Move to multi crates project
ci/woodpecker/push/tests Pipeline was successful

Starting impl devcontainer spec
This commit is contained in:
2026-07-31 20:06:37 +00:00
parent d711153553
commit 15f619ccf7
23 changed files with 271 additions and 30 deletions
+30
View File
@@ -0,0 +1,30 @@
[package]
name = "herald-server"
version = "1.2.0"
edition = "2024"
[dependencies]
reqwest = { workspace = true }
tokio = { workspace = true }
tokio-stream = { workspace = true }
tokio-util = { workspace = true }
futures-util = { workspace = true }
serde_json = { workspace = true }
serde = { workspace = true }
sentry = { workspace = true }
sentry-anyhow = { workspace = true }
openrouter-rs = { workspace = true }
dotenvy = { workspace = true }
tower = { workspace = true }
tower-http = { workspace = true }
tracing = { workspace = true }
tracing-subscriber = { workspace = true }
axum = { workspace = true }
anyhow = { workspace = true }
thiserror = { workspace = true }
ring = { workspace = true }
hex = { workspace = true }
bytes = { workspace = true }
metrics = { workspace = true }
metrics-exporter-prometheus = { workspace = true }
devcontainer-rs = { path = "../devcontainer-rs" }
+147
View File
@@ -0,0 +1,147 @@
use axum::body::{Bytes, to_bytes};
use axum::extract::{FromRef, FromRequest, State};
use axum::http::Request;
use axum::response::IntoResponse;
use axum::routing::{get, post};
use axum::{Json, Router};
use reqwest::StatusCode;
use ring::hmac;
use sentry::integrations::tower::{NewSentryLayer, SentryHttpLayer};
use serde_json::Value;
use tower::ServiceBuilder;
use tower_http::trace::TraceLayer;
use tracing::{info, instrument};
use tokio_util::sync::CancellationToken;
use crate::consts::{GITEA_EVENT_TYPE_HEADER_NAME, GITEA_SIG_HEADER_NAME, MAX_WEBHOOK_BODY_SIZE};
use crate::errors::AppError;
use crate::gitea::WebhookType;
use crate::metrics;
use crate::state::AppState;
pub async fn start(app_state: AppState, shutdown: CancellationToken) -> anyhow::Result<()> {
let http_port = app_state.config.http_port;
let app = Router::new()
.route("/", get(root))
.route("/webhook", post(webhook))
.layer(
ServiceBuilder::new()
.layer(NewSentryLayer::<Request<_>>::new_from_top())
.layer(SentryHttpLayer::new())
.layer(TraceLayer::new_for_http()),
)
.with_state(app_state);
let listener = tokio::net::TcpListener::bind(format!("0.0.0.0:{}", http_port)).await?;
info!("Listening API on port {}", http_port);
axum::serve(listener, app)
.with_graceful_shutdown(async move { shutdown.cancelled().await })
.await
.map_err(anyhow::Error::from)
.inspect(|_| info!("API shutting down complete"))?;
Ok(())
}
async fn root() -> &'static str {
"Hi, i'm Herald :)"
}
#[instrument(skip(app_state), fields(webhook_type), err)]
async fn webhook(
State(app_state): State<AppState>,
WebhookExtract(wb): WebhookExtract,
) -> Result<impl IntoResponse, AppError> {
tracing::Span::current().record("webhook_type", tracing::field::debug(&wb));
let event_type = wb.event_type_str();
metrics::webhook_received(event_type);
let event_id = wb.event_id();
if app_state.bot.check_and_mark(event_id).await {
metrics::webhook_duplicate(event_type);
return Err(AppError::AlreadyProcessedErr);
}
if app_state.bot_tx.try_send(wb).is_err() {
app_state.bot.unmark(event_id).await;
metrics::webhook_channel_full(event_type);
return Err(AppError::ChannelFullErr);
}
Ok((StatusCode::CREATED, "Task started"))
}
pub struct WebhookExtract(pub WebhookType);
impl<S> FromRequest<S> for WebhookExtract
where
AppState: FromRef<S>,
S: Send + Sync,
{
type Rejection = AppError;
#[instrument(skip(req, state), err)]
async fn from_request(req: axum::extract::Request, state: &S) -> Result<Self, Self::Rejection> {
let app_state = AppState::from_ref(state);
let headers = req.headers();
let sig_header = extract_header(GITEA_SIG_HEADER_NAME, headers)?;
let type_header = extract_header(GITEA_EVENT_TYPE_HEADER_NAME, headers)?;
let body_bytes = read_body(req.into_body()).await?;
verify_signature(
app_state.config.webhook_secret.as_bytes(),
&sig_header,
&body_bytes,
)?;
let body_str = String::from_utf8_lossy(&body_bytes).into_owned();
sentry::configure_scope(|scope| {
scope.add_event_processor(move |mut event| {
let mut request = event.request.take().unwrap_or_default();
request.data = Some(body_str.clone());
event.request = Some(request);
Some(event)
});
});
let webhook = parse_webhook(&type_header, &app_state.bot.name(), &body_bytes)?;
Ok(WebhookExtract(webhook))
}
}
fn extract_header(key: &str, headers: &axum::http::HeaderMap) -> Result<String, AppError> {
let value = headers
.get(key)
.ok_or(AppError::WebHookMissingHeaderErr(key.into()))?
