typos/crates/wikipedia-dict/codegen/src/main.rs
Ed Page a1e95bc7c0 refactor(dict): Pull out table-lookup logic
Before, only some dicts did we guarentee were pre-sorted.  Now, all are
for-sure pre-sorted.

This also gives each dict the size-check to avoid lookup.

But this is really about refactoring in prep for playing with other
lookup options, like tries.
2021-06-30 10:12:17 -05:00

71 lines
1.6 KiB
Rust

use structopt::StructOpt;
pub const DICT: &str = include_str!("../../assets/dictionary.txt");
fn parse_dict(raw: &str) -> impl Iterator<Item = (&str, Vec<&str>)> {
raw.lines().map(|s| {
let mut parts = s.splitn(2, "->");
let typo = parts.next().unwrap().trim();
let corrections = parts
.next()
.unwrap()
.split(',')
.filter_map(|c| {
let c = c.trim();
if c.is_empty() {
None
} else {
Some(c)
}
})
.collect();
(typo, corrections)
})
}
fn generate<W: std::io::Write>(file: &mut W) {
writeln!(
file,
"// This file is code-genned by {}",
env!("CARGO_PKG_NAME")
)
.unwrap();
writeln!(file).unwrap();
let dict = parse_dict(DICT);
dictgen::generate_table(
file,
"WORD_DICTIONARY",
"&[&str]",
dict.map(|kv| (kv.0, format!("&{:?}", kv.1))),
)
.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);
}