YGGverse/aquatic-crawler
1
0
Fork 0
mirror of https://github.com/YGGverse/aquatic-crawler.git synced 2026-03-31 09:05:33 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

use librqbit Id20 impl to validate info-hash bytes, require valid type for public methods

Browse source
This commit is contained in:
yggverse 2025-08-08 04:10:43 +03:00
parent d61af4b970
commit 704a2e5c29
4 changed files with 24 additions and 55 deletions
Download patch file Download diff file Expand all files Collapse all files

7
src/api.rs
View file

@ -1,11 +1,10 @@
mod info_hash;
use info_hash::InfoHash;
use librqbit::dht::Id20;
/// Parse infohash from the source filepath,
/// decode hash bytes to `InfoHash` array on success.
///
/// * return `None` if the `path` is not reachable
pub fn get(path: &str, capacity: usize) -> Option<Vec<InfoHash>> {
pub fn get(path: &str, capacity: usize) -> Option<Vec<Id20>> {
use std::io::Read;
if !path.ends_with(".bin") {
todo!("Only sources in the `.bin` format are supported!")
@ -21,7 +20,7 @@ pub fn get(path: &str, capacity: usize) -> Option<Vec<InfoHash>> {
if f.read(&mut b).ok()? != L {
break;
}
r.push(InfoHash::V1(b))
r.push(Id20::from_bytes(&b).ok()?)
}
Some(r)
}

15
src/api/info_hash.rs
View file

@ -1,15 +0,0 @@
pub enum InfoHash {
V1([u8; 20]),
}
impl std::fmt::Display for InfoHash {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
Self::V1(i) => write!(
f,
"{}",
i.iter().map(|b| format!("{b:02x}")).collect::<String>()
),
}
}
}

24
src/main.rs
View file

@ -89,18 +89,18 @@ async fn main() -> Result<()> {
}
} {
// convert to string once
let i = i.to_string();
let is = i.as_string();
if preload.contains_torrent(&i)? {
continue;
}
log::debug!("Index `{i}`...");
log::debug!("Index `{is}`...");
// run the crawler in single thread for performance reasons,
// use `timeout` argument option to skip the dead connections.
match time::timeout(
Duration::from_secs(config.add_torrent_timeout),
session.add_torrent(
AddTorrent::from_url(magnet(
&i,
&is,
if config.tracker.is_empty() {
None
} else {
@ -143,20 +143,20 @@ async fn main() -> Result<()> {
.is_some_and(|limit| only_files.len() + 1 > limit)
{
log::debug!(
"file count limit reached, skip `{id}` for `{i}`"
"file count limit reached, skip `{id}` for `{is}`"
);
break;
}
if preload.max_filesize.is_some_and(|limit| info.len > limit) {
log::debug!(
"file size limit reached, skip `{id}` for `{i}`"
"file size limit reached, skip `{id}` for `{is}`"
);
continue;
}
if preload.regex.as_ref().is_some_and(|r| {
!r.is_match(&info.relative_filename.to_string_lossy())
}) {
log::debug!("regex filter, skip `{id}` for `{i}`");
log::debug!("regex filter, skip `{id}` for `{is}`");
continue;
}
assert!(keep_files.insert(info.relative_filename.clone()));
@ -175,12 +175,12 @@ async fn main() -> Result<()> {
session
.delete(librqbit::api::TorrentIdOrHash::Id(id), false)
.await?;
log::debug!("torrent `{i}` indexed.")
log::debug!("torrent `{is}` indexed.")
}
Ok(_) => panic!(),
Err(e) => log::debug!("Failed to resolve `{i}`: `{e}`."),
Err(e) => log::debug!("Failed to resolve `{is}`: `{e}`."),
},
Err(e) => log::debug!("failed to resolve `{i}`: `{e}`"),
Err(e) => log::debug!("failed to resolve `{is}`: `{e}`"),
}
}
}
@ -200,11 +200,7 @@ async fn main() -> Result<()> {
/// Build magnet URI
fn magnet(info_hash: &str, trackers: Option<&Vec<Url>>) -> String {
let mut m = if info_hash.len() == 40 {
format!("magnet:?xt=urn:btih:{info_hash}")
} else {
todo!("infohash v2 is not supported by librqbit")
};
let mut m = format!("magnet:?xt=urn:btih:{info_hash}");
if let Some(t) = trackers {
for tracker in t {
m.push_str("&tr=");

33
src/preload.rs
View file

@ -1,4 +1,5 @@
use anyhow::{Result, bail};
use librqbit::dht::Id20;
use regex::Regex;
use std::{collections::HashSet, fs, path::PathBuf};
@ -36,14 +37,14 @@ impl Preload {
/// cleanup tmp data on success (see rqbit#408)
pub fn commit(
&self,
info_hash: &str,
info_hash: &Id20,
torrent_bytes: Vec<u8>,
persist_files: Option<HashSet<PathBuf>>,
) -> Result<()> {
validate_info_hash(info_hash)?;
let i = info_hash.as_string();
// persist preload files
let mut d = PathBuf::from(&self.root);
d.push(info_hash);
d.push(&i);
if d.exists() {
// clean previous data
fs::remove_dir_all(&d)?;
@ -87,7 +88,7 @@ impl Preload {
log::debug!("clean tmp data `{}`", tmp.to_string_lossy())
}
// persist torrent bytes to file (on previous operations success)
let t = self.torrent(info_hash);
let t = self.torrent(i);
fs::write(&t, torrent_bytes)?;
log::debug!("persist torrent bytes for `{}`", t.to_string_lossy());
Ok(())
@ -97,10 +98,9 @@ impl Preload {
/// Get absolute path to the temporary directory
/// * optionally creates directory if not exists
pub fn tmp(&self, info_hash: &str, is_create: bool) -> Result<PathBuf> {
validate_info_hash(info_hash)?;
pub fn tmp(&self, info_hash: &Id20, is_create: bool) -> Result<PathBuf> {
let mut p = PathBuf::from(&self.root);
p.push(tmp_component(info_hash));
p.push(tmp_component(info_hash.as_string()));
if p.is_file() {
bail!("Output directory `{}` is file", p.to_string_lossy())
}
@ -117,30 +117,19 @@ impl Preload {
}
/// Check the given hash is contain resolved torrent file
pub fn contains_torrent(&self, info_hash: &str) -> Result<bool> {
validate_info_hash(info_hash)?;
Ok(fs::exists(self.torrent(info_hash))?)
pub fn contains_torrent(&self, info_hash: &Id20) -> Result<bool> {
Ok(fs::exists(self.torrent(info_hash.as_string()))?)
}
/// Get absolute path to the torrent file
fn torrent(&self, info_hash: &str) -> PathBuf {
fn torrent(&self, info_hash: String) -> PathBuf {
let mut p = PathBuf::from(&self.root);
p.push(format!("{info_hash}.torrent"));
p
}
}
/// Non-expensive method to make sure the given string is safe to use in path builders
/// @TODO implement custom type?
fn validate_info_hash(value: &str) -> Result<()> {
if value.len() == 40 && value.chars().all(|c| c.is_ascii_hexdigit()) {
Ok(())
} else {
bail!("Invalid info-hash value `{value}`")
}
}
/// Build constant path component
fn tmp_component(info_hash: &str) -> String {
fn tmp_component(info_hash: String) -> String {
format!(".{info_hash}")
}