.to_str()
.map_err(anyhow::Error::from)?;
Ok(value.to_owned())
}
async fn read_body(body: axum::body::Body) -> Result<Bytes, AppError> {
to_bytes(body, MAX_WEBHOOK_BODY_SIZE)
.await
.map_err(anyhow::Error::from)
.map_err(AppError::from)
}
fn parse_webhook(header: &str, bot_name: &str, body_bytes: &[u8]) -> Result<WebhookType, AppError> {
let Json(value) =
Json::<Value>::from_bytes(body_bytes).map_err(|_| AppError::MalformedJsonErr)?;
WebhookType::from_event(header, bot_name, value)
}
fn verify_signature(secret_key: &[u8], sig_header: &str, body: &[u8]) -> Result<(), AppError> {
let sig_header_decoded =
hex::decode(sig_header).map_err(|_| AppError::WebHookSigHeaderInvalidErr)?;
let key = hmac::Key::new(hmac::HMAC_SHA256, secret_key);
hmac::verify(&key, body, &sig_header_decoded).map_err(|_| AppError::WebHookSigHeaderInvalidErr)
}
+221
View File
@@ -0,0 +1,221 @@
use crate::{
gitea::{GiteaAPI, WebhookType}, metrics, open_router::OpenRouterClient,
};
use serde::Deserialize;
use std::{collections::HashSet, sync::Arc};
use tokio::sync::Mutex;
use tokio_util::sync::CancellationToken;
use tracing::{error, info, instrument};
#[derive(Deserialize, Debug)]
pub struct ReviewResult {
pub reviews: Vec<ReviewItem>,
pub comment: String,
pub cost: Option<f64>,
}
#[derive(Deserialize, Debug)]
pub struct ReviewItem {
pub filename: String,
pub line: Option<u64>,
pub code: String,
pub message: String,
}
#[derive(Clone)]
pub struct Bot {
bot_name: String,
gitea_api: GiteaAPI,
open_router_client: OpenRouterClient,
http_client: reqwest::Client,
max_concurrent: usize,
open_router_model: String,
actions_handled: Arc<Mutex<HashSet<u64>>>,
}
impl Bot {
pub fn new(
bot_name: String,
gitea_api: GiteaAPI,
open_router_client: OpenRouterClient,
http_client: reqwest::Client,
max_concurrent: usize,
open_router_model: String,
) -> Self {
Self {
bot_name,
gitea_api,
open_router_client,
http_client,
max_concurrent,
open_router_model,
actions_handled: Arc::new(Mutex::new(HashSet::new())),
}
}
pub fn name(&self) -> String {
self.bot_name.clone()
}
pub async fn start(
&self,
mut rx: tokio::sync::mpsc::Receiver<WebhookType>,
shutdown: CancellationToken,
) -> anyhow::Result<()> {
info!("Bot started");
let sem = Arc::new(tokio::sync::Semaphore::new(self.max_concurrent));
let mut tasks = tokio::task::JoinSet::new();
loop {
let wb = tokio::select! {
biased;
_ = shutdown.cancelled() => break,
msg = rx.recv() => match msg {
Some(wb) => wb,
None => break,
},
};
while let Some(res) = tasks.try_join_next() {
if let Err(e) = res {
error!("Task panicked: {e}");
}
}
info!(queued = rx.len(), active = tasks.len(), "Webhook received");
let permit = sem.clone().acquire_owned().await?;
let self_clone = self.clone();
metrics::increment_task_active();
tasks.spawn(async move {
self_clone.exec(wb).await;
drop(permit);
metrics::decrement_task_active();
});
}
tasks.join_all().await;
info!("Bot shutting down complete");
Ok(())
}
#[instrument(skip(self, webhook), fields(repo, pr))]
pub async fn exec(&self, webhook: WebhookType) {
let event_type_str = webhook.event_type_str();
match &webhook {
WebhookType::Review(p) => {
tracing::Span::current().record("repo", &p.repository.full_name);
tracing::Span::current().record("pr", p.pull_request.number);
}
};
let exec_result = match webhook {
WebhookType::Review(review_payload) => crate::bot_actions::review::exec_review(
&self.gitea_api,
&self.open_router_client,
&self.http_client,
&self.open_router_model,
