isspam/jest_rust/src/main.rs

mod stats;
mod trie;

use stats::Stats;
use std::{cell::LazyCell, env, fs};
use tokio::sync::mpsc;
use trie::Trie;

thread_local! {
	static FORBIDON: LazyCell<Trie> = LazyCell::new(|| {
		let mut trie = Trie::default();
		for word in [
			"recovery",
			"techie",
			"http",
			"https",
			"digital",
			"hack",
			"::",
			"//",
			"@",
			"com",
			"crypto",
			"bitcoin",
			"wallet",
			"hacker",
			"welcome",
			"whatsapp",
			"email",
			"cryptocurrency",
			"stolen",
			"freeze",
			"quick",
			"crucial",
			"tracing",
			"scammers",
			"expers",
			"hire",
			"century",
			"transaction",
			"essential",
			"managing",
			"contact",
			"contacting",
			"understanding",
			"assets",
			"funds",
		] {
			trie.insert(word);
		}
		trie
	});
}

impl Stats {
	pub fn process(&mut self, file: &str) {
		let Ok(text) = fs::read_to_string(&file) else {
			self.failed_file_count += 1;
			return;
		};
		self.file_count += 1;
		// self.muncher(&text);
		self.for_loops(&text);
	}
	#[allow(dead_code)]
	/// probably buggy. for example, are new lines sentences? what if the text has no last period?
	/// 500ms is without forbidden words check, but...
	/// 6000ms if adding forbidden words.. so not faster
	/// with trie this is 2600ms
	fn muncher(&mut self, text: &str) {
		let mut capitalized = true;
		let mut whitespaced = false;
		let mut dotted = false;
		let mut word = String::new();
		for char in text.chars() {
			if whitespaced {
				if !char.is_whitespace() {
					whitespaced = false; //end whiteness
				}
				continue;
			} else if char.is_whitespace() {
				whitespaced = true;
				self.word_count += 1; //end of word
				if capitalized {
					self.capitalized_count += 1;
				} else {
					//reset capitalized word
					capitalized = true;
				}
				let lowercase_word = word.to_lowercase();
				FORBIDON.with(|trie| {
					if trie.contains(&lowercase_word) {
						self.forbidden_count += 1;
					}
				});
				word = String::new();
				continue;
			}
			if dotted {
				if char != '.' {
					dotted = false; //end sentencing
				}
				continue;
			} else if char == '.' {
				dotted = true;
				self.sentence_count += 1;
				self.word_count += 1; //end of word
				if capitalized {
					self.capitalized_count += 1;
				} else {
					//reset capitalized word
					capitalized = true;
				}
				let lowercase_word = word.to_lowercase();
				FORBIDON.with(|trie| {
					if trie.contains(&lowercase_word) {
						self.forbidden_count += 1;
					}
				});
				word = String::new();
				continue;
			}
			word += &char.to_string();
			if char.is_numeric() {
				self.numeric_count += 1;
				capitalized = false;
			}
			if !char.is_ascii_uppercase() {
				capitalized = false;
			}
		}
	}
	#[allow(dead_code)]
	/// typically 5000ms
	/// with trie this is 1600ms
	fn for_loops(&mut self, text: &str) {
		for sentence in text
			.split('.')
			.map(|s| s.trim())
			.filter(|s| !s.is_empty())
		{
			self.sentence_count += 1;
			for word in sentence
				.split_whitespace()
				.map(|s| s.trim())
				.filter(|s| !s.is_empty())
			{
				self.word_count += 1;
				//get all numbers counted
				let mut all_capitalized = true;
				for char in word.chars() {
					if char.is_numeric() {
						self.numeric_count += 1;
						//TODO are numbers capitalized or not? I don't know!
					}
					if !char.is_ascii_uppercase() {
						all_capitalized = false;
					}
				}
				if all_capitalized {
					self.capitalized_count += 1;
				}
				let lowercase_word = word.to_lowercase();
				FORBIDON.with(|trie| {
					if trie.contains(&lowercase_word) {
						self.forbidden_count += 1;
					}
				});
			}
		}
	}
}

#[tokio::main]
async fn main() {
	let files = env::args().skip(1);

	let mut rx = {
		let (tx, rx) = mpsc::unbounded_channel();
		for file in files {
			let tx = tx.clone();
			tokio::spawn(async move {
				let mut stats = Stats::default();
				stats.process(&file);
				tx.send(stats).unwrap();
			});
		}
		rx
	};
	let mut stats = Stats::default();
	while let Some(file_stat) = rx.recv().await {
		stats += file_stat;
	}
	println!("{stats}");
}

#[test]
fn test() {
	use std::{env, fs, process::Command, time::Instant};
	println!("cwd: {}", env::current_dir().unwrap().display());

	//compile
	let mut compile = Command::new("cargo");
	let compile_arged = compile.arg("build").arg("--release");
	match compile_arged.output() {
		Ok(output) => println!("compiled {}", String::from_utf8_lossy(&output.stdout)),
		Err(err) => eprintln!("compile failed: {err}"),
	}

	//get test files
	let files = fs::read_dir("test_files")
		.unwrap()
		.map(|f| {
			f.unwrap()
				.path()
				.canonicalize()
				.unwrap()
				.to_str()
				.unwrap()
				.to_string()
		})
		.collect::<Vec<_>>();
	println!("test files found: {:#?}", files);

	//benchmark run
	let benchmark = Instant::now();
	let mut run = Command::new("target/release/jisspam");
	let run_arged = run.args(files);
	match run_arged.output() {
		Ok(output) => println!("{}", String::from_utf8_lossy(&output.stdout)),
		Err(err) => eprintln!("run failed: {err}"),
	}
	println!("benchmark: {}ms", benchmark.elapsed().as_millis());
}
#[test]
fn books_test() {
	use std::{env, fs, process::Command, time::Instant};
	println!("cwd: {}", env::current_dir().unwrap().display());

	//compile
	let mut compile = Command::new("cargo");
	let compile_arged = compile.arg("build").arg("--release");
	match compile_arged.output() {
		Ok(output) => println!("compiled {}", String::from_utf8_lossy(&output.stdout)),
		Err(err) => eprintln!("compile failed: {err}"),
	}

	//get test files
	let files = fs::read_dir("../books")
		.unwrap()
		.map(|f| {
			f.unwrap()
				.path()
				.canonicalize()
				.unwrap()
				.to_str()
				.unwrap()
				.to_string()
		})
		.collect::<Vec<_>>();
	println!("test files found: {:#?}", files);

	//benchmark run
	let benchmark = Instant::now();
	let mut run = Command::new("target/release/jisspam");
	let run_arged = run.args(files);
	match run_arged.output() {
		Ok(output) => println!("{}", String::from_utf8_lossy(&output.stdout)),
		Err(err) => eprintln!("run failed: {err}"),
	}
	println!("benchmark: {}ms", benchmark.elapsed().as_millis());
}