retoor/isspam
2
0
Fork 0
Code Issues 2 Pull Requests Actions Packages Projects Releases Wiki Activity

refactor: extract parser module with muncher and for_loops submodules from main.rs

Browse Source 
Move Stats processing methods into dedicated parser module with separate muncher and for_loops implementations, update .gitignore to exclude test_books directory, and simplify README to remove outdated benchmark instructions
This commit is contained in:
JestDotty
2025-10-04 13:18:03 +00:00
parent 305d2d5243
commit 93cf77cb96
6 changed files with 183 additions and 242 deletions
Download Patch File Download Diff File Expand all files Collapse all files
jest_rust/src/main.rs
+15 -157
View File
@@ -1,3 +1,4 @@
mod parser;
mod stats;
mod trie;
@@ -50,114 +51,6 @@ static FORBIDDEN_WORDS: LazyLock<Trie> = LazyLock::new(|| {
trie
});
impl Stats {
pub fn process(&mut self, text: &str) {
// 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();
if FORBIDDEN_WORDS.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();
if FORBIDDEN_WORDS.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();
if FORBIDDEN_WORDS.contains(&lowercase_word) {
self.forbidden_count += 1;
}
}
}
}
}
#[tokio::main]
async fn main() {
let files = env::args().skip(1);
@@ -166,17 +59,17 @@ async fn main() {
let (tx, rx) = mpsc::unbounded_channel();
for file in files {
//reading files not sequentially average shaves 30ms (of 1250ms), and that's on a NVMe SSD so why not
let Ok(text) = fs::read_to_string(&file) else {
if let Ok(text) = fs::read_to_string(&file) {
stats.file_count += 1;
let tx = tx.clone();
tokio::spawn(async move {
let mut stats = Stats::default();
parser::for_loops::parse(&mut stats, &text);
let _ = tx.send(stats);
});
} else {
stats.failed_file_count += 1;
continue;
};
stats.file_count += 1;
let tx = tx.clone();
tokio::spawn(async move {
let mut stats = Stats::default();
stats.process(&text);
tx.send(stats).unwrap();
});
}
}
rx
};
@@ -186,6 +79,7 @@ async fn main() {
println!("{stats}");
}
/// needs ../books.tar.gz to be extracted
#[test]
fn test() {
use std::{env, fs, process::Command, time::Instant};
@@ -199,44 +93,6 @@ fn test() {
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()
@@ -250,7 +106,9 @@ fn books_test() {
.to_string()
})
.collect::<Vec<_>>();
println!("test files found: {:#?}", files);
println!("test files found: {}", files.len());
println!();
//benchmark run
let benchmark = Instant::now();
jest_rust/src/parser/for_loops.rs
+39
View File
@@ -0,0 +1,39 @@
use crate::{FORBIDDEN_WORDS, stats::Stats};
#[allow(dead_code)]
/// typically 5000ms
/// with trie this is 1600ms
pub fn parse(stats: &mut Stats, text: &str) {
for sentence in text
.split('.')
.map(|s| s.trim())
.filter(|s| !s.is_empty())
{
stats.sentence_count += 1;
for word in sentence
.split_whitespace()
.map(|s| s.trim())
.filter(|s| !s.is_empty())
{
stats.word_count += 1;
//get all numbers counted
let mut all_capitalized = true;
for char in word.chars() {
if char.is_numeric() {
stats.numeric_count += 1;
//TODO are numbers capitalized or not? I don't know!
}
if !char.is_ascii_uppercase() {
all_capitalized = false;
}
}
if all_capitalized {
stats.capitalized_count += 1;
}
let lowercase_word = word.to_lowercase();
if FORBIDDEN_WORDS.contains(&lowercase_word) {
stats.forbidden_count += 1;
}
}
}
}
jest_rust/src/parser/mod.rs
+2
View File
@@ -0,0 +1,2 @@
pub mod muncher;
pub mod for_loops;
jest_rust/src/parser/muncher.rs
+66
View File
@@ -0,0 +1,66 @@
use crate::{FORBIDDEN_WORDS, stats::Stats};
#[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
pub fn parse(stats: &mut Stats, 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;
stats.word_count += 1; //end of word
if capitalized {
stats.capitalized_count += 1;
} else {
//reset capitalized word
capitalized = true;
}
let lowercase_word = word.to_lowercase();
if FORBIDDEN_WORDS.contains(&lowercase_word) {
stats.forbidden_count += 1;
}
word = String::new();
continue;
}
if dotted {
if char != '.' {
dotted = false; //end sentencing
}
continue;
} else if char == '.' {
dotted = true;
stats.sentence_count += 1;
stats.word_count += 1; //end of word
if capitalized {
stats.capitalized_count += 1;
} else {
//reset capitalized word
capitalized = true;
}
let lowercase_word = word.to_lowercase();
if FORBIDDEN_WORDS.contains(&lowercase_word) {
stats.forbidden_count += 1;
}
word = String::new();
continue;
}
word += &char.to_string();
if char.is_numeric() {
stats.numeric_count += 1;
capitalized = false;
}
if !char.is_ascii_uppercase() {
capitalized = false;
}
}
}