review_payload,
),
}
.await;
match exec_result {
Ok(_) => {
metrics::task_completed(event_type_str);
info!("Task completed");
}
Err(err) => {
metrics::task_failed(event_type_str);
error!(%err, "Task error");
sentry_anyhow::capture_anyhow(&err);
}
}
}
pub async fn check_and_mark(&self, event_id: u64) -> bool {
let mut action_handled_lock = self.actions_handled.lock().await;
!action_handled_lock.insert(event_id)
}
pub async fn unmark(&self, event_id: u64) {
let mut action_handled_lock = self.actions_handled.lock().await;
action_handled_lock.remove(&event_id);
}
}
#[cfg(test)]
mod tests {
use super::*;
fn make_actions_handled() -> Arc<Mutex<HashSet<u64>>> {
Arc::new(Mutex::new(HashSet::new()))
}
async fn check_and_mark(actions_handled: &Arc<Mutex<HashSet<u64>>>, event_id: u64) -> bool {
let mut lock = actions_handled.lock().await;
!lock.insert(event_id)
}
async fn unmark(actions_handled: &Arc<Mutex<HashSet<u64>>>, event_id: u64) {
let mut lock = actions_handled.lock().await;
lock.remove(&event_id);
}
#[tokio::test]
async fn test_check_and_mark_first_call_returns_false() {
let actions_handled = make_actions_handled();
assert!(!check_and_mark(&actions_handled, 1).await);
}
#[tokio::test]
async fn test_check_and_mark_second_call_returns_true() {
let actions_handled = make_actions_handled();
check_and_mark(&actions_handled, 1).await;
assert!(check_and_mark(&actions_handled, 1).await);
}
#[tokio::test]
async fn test_check_and_mark_different_ids_are_independent() {
let actions_handled = make_actions_handled();
check_and_mark(&actions_handled, 1).await;
assert!(!check_and_mark(&actions_handled, 2).await);
}
#[tokio::test]
async fn test_unmark_allows_reprocessing() {
let actions_handled = make_actions_handled();
check_and_mark(&actions_handled, 1).await;
unmark(&actions_handled, 1).await;
assert!(!check_and_mark(&actions_handled, 1).await);
}
#[tokio::test]
async fn test_unmark_nonexistent_id_is_noop() {
let actions_handled = make_actions_handled();
unmark(&actions_handled, 99).await;
assert!(!check_and_mark(&actions_handled, 99).await);
}
#[tokio::test]
async fn test_check_and_mark_concurrent_same_id() {
let actions_handled = make_actions_handled();
let actions_handled2 = Arc::clone(&actions_handled);
let t1 = tokio::spawn(async move { check_and_mark(&actions_handled, 42).await });
let t2 = tokio::spawn(async move { check_and_mark(&actions_handled2, 42).await });
let (r1, r2) = tokio::join!(t1, t2);
let results = [r1.unwrap(), r2.unwrap()];
// exactement un seul des deux doit retourner false (non traité)
assert_eq!(results.iter().filter(|&&r| !r).count(), 1);
// l'autre doit retourner true (déjà traité)
assert_eq!(results.iter().filter(|&&r| r).count(), 1);
}
}
@@ -0,0 +1 @@
pub mod review;
@@ -0,0 +1,214 @@
use futures_util::stream::TryStreamExt;
use tokio::io::AsyncReadExt;
use tokio_util::io::StreamReader;
use tracing::instrument;
use crate::{
bot::ReviewResult, consts::{BOT_PROCESS_MSG, MAX_DIFF_SIZE, REVIEW_PROMPT}, gitea::{GiteaAPI, ReviewPayload}, metrics, open_router::OpenRouterClient,
};
#[instrument(skip(gitea_api, open_router_client, http_client, review_payload))]
pub async fn exec_review(
gitea_api: &GiteaAPI,
open_router_client: &OpenRouterClient,
http_client: &reqwest::Client,
model: &str,
review_payload: ReviewPayload,
) -> anyhow::Result<()> {
tracing::info!(
repo = %review_payload.repository.full_name,
pr = review_payload.pull_request.number,
action = %review_payload.action,
"Starting review"
