Skip to content

refactor: use struct and impl method for link check #241

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
reddevilmidzy merged 6 commits into from
Oct 5, 2025
Merged

refactor: use struct and impl method for link check #241

Show file tree
Hide file tree
Changes from 5 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
c74370c
refact: change module name
reddevilmidzy Oct 4, 2025
9af3bb8
refactor: use struct and impl method for link check
reddevilmidzy Oct 4, 2025
c04ec47
feat: add default impl for LinkChecker
reddevilmidzy Oct 4, 2025
d9ad8f1
refactor: change LinckChecker::new() method return type into Result<>
reddevilmidzy Oct 5, 2025
38c3433
fix: fix typo in error message
reddevilmidzy Oct 5, 2025
74f3455
refactor: use error propagation
reddevilmidzy Oct 5, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
122 changes: 73 additions & 49 deletions src/link_checker/link.rs → src/link_checker/checker.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,67 @@
use crate::{GitHubUrl, RepoManager};
use url::Url;

pub struct LinkChecker {
client: reqwest::Client,
}

impl LinkChecker {
pub fn new() -> Result<Self, String> {
let client = reqwest::Client::builder()
.timeout(std::time::Duration::from_secs(5))
.redirect(reqwest::redirect::Policy::none())
.build();

if let Ok(client) = client {
Ok(LinkChecker { client })
} else {
Err("failed to create Client".to_string())
}
}

pub async fn check_link(&self, url: &str) -> LinkCheckResult {
let mut attempts = 3;
while attempts > 0 {
match self.client.get(url).send().await {
Ok(res) => {
let status = res.status();
if status.is_success() {
return LinkCheckResult::Valid;
} else if status.is_redirection() {
if let Some(redirect_url) = res.headers().get("location")
&& let Ok(redirect_str) = redirect_url.to_str()
{
if is_trivial_redirect(url, redirect_str) {
return LinkCheckResult::Valid;
}
return LinkCheckResult::Redirect(redirect_str.to_string());
}
return LinkCheckResult::Valid;
} else if status.as_u16() == 404 && url.contains("github.com") {
return handle_github_404(url);
Comment on lines +36 to +37
Copy link
Contributor

@coderabbitai coderabbitai bot Oct 5, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

URL host check is fragile.

Using url.contains("github.com") can match unintended URLs like https://notgithub.com or paths containing the substring. Consider parsing the URL and checking the host properly.

-} else if status.as_u16() == 404 && url.contains("github.com") {
+} else if status.as_u16() == 404 && is_github_url(url) {
     return handle_github_404(url);

Add a helper function:

fn is_github_url(url: &str) -> bool {
    Url::parse(url)
        .ok()
        .and_then(|u| u.host_str())
        .map(|h| h == "github.com" || h.ends_with(".github.com"))
        .unwrap_or(false)
}
🤖 Prompt for AI Agents 
In src/link_checker/checker.rs around lines 40-41, the check using
url.contains("github.com") is fragile and can match unintended strings; replace
it by parsing the URL and checking the host properly (use
Url::parse(url).ok().and_then(|u| u.host_str()).map(|h| h == "github.com" ||
h.ends_with(".github.com")).unwrap_or(false))—add a small helper fn
is_github_url(url: &str) -> bool that implements this logic, use that helper in
place of the contains() call, and ensure the url crate is in scope and parsing
failures are handled by returning false.

} else {
return LinkCheckResult::Invalid(format!("HTTP status code: {status}"));
}
}
Err(e) => {
if attempts == 1 {
return LinkCheckResult::Invalid(format!("Request error: {e}"));
}
}
}
attempts -= 1;
tokio::time::sleep(tokio::time::Duration::from_secs(1)).await;
}
LinkCheckResult::Invalid("Max retries exceeded".to_string())
}
}

impl Default for LinkChecker {
fn default() -> Self {
Self::new().expect("failed to create LinkChecker")
}
}

