use std::fmt::Write;
use std::path::Path;
use sha2::{Digest, Sha256};
use tokio::io::{AsyncReadExt, AsyncWrite, AsyncWriteExt};
use xxhash_rust::xxh3::Xxh3;
use xxhash_rust::xxh64::Xxh64;
use crate::CoreError;
use crate::error::Result;
const BUF_SIZE: usize = 64 * 1024;
fn hash_mismatch(path: &Path, expected: u64, actual: u64) -> CoreError {
CoreError::HashMismatch {
path: path.to_path_buf(),
expected: format!("{expected:016x}"),
actual: format!("{actual:016x}"),
}
}
pub async fn hash_file_xxhash(path: &Path) -> Result<u64> {
let mut file = tokio::fs::File::open(path).await?;
let mut hasher = Xxh3::new();
let mut buf = vec![0u8; BUF_SIZE];
loop {
let n = file.read(&mut buf).await?;
if n == 0 {
break;
}
hasher.update(&buf[..n]);
}
Ok(hasher.digest())
}
pub async fn verify_xxhash(path: &Path, expected: u64) -> Result<()> {
let actual = hash_file_xxhash(path).await?;
if actual != expected {
return Err(hash_mismatch(path, expected, actual));
}
Ok(())
}
pub async fn verify_xxh64(path: &Path, expected: u64) -> Result<()> {
let actual = hash_file_xxh64(path).await?;
if actual != expected {
return Err(hash_mismatch(path, expected, actual));
}
Ok(())
}
pub async fn hash_file_xxh64(path: &Path) -> Result<u64> {
let mut file = tokio::fs::File::open(path).await?;
let mut hasher = Xxh64::new(0);
let mut buf = vec![0u8; BUF_SIZE];
loop {
let n = file.read(&mut buf).await?;
if n == 0 {
break;
}
hasher.update(&buf[..n]);
}
Ok(hasher.digest())
}
pub async fn verify_xxhash_compat(path: &Path, expected: u64) -> Result<u64> {
let mut file = tokio::fs::File::open(path).await?;
let mut xxh64_hasher = Xxh64::new(0);
let mut xxh3_hasher = Xxh3::new();
let mut buf = vec![0u8; BUF_SIZE];
loop {
let n = file.read(&mut buf).await?;
if n == 0 {
break;
}
xxh64_hasher.update(&buf[..n]);
xxh3_hasher.update(&buf[..n]);
}
let h64 = xxh64_hasher.digest();
if h64 == expected || xxh3_hasher.digest() == expected {
return Ok(expected);
}
Err(hash_mismatch(path, expected, h64))
}
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum CompatHashMatch {
Xxh64,
Xxh3,
}
pub async fn copy_and_hash_compat<R, W>(
reader: &mut R,
writer: &mut W,
expected: u64,
) -> Result<(u64, CompatHashMatch)>
where
R: tokio::io::AsyncRead + Unpin,
W: AsyncWrite + Unpin,
{
let mut xxh64_hasher = Xxh64::new(0);
let mut xxh3_hasher = Xxh3::new();
let mut buf = vec![0u8; BUF_SIZE];
loop {
let n = reader.read(&mut buf).await?;
if n == 0 {
break;
}
xxh64_hasher.update(&buf[..n]);
xxh3_hasher.update(&buf[..n]);
writer.write_all(&buf[..n]).await?;
}
writer.flush().await?;
let h64 = xxh64_hasher.digest();
if h64 == expected {
return Ok((h64, CompatHashMatch::Xxh64));
}
let h3 = xxh3_hasher.digest();
if h3 == expected {
return Ok((h3, CompatHashMatch::Xxh3));
}
Err(hash_mismatch(Path::new("<stream>"), expected, h64))
}
pub async fn hash_file_sha256(path: &Path) -> Result<String> {
let mut file = tokio::fs::File::open(path).await?;
let mut hasher = Sha256::new();
let mut buf = vec![0u8; BUF_SIZE];
loop {
let n = file.read(&mut buf).await?;
if n == 0 {
break;
}
hasher.update(&buf[..n]);
}
let mut hex = String::with_capacity(64);
for byte in hasher.finalize() {
write!(&mut hex, "{byte:02x}").expect("writing to String cannot fail");
}
Ok(hex)
}
pub async fn verify_sha256(path: &Path, expected: &str) -> Result<()> {
let actual = hash_file_sha256(path).await?;
if actual != expected {
return Err(CoreError::HashMismatch {
path: path.to_path_buf(),
expected: expected.to_string(),
actual,
});
}
Ok(())
}
#[cfg(test)]
mod tests {
use super::*;
use std::io::Write;