);
let new_comment = gitea_api
.comment(
&BOT_PROCESS_MSG.replace("{model}", model),
&review_payload.repository.full_name,
review_payload.pull_request.number,
)
.await?;
let bot_result: Result<ReviewResult, anyhow::Error> = async {
let git_diff =
download_git_diff(http_client, &review_payload.pull_request.diff_url).await?;
let diff_for_llm = format_diff_for_review(&git_diff);
let bot_request = REVIEW_PROMPT
.replace("{subject}", &review_payload.pull_request.title)
.replace("{comment}", &review_payload.comment.body)
.replace("{diff}", &diff_for_llm);
let chat_result = open_router_client.chat(&bot_request).await?;
let mut review_result = serde_json::from_str::<ReviewResult>(&chat_result.message)?;
review_result.cost = chat_result.cost;
if let Some(cost) = review_result.cost {
metrics::openrouter_cost_usd(cost);
}
let final_review_markdown = review_result_to_markdown(&review_result);
gitea_api
.post_pull_request_review(
&review_result,
&final_review_markdown,
&review_payload.repository.full_name,
review_payload.pull_request.number,
)
.await?;
Ok(review_result)
}
.await;
match bot_result {
Ok(_) => {
gitea_api
.delete_comment(&review_payload.repository.full_name, new_comment.id)
.await
}
Err(e) => {
gitea_api
.edit_comment(
&format!("Error while reviewing: {}", e),
&review_payload.repository.full_name,
new_comment.id,
)
.await
}
}
}
fn review_result_to_markdown(review_result: &ReviewResult) -> String {
if review_result.reviews.is_empty() {
return String::from("No issues found. ✅");
}
let mut md = String::from("## Review Feedback\n\n");
md.push_str(&format!(
"### {} issues found.\n\n",
review_result.reviews.len()
));
if !review_result.comment.is_empty() {
md.push_str("\n---\n\n");
md.push_str("### Summary\n\n");
md.push_str(&review_result.comment);
md.push('\n');
}
if let Some(cost) = review_result.cost {
md.push_str("\n---\n\n");
md.push_str(&format!("### Cost: ${}", cost));
md.push('\n');
}
md
}
async fn download_git_diff(http_client: &reqwest::Client, url: &str) -> anyhow::Result<String> {
let response = http_client.get(url).send().await?;
let stream = response.bytes_stream().map_err(std::io::Error::other);
let mut buf = Vec::with_capacity(MAX_DIFF_SIZE);
StreamReader::new(stream)
.take((MAX_DIFF_SIZE + 1) as u64)
.read_to_end(&mut buf)
.await?;
if buf.len() > MAX_DIFF_SIZE {
anyhow::bail!("Git diff exceeds the maximum allowed size of 1 Mo");
}
Ok(String::from_utf8_lossy(&buf).into_owned())
}
fn format_diff_for_review(git_diff: &str) -> String {
let mut output = String::new();
let mut current_file: Option<&str> = None;
let mut new_line: u64 = 0;
for line in git_diff.lines() {
if let Some(rest) = line.strip_prefix("diff --git a/") {
if let Some(end) = rest.find(' ') {
current_file = Some(&rest[..end]);
}
new_line = 0;
continue;
}
if line.starts_with("---") || line.starts_with("+++") {
continue;
}
if line.starts_with("@@") && line.contains('+') {
if let Some(start) = parse_hunk_new_start(line) {
new_line = start;
}
continue;
}
let Some(filename) = current_file else {
continue;
};
if line.starts_with(' ') {
new_line += 1;
continue;
}
if let Some(code) = line.strip_prefix('+') {
use std::fmt::Write;
let _ = writeln!(output, "{filename}:{new_line}:{code}");
new_line += 1;
}
}
output
}
fn parse_hunk_new_start(hunk_header: &str) -> Option<u64> {
let plus_part = hunk_header.split('+').nth(1)?;
let num_str = plus_part.split(|c: char| !c.is_ascii_digit()).next()?;
num_str.parse::<u64>().ok()
}
#[cfg(test)]
#[test]
fn test_format_diff_for_review() {
let diff = concat!(
"diff --git a/src/foo.rs b/src/foo.rs\n",
"--- a/src/foo.rs\n",
"+++ b/src/foo.rs\n",
"@@ -1,3 +1,6 @@\n",
" fn main() {\n",
"+ let x = 1;\n",
" println!(\"hello\");\n",
"+ let y = 2;\n",
"+ let z = 3;\n",
" }\n",
"diff --git a/src/bar.rs b/src/bar.rs\n",
"--- a/src/bar.rs\n",
"+++ b/src/bar.rs\n",
"@@ -10,4 +10,6 @@\n",
" old context\n",
"+ let a = 10;\n",
" more context\n",
"+ let b = 20;\n",
);
let result = format_diff_for_review(diff);
let expected = concat!(
"src/foo.rs:2: let x = 1;\n",
"src/foo.rs:4: let y = 2;\n",
"src/foo.rs:5: let z = 3;\n",
"src/bar.rs:11: let a = 10;\n",
"src/bar.rs:13: let b = 20;\n",
);
assert_eq!(result, expected);
}
+44
View File
@@ -0,0 +1,44 @@
pub const GITEA_SIG_HEADER_NAME: &str = "x-gitea-signature";
pub const GITEA_EVENT_TYPE_HEADER_NAME: &str = "x-gitea-event-type";
pub const MAX_WEBHOOK_BODY_SIZE: usize = 1024 * 1024; // 1 MiB
pub const MAX_DIFF_SIZE: usize = 1024 * 1024; // 1 MiB
pub const BOT_PROCESS_MSG: &str = "
Review in progress with the model \"{model}\"...
";
pub const REVIEW_PROMPT: &str = "
You are a senior software engineer reviewing code changes.
Check good practices and code quality.
This is the pull request subject: \"{subject}\"
This is the user comment: \"{comment}\"
The code changes (only added lines, with line numbers):
{diff}
Please review the code changes and provide feedback.
IMPORTANT: the `line` field must be the line number shown before each line.
The provided code has the format: `filename:line:code`
Return your feedback, in french, with only this json format, reviews must contain each review
All fields are mandatory.
(filename field must contain the full path with extension) and comment must contain a final summary:
{
\"reviews\": [
{
\"filename\": \"\",
\"line\": ,
\"code\": \"\",
\"message\": \"\"
}
],
\"comment\": \"\"
}
";
+82
View File
@@ -0,0 +1,82 @@
use anyhow::anyhow;
#[derive(Clone)]
pub struct EnvConfig {
pub http_port: u16,
pub webhook_secret: String,
pub open_router_api_key: String,
pub open_router_model: String,
pub open_router_timeout: u64,
pub bot_max_concurrent: usize,
pub gitea_url: String,
pub gitea_token: String,
pub gitea_timeout: u64,
pub metrics_bind_addr: Option<String>,
}
pub fn load_config() -> anyhow::Result<EnvConfig> {
let http_port = try_get_env("HTTP_PORT")?.parse()?;
let webhook_secret = try_get_env("WEBHOOK_SIG_HEADER_SECRET")?;
let open_router_api_key = try_get_env("OPEN_ROUTER_API_KEY")?;
let open_router_model = try_get_env("OPEN_ROUTER_MODEL")?;
let open_router_timeout = try_get_env("OPEN_ROUTER_TIMEOUT")?.parse()?;
let bot_max_concurrent = try_get_env("BOT_MAX_CONCURRENT")?.parse()?;
let gitea_url = try_get_env("GITEA_URL")?;
let gitea_token = try_get_env("GITEA_TOKEN")?;
let gitea_timeout = try_get_env("GITEA_TIMEOUT")?.parse()?;
let metrics_bind_addr = std::env::var("METRICS_BIND_ADDR").ok();
Ok(EnvConfig {
http_port,
webhook_secret,
open_router_api_key,
open_router_model,
open_router_timeout,
bot_max_concurrent,
gitea_url,
gitea_token,
gitea_timeout,
metrics_bind_addr,
})
}
pub fn try_get_env(key: &str) -> anyhow::Result<String> {
let env_value = std::env::var(key).map_err(|e| anyhow::anyhow!("{}: {}", key, e))?;
if env_value.trim().is_empty() {
return Err(anyhow!(format!("env var {} is empty", key)));
}
Ok(env_value)
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_try_get_env_returns_value() {
