mirror of
https://github.com/crate-ci/typos.git
synced 2024-12-01 21:40:55 -05:00
beaa0f4091
Bypass hashing when we know (through str::len) that a word won't be in the dict. Master: ``` real 0m26.675s user 0m33.683s sys 0m4.535s ``` With this change: ``` real 0m24.060s user 0m31.559s sys 0m4.258s ```
70 lines
2 KiB
Rust
70 lines
2 KiB
Rust
use structopt::StructOpt;
|
|
|
|
const DICT: &[u8] = include_bytes!("../../assets/words.csv");
|
|
|
|
fn generate<W: std::io::Write>(file: &mut W) {
|
|
writeln!(
|
|
file,
|
|
"// This file is code-genned by {}",
|
|
env!("CARGO_PKG_NAME")
|
|
)
|
|
.unwrap();
|
|
writeln!(file, "#![allow(clippy::unreadable_literal)]",).unwrap();
|
|
writeln!(file).unwrap();
|
|
writeln!(file, "use unicase::UniCase;").unwrap();
|
|
|
|
let mut smallest = usize::MAX;
|
|
let mut largest = usize::MIN;
|
|
|
|
writeln!(
|
|
file,
|
|
"pub static WORD_DICTIONARY: phf::Map<unicase::UniCase<&'static str>, &'static str> = "
|
|
)
|
|
.unwrap();
|
|
let mut builder = phf_codegen::Map::new();
|
|
let records: Vec<_> = csv::ReaderBuilder::new()
|
|
.has_headers(false)
|
|
.from_reader(DICT)
|
|
.records()
|
|
.map(|r| r.unwrap())
|
|
.collect();
|
|
for record in &records {
|
|
smallest = std::cmp::min(smallest, record[0].len());
|
|
largest = std::cmp::max(largest, record[0].len());
|
|
let value = format!(r#""{}""#, &record[1]);
|
|
builder.entry(unicase::UniCase::new(&record[0]), &value);
|
|
}
|
|
let codegenned = builder.build();
|
|
writeln!(file, "{}", codegenned).unwrap();
|
|
writeln!(file, ";").unwrap();
|
|
writeln!(file).unwrap();
|
|
writeln!(file, "pub const WORD_MIN: usize = {};", smallest).unwrap();
|
|
writeln!(file, "pub const WORD_MAX: usize = {};", largest).unwrap();
|
|
}
|
|
|
|
#[derive(Debug, StructOpt)]
|
|
#[structopt(rename_all = "kebab-case")]
|
|
struct Options {
|
|
#[structopt(flatten)]
|
|
codegen: codegenrs::CodeGenArgs,
|
|
#[structopt(flatten)]
|
|
rustmft: codegenrs::RustfmtArgs,
|
|
}
|
|
|
|
fn run() -> Result<i32, Box<dyn std::error::Error>> {
|
|
let options = Options::from_args();
|
|
|
|
let mut content = vec![];
|
|
generate(&mut content);
|
|
|
|
let content = String::from_utf8(content)?;
|
|
let content = options.rustmft.reformat(&content)?;
|
|
options.codegen.write_str(&content)?;
|
|
|
|
Ok(0)
|
|
}
|
|
|
|
fn main() {
|
|
let code = run().unwrap();
|
|
std::process::exit(code);
|
|
}
|