
Indent::Spaces(n) => (' ', n),
Indent::Tabs(n) => ('\t', n),
};
for _ in 0..n.get() {
write!(f, "{c}")?;
}
Ok(())
}
PostToken::Literal(value) => write!(f, "{value}"),
}
}
}
Display {
token: self,
config,
}
}
}
/// Current position in a line.
#[derive(Default, Eq, PartialEq)]
enum LinePosition {
/// The start of a line.
#[default]
StartOfLine,
/// The middle of a line.
MiddleOfLine,
}
/// A postprocessor of [tokens](PreToken).
#[derive(Default)]
pub struct Postprocessor {
/// The current position in the line.
position: LinePosition,
/// The current indentation level.
indent_level: usize,
/// Whether the current line has been interrupted by trivia.
interrupted: bool,
/// Whether blank lines are allowed in the current context.
line_spacing_policy: LineSpacingPolicy,
}
impl Postprocessor {
/// Runs the postprocessor.
pub fn run(&mut self, input: TokenStream<PreToken>) -> TokenStream<PostToken> {
let mut output = TokenStream::<PostToken>::default();
let mut stream = input.into_iter().peekable();
while let Some(token) = stream.next() {
self.step(token, stream.peek(), &mut output);
}
self.trim_whitespace(&mut output);
output.push(PostToken::Newline);
output
}
/// Takes a step of a [`PreToken`] stream and processes the appropriate
/// [`PostToken`]s.
pub fn step(
&mut self,
token: PreToken,
next: Option<&PreToken>,
stream: &mut TokenStream<PostToken>,
) {
match token {
PreToken::BlankLine => {
self.blank_line(stream);
}
PreToken::LineEnd => {
self.interrupted = false;
self.end_line(stream);
}
PreToken::WordEnd => {
stream.trim_end(&PostToken::Space);
if self.position == LinePosition::MiddleOfLine {
stream.push(PostToken::Space);
} else {
// We're at the start of a line, so we don't need to add a
// space.
}
}
PreToken::IndentStart => {
self.indent_level += 1;
self.end_line(stream);
}
PreToken::IndentEnd => {
self.indent_level = self.indent_level.saturating_sub(1);
self.end_line(stream);
}
PreToken::LineSpacingPolicy(policy) => {
self.line_spacing_policy = policy;
}
PreToken::Literal(value, kind) => {
assert!(kind != SyntaxKind::Comment && kind != SyntaxKind::Whitespace);
if self.interrupted
&& matches!(
kind,
SyntaxKind::OpenBrace
| SyntaxKind::OpenBracket
| SyntaxKind::OpenParen
| SyntaxKind::OpenHeredoc
)
&& stream.0.last() == Some(&PostToken::Indent)
{
stream.0.pop();
}
stream.push(PostToken::Literal(value));
self.position = LinePosition::MiddleOfLine;
}
PreToken::Trivia(trivia) => match trivia {
Trivia::BlankLine => match self.line_spacing_policy {
LineSpacingPolicy::Always => {
self.blank_line(stream);
}
LineSpacingPolicy::BeforeComments => {
if matches!(next, Some(&PreToken::Trivia(Trivia::Comment(_)))) {
self.blank_line(stream);
}
}
},
Trivia::Comment(comment) => {
match comment {
Comment::Preceding(value) => {
if !matches!(
stream.0.last(),
Some(&PostToken::Newline) | Some(&PostToken::Indent) | None
) {
self.interrupted = true;
}
self.end_line(stream);
stream.push(PostToken::Literal(value));
self.position = LinePosition::MiddleOfLine;
}
Comment::Inline(value) => {
assert!(self.position == LinePosition::MiddleOfLine);
if let Some(next) = next {
if next != &PreToken::LineEnd {
self.interrupted = true;
}
}
self.trim_last_line(stream);
stream.push(PostToken::Space);
stream.push(PostToken::Space);
stream.push(PostToken::Literal(value));
}
}
self.end_line(stream);
}
},
}
}
/// Trims any and all whitespace from the end of the stream.
fn trim_whitespace(&mut self, stream: &mut TokenStream<PostToken>) {
stream.trim_while(|token| {
matches!(
token,
PostToken::Space | PostToken::Newline | PostToken::Indent
)
});
}
/// Trims spaces and indents (and not newlines) from the end of the stream.
fn trim_last_line(&mut self, stream: &mut TokenStream<PostToken>) {
stream.trim_while(|token| matches!(token, PostToken::Space | PostToken::Indent));
}
/// Ends the current line without resetting the interrupted flag.
///
/// Removes any trailing spaces or indents and adds a newline only if state
/// is not [`LinePosition::StartOfLine`]. State is then set to
/// [`LinePosition::StartOfLine`]. Safe to call multiple times in a row.
fn end_line(&mut self, stream: &mut TokenStream<PostToken>) {
self.trim_last_line(stream);
if self.position != LinePosition::StartOfLine {
stream.push(PostToken::Newline);
}
self.position = LinePosition::StartOfLine;
self.indent(stream);
}
/// Pushes the current indentation level to the stream.
/// This should only be called when the state is
/// [`LinePosition::StartOfLine`].
fn indent(&self, stream: &mut TokenStream<PostToken>) {
assert!(self.position == LinePosition::StartOfLine);
let level = if self.interrupted {
self.indent_level + 1
} else {
self.indent_level
};
for _ in 0..level {
stream.push(PostToken::Indent);
}
}
/// Creates a blank line and then indents.
fn blank_line(&mut self, stream: &mut TokenStream<PostToken>) {
self.trim_whitespace(stream);
stream.push(PostToken::Newline);
stream.push(PostToken::Newline);
self.position = LinePosition::StartOfLine;
self.indent(stream);
}
}