unsafe { std::env::set_var("TEST_ENV_PRESENT", "hello") };
assert_eq!(try_get_env("TEST_ENV_PRESENT").unwrap(), "hello");
}
#[test]
fn test_try_get_env_missing_var_returns_error() {
unsafe { std::env::remove_var("TEST_ENV_MISSING") };
assert!(try_get_env("TEST_ENV_MISSING").is_err());
}
#[test]
fn test_try_get_env_empty_value_returns_error() {
unsafe { std::env::set_var("TEST_ENV_EMPTY", "") };
let err = try_get_env("TEST_ENV_EMPTY").unwrap_err();
assert!(err.to_string().contains("TEST_ENV_EMPTY"));
}
#[test]
fn test_try_get_env_whitespace_only_returns_error() {
unsafe { std::env::set_var("TEST_ENV_WHITESPACE", " ") };
let err = try_get_env("TEST_ENV_WHITESPACE").unwrap_err();
assert!(err.to_string().contains("TEST_ENV_WHITESPACE"));
}
}
+82
View File
@@ -0,0 +1,82 @@
use anyhow::anyhow;
use axum::response::IntoResponse;
use reqwest::StatusCode;
#[derive(thiserror::Error, Debug)]
pub enum AppError {
#[error("Unauthorized user id")]
UnauthorizedUserErr,
#[error("Unknow gitea event")]
UnknownEventErr,
#[error("Malformed Json")]
MalformedJsonErr,
#[error("WebHook header not found")]
WebHookMissingHeaderErr(String),
#[error("WebHook sig header is invalid")]
WebHookSigHeaderInvalidErr,
#[error("WebHook have bad action")]
InvalidActionErr,
#[error("Channel full")]
ChannelFullErr,
#[error("Already processed")]
AlreadyProcessedErr,
#[error(transparent)]
BadJsonStructErr(#[from] serde_json::Error),
#[error(transparent)]
Other(#[from] anyhow::Error),
}
impl IntoResponse for AppError {
fn into_response(self) -> axum::response::Response {
match self {
AppError::InvalidActionErr => (
StatusCode::UNPROCESSABLE_ENTITY,
"WebHook have bad action".to_string(),
),
AppError::UnknownEventErr => {
(StatusCode::BAD_REQUEST, "Unknow gitea event".to_string())
}
AppError::UnauthorizedUserErr => (
StatusCode::UNAUTHORIZED,
"Unauthorized user name".to_string(),
),
AppError::MalformedJsonErr => (StatusCode::BAD_REQUEST, "Malformed Json".to_string()),
AppError::BadJsonStructErr(err) => (
StatusCode::BAD_REQUEST,
format!("Json not contains mandatory fields: {}", err),
),
AppError::WebHookMissingHeaderErr(h) => {
(StatusCode::BAD_REQUEST, format!("header {} is missing", h))
}
AppError::WebHookSigHeaderInvalidErr => (
StatusCode::UNAUTHORIZED,
"WebHook sig header is invalid".to_string(),
),
AppError::AlreadyProcessedErr => (StatusCode::OK, "Already processed".to_string()),
AppError::ChannelFullErr => {
sentry_anyhow::capture_anyhow(&anyhow!("Max concurrent tasks reached"));
(
StatusCode::SERVICE_UNAVAILABLE,
"Max concurrent tasks reached".to_string(),
)
}
AppError::Other(err) => {
sentry_anyhow::capture_anyhow(&err);
(
StatusCode::INTERNAL_SERVER_ERROR,
"Internal server error".to_string(),
)
}
}
.into_response()
}
}
+444
View File
@@ -0,0 +1,444 @@
use std::time::Duration;
use serde::Deserialize;
use serde_json::{Value, json};
use tracing::instrument;
use crate::{bot::ReviewResult, errors::AppError};
#[derive(Clone)]
pub struct GiteaAPI {
base_url: String,
client: reqwest::Client,
}
impl GiteaAPI {
pub fn new(base_url: &str, token: &str, timeout: u64) -> anyhow::Result<Self> {
let mut default_headers = reqwest::header::HeaderMap::new();
default_headers.insert(
reqwest::header::HeaderName::from_static("authorization"),
reqwest::header::HeaderValue::from_str(&format!("Bearer {}", token))?,
);
Ok(Self {
base_url: String::from(base_url),
client: reqwest::Client::builder()
.timeout(Duration::from_secs(timeout))
.default_headers(default_headers)