use tempfile::NamedTempFile;
use xxhash_rust::xxh3::xxh3_64;
use xxhash_rust::xxh64::xxh64;
fn create_temp_file(content: &[u8]) -> NamedTempFile {
let mut f = NamedTempFile::new().unwrap();
f.write_all(content).unwrap();
f
}
#[tokio::test]
async fn test_xxhash_roundtrip() {
let f = create_temp_file(b"hello world");
let hash = hash_file_xxhash(f.path()).await.unwrap();
verify_xxhash(f.path(), hash).await.unwrap();
}
#[tokio::test]
async fn test_xxhash_mismatch() {
let f = create_temp_file(b"hello world");
let result = verify_xxhash(f.path(), 0).await;
assert!(result.is_err());
}
#[tokio::test]
async fn test_sha256_known_value() {
let f = create_temp_file(b"hello world");
let hash = hash_file_sha256(f.path()).await.unwrap();
assert_eq!(
hash,
"b94d27b9934d3e08a52e52d7da7dabfac484efe37a5380ee9088f7ace2efcde9"
);
}
#[tokio::test]
async fn test_sha256_verify() {
let f = create_temp_file(b"test data");
let hash = hash_file_sha256(f.path()).await.unwrap();
verify_sha256(f.path(), &hash).await.unwrap();
}
#[tokio::test]
async fn test_sha256_mismatch() {
let f = create_temp_file(b"test data");
let result = verify_sha256(f.path(), "0000").await;
assert!(result.is_err());
}
#[tokio::test]
async fn test_xxhash_empty_file() {
let f = create_temp_file(b"");
let hash = hash_file_xxhash(f.path()).await.unwrap();
let expected = xxh3_64(b"");
assert_eq!(hash, expected);
}
#[tokio::test]
async fn test_xxhash_large_file() {
let data = vec![0xABu8; 1024 * 1024]; let f = create_temp_file(&data);
let hash = hash_file_xxhash(f.path()).await.unwrap();
let expected = xxh3_64(&data);
assert_eq!(hash, expected);
}
#[tokio::test]
async fn test_sha256_empty_file() {
let f = create_temp_file(b"");
let hash = hash_file_sha256(f.path()).await.unwrap();
assert_eq!(
hash,
"e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
);
}
#[tokio::test]
async fn test_xxhash_nonexistent_file() {
let result = hash_file_xxhash(Path::new("/tmp/nonexistent_file_xxhash_test")).await;
assert!(result.is_err());
}
#[tokio::test]
async fn test_sha256_nonexistent_file() {
let result = hash_file_sha256(Path::new("/tmp/nonexistent_file_sha256_test")).await;
assert!(result.is_err());
}
#[tokio::test]
async fn test_xxhash_different_content_different_hash() {
let f1 = create_temp_file(b"content alpha");
let f2 = create_temp_file(b"content beta");
let h1 = hash_file_xxhash(f1.path()).await.unwrap();
let h2 = hash_file_xxhash(f2.path()).await.unwrap();
assert_ne!(h1, h2);
}
#[tokio::test]
async fn test_xxhash_same_content_same_hash() {
let f1 = create_temp_file(b"identical content");
let f2 = create_temp_file(b"identical content");
let h1 = hash_file_xxhash(f1.path()).await.unwrap();
let h2 = hash_file_xxhash(f2.path()).await.unwrap();
assert_eq!(h1, h2);
}
#[tokio::test]
async fn copy_and_hash_matches_existing_helpers() {
let content = b"streamed hash content";
let expected = xxh64(content, 0);
let mut reader = tokio::io::BufReader::new(&content[..]);
let mut writer = Vec::new();
let (hash, matched) = copy_and_hash_compat(&mut reader, &mut writer, expected)
.await
.unwrap();
assert_eq!(writer, content);
assert_eq!(hash, expected);
assert_eq!(matched, CompatHashMatch::Xxh64);
}
#[tokio::test]
async fn copy_and_hash_detects_mismatch() {
let content = b"streamed hash content";
let mut reader = tokio::io::BufReader::new(&content[..]);
let mut writer = Vec::new();
let err = copy_and_hash_compat(&mut reader, &mut writer, 0)
.await
.unwrap_err();
assert!(matches!(err, CoreError::HashMismatch { .. }));
assert_eq!(writer, content);
}
}