#[derive(Debug, Eq, PartialEq)]
pub enum LinkCheckResult {
Valid,
Expand Down Expand Up @@ -32,48 +93,6 @@ fn handle_github_404(url: &str) -> LinkCheckResult {
}
}

pub async fn check_link(url: &str) -> LinkCheckResult {
let client = reqwest::Client::builder()
.timeout(std::time::Duration::from_secs(5))
.redirect(reqwest::redirect::Policy::none())
.build()
.unwrap();

let mut attempts = 3;
while attempts > 0 {
match client.get(url).send().await {
Ok(res) => {
let status = res.status();
if status.is_success() {
return LinkCheckResult::Valid;
} else if status.is_redirection() {
if let Some(redirect_url) = res.headers().get("location")
&& let Ok(redirect_str) = redirect_url.to_str()
{
if is_trivial_redirect(url, redirect_str) {
return LinkCheckResult::Valid;
}
return LinkCheckResult::Redirect(redirect_str.to_string());
}
return LinkCheckResult::Valid;
} else if status.as_u16() == 404 && url.contains("github.com") {
return handle_github_404(url);
} else {
return LinkCheckResult::Invalid(format!("HTTP status code: {status}"));
}
}
Err(e) => {
if attempts == 1 {
return LinkCheckResult::Invalid(format!("Request error: {e}"));
}
}
}
attempts -= 1;
tokio::time::sleep(tokio::time::Duration::from_secs(1)).await;
}
LinkCheckResult::Invalid("Max retries exceeded".to_string())
}

fn is_trivial_redirect(original: &str, redirect: &str) -> bool {
let orig_url = match Url::parse(original) {
Ok(url) => url,
Expand Down Expand Up @@ -109,29 +128,32 @@ mod tests {

#[tokio::test]
async fn validate_link() {
let link_checker = LinkChecker::default();
let link = "https://redddy.ai";
assert!(matches!(
check_link(link).await,
link_checker.check_link(link).await,
LinkCheckResult::Invalid(_)
));
let link = "https://lazypazy.tistory.com";
assert_eq!(check_link(link).await, LinkCheckResult::Valid);
assert_eq!(link_checker.check_link(link).await, LinkCheckResult::Valid);
}

#[tokio::test]
async fn change_organization_name() {
let link_checker = LinkChecker::default();
let link = "https://github.com/Bibimbap-Team/git-playground";
assert_eq!(
check_link(link).await,
link_checker.check_link(link).await,
LinkCheckResult::Redirect("https://github.com/Coduck-Team/git-playground".to_string())
);
}

#[tokio::test]
async fn change_branch_name() {
let link_checker = LinkChecker::default();
let link = "https://github.com/reddevilmidzy/kingsac/tree/forever";
assert_eq!(
check_link(link).await,
link_checker.check_link(link).await,
LinkCheckResult::Redirect(
"https://github.com/reddevilmidzy/kingsac/tree/lie".to_string()
)
Expand All @@ -140,24 +162,26 @@ mod tests {

#[tokio::test]
async fn change_repository_name() {
let link_checker = LinkChecker::default();
let link = "https://github.com/reddevilmidzy/test-queensac";
assert_eq!(
check_link(link).await,
link_checker.check_link(link).await,
LinkCheckResult::Redirect("https://github.com/reddevilmidzy/kingsac".to_string())
);
}

#[tokio::test]
async fn check_redirect_url() {
let link_checker = LinkChecker::default();
let link = "https://gluesql.org/docs";
assert_eq!(
check_link(link).await,
link_checker.check_link(link).await,
LinkCheckResult::Valid,
"check trivial redirect"
);
let link = "https://gluesql.org/docs/";
assert_eq!(
check_link(link).await,
link_checker.check_link(link).await,
LinkCheckResult::Valid,
"check trivial redirect"
);
Expand Down
4 changes: 2 additions & 2 deletions src/link_checker/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
mod link;
mod checker;
mod service;

pub use link::{LinkCheckResult, check_link};
pub use checker::{LinkCheckResult, LinkChecker};
pub use service::{InvalidLinkInfo, LinkCheckEvent, check_links};
5 changes: 3 additions & 2 deletions src/link_checker/service.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
use tracing::{error, info, instrument};

use crate::{LinkCheckResult, check_link, git};
use crate::{LinkCheckResult, LinkChecker, git};

#[derive(Debug)]
pub struct LinkCheckEvent {
Expand Down Expand Up @@ -100,11 +100,12 @@ pub async fn check_links(
}
};

let link_checker = LinkChecker::default();
let mut counters = LinkCheckCounters::new();
let mut invalid_links = Vec::new();

for link in links {
let result = check_link(&link.url).await;
let result = link_checker.check_link(&link.url).await;

counters.increment_total();

Expand Down