.build()?,
})
}
#[instrument(skip(self))]
pub async fn get_authorized_user(&self) -> anyhow::Result<User> {
let url = format!("{}/api/v1/user", self.base_url);
let res = self.client.get(url).send().await?;
if !res.status().is_success() {
return Err(anyhow::anyhow!(
"Failed to get authorized user: {}",
res.status()
));
}
res.json::<User>().await.map_err(anyhow::Error::from)
}
#[instrument(skip(self))]
pub async fn comment(
&self,
body: &str,
full_name: &str,
index: u64,
) -> anyhow::Result<Comment> {
let url = format!(
"{}/api/v1/repos/{}/issues/{}/comments",
self.base_url, full_name, index
);
let res = self
.client
.post(url)
.json(&json!({
"body": body
}))
.send()
.await?;
if !res.status().is_success() {
return Err(anyhow::anyhow!("Failed to comment: {}", res.status()));
}
res.json::<Comment>().await.map_err(anyhow::Error::from)
}
#[instrument(skip(self))]
pub async fn edit_comment(
&self,
body: &str,
full_name: &str,
comment_id: u64,
) -> anyhow::Result<()> {
let url = format!(
"{}/api/v1/repos/{}/issues/comments/{}",
self.base_url, full_name, comment_id
);
let res = self
.client
.patch(url)
.json(&json!({
"body": body
}))
.send()
.await?;
if !res.status().is_success() {
return Err(anyhow::anyhow!("Failed to comment: {}", res.status()));
}
Ok(())
}
#[instrument(skip(self))]
pub async fn delete_comment(&self, full_name: &str, comment_id: u64) -> anyhow::Result<()> {
let url = format!(
"{}/api/v1/repos/{}/issues/comments/{}",
self.base_url, full_name, comment_id
);
let res = self.client.delete(url).send().await?;
if !res.status().is_success() {
return Err(anyhow::anyhow!(
"Failed to delete comment: {}",
res.status()
));
}
Ok(())
}
#[instrument(skip(self, review_result))]
pub async fn post_pull_request_review(
&self,
review_result: &ReviewResult,
final_comment: &str,
full_name: &str,
index: u64,
) -> anyhow::Result<()> {
let url = format!(
"{}/api/v1/repos/{}/pulls/{}/reviews",
self.base_url, full_name, index
);
let comments = &review_result
.reviews
.iter()
.filter(|r| r.line.is_some())
.map(|r| {
let path = r.filename.clone();
let line = r.line.unwrap_or(0);
let body = r.message.clone();
json!({
"path": path,
"new_position": line,
"body": body
})
})
.collect::<Vec<_>>();
let res = self
.client
.post(url)
.json(&json!({
"event": "COMMENT",
"body": final_comment,
"comments": comments
}))
.send()
.await?;
if !res.status().is_success() {
return Err(anyhow::anyhow!("Failed to post review: {}", res.status()));
}
Ok(())
}
}
#[derive(Debug)]
pub enum WebhookType {
Review(ReviewPayload),
}
impl WebhookType {
pub fn event_id(&self) -> u64 {
match self {
WebhookType::Review(payload) => payload.comment.id,
}
}
pub fn event_type_str(&self) -> &'static str {
match self {
WebhookType::Review(_) => "review",
}
}
}
#[derive(Deserialize, Debug)]
pub struct ReviewPayload {
pub action: String,
pub pull_request: PullRequest,
pub repository: Repository,
pub comment: Comment,
}
#[derive(Deserialize, Debug)]
pub struct PullRequest {
pub id: u64,
pub diff_url: String,
pub number: u64,
pub title: String,
}
#[derive(Deserialize, Debug)]
pub struct Comment {
pub id: u64,
pub body: String,
pub user: User,
}
#[derive(Deserialize, Debug)]
pub struct User {
pub id: u64,
pub login: String,
}
#[derive(Deserialize, Debug)]
pub struct Repository {
pub full_name: String,
}
impl WebhookType {
pub fn from_event(event: &str, bot_name: &str, json: Value) -> Result<Self, AppError> {
let wb = match event {
"pull_request_comment" => Ok(WebhookType::Review(serde_json::from_value(json)?)),
_ => Err(AppError::UnknownEventErr),
}?;
let action = match &wb {
WebhookType::Review(review_payload) => &review_payload.action,
};
if action != "created" {
return Err(AppError::InvalidActionErr);
}
let pr_body = match &wb {
WebhookType::Review(review_payload) => &review_payload.comment.body,
};
if !pr_body.starts_with(&format!("@{}", bot_name)) {
return Err(AppError::UnauthorizedUserErr);
}
Ok(wb)
}
}
#[cfg(test)]
mod tests {
use super::*;
use serde_json::json;
#[test]
fn test_from_event_valid_pull_request_comment() {
let json = json!({
"action": "created",
"pull_request": {
"id": 42,
"diff_url": "https://mydiff.fr",
"number": 1,
"title": "My PR"
},
"repository": {
"full_name": "owner/repo"
},
"comment": {
"id": 7,
"body": "@test_bot LGTM",
"user": {
"id": 100,
"login": "test_user"
}
}
});
let result = WebhookType::from_event("pull_request_comment", "test_bot", json);
assert!(result.is_ok());
match result.unwrap() {
WebhookType::Review(payload) => {
assert_eq!(payload.action, "created");
assert_eq!(payload.pull_request.id, 42);
assert_eq!(payload.comment.id, 7);
assert_eq!(payload.comment.body, "@test_bot LGTM");
assert_eq!(payload.comment.user.id, 100);
}
}
}
#[test]
fn test_from_event_unknown_event() {
let json = json!({});
let result = WebhookType::from_event("push", "test_bot", json);
assert!(result.is_err());
match result.unwrap_err() {
AppError::UnknownEventErr => {}
_ => panic!("expected UnknownEventErr"),
}
}
#[test]
fn test_from_event_malformed_json() {
let json = json!({
"action": "created"
// pull_request and comment are missing
});
let result = WebhookType::from_event("pull_request_comment", "test_bot", json);
assert!(result.is_err());
match result.unwrap_err() {
AppError::BadJsonStructErr(_) => {}
_ => panic!("expected BadJsonStructErr"),
}
}
#[test]
fn test_from_event_rejects_non_created_action() {
let json = json!({
"action": "edited",
"pull_request": {
"id": 1,
"diff_url": "https://mydiff.fr",
"number": 1,
"title": "My PR"
},
"repository": {
"full_name": "owner/repo"
},
"comment": {
"id": 1,
"body": "@test_bot body",
"user": {
"id": 1,
"login": "test_user"
}
}
});
let result = WebhookType::from_event("pull_request_comment", "test_bot", json);
assert!(result.is_err());
match result.unwrap_err() {
AppError::InvalidActionErr => {}
_ => panic!("expected InvalidActionErr"),
}
}
#[test]
fn test_deserialize_review_payload() {
let json = json!({
"action": "created",
"pull_request": {
"id": 99,
"diff_url": "https://mydiff.fr",
"number": 5,
"title": "My PR"
},
"repository": {
"full_name": "owner/repo"
},
"comment": {
"id": 12,
"body": "Needs work",
"user": {
"id": 200,
"login": "test_user"
}
}
});
let payload: ReviewPayload = serde_json::from_value(json).unwrap();
assert_eq!(payload.action, "created");
assert_eq!(payload.pull_request.id, 99);
assert_eq!(payload.comment.id, 12);
assert_eq!(payload.comment.body, "Needs work");
assert_eq!(payload.comment.user.id, 200);
}
#[test]
fn test_from_event_empty_json() {
let result = WebhookType::from_event("pull_request_comment", "test_bot", json!({}));
assert!(result.is_err());
assert!(matches!(result.unwrap_err(), AppError::BadJsonStructErr(_)));
}
#[test]
fn test_from_event_rejects_wrong_bot_name() {
let json = json!({
"action": "created",
"pull_request": {
"id": 1,
"diff_url": "https://mydiff.fr",
"number": 1,
"title": "My PR"
},
"repository": {
"full_name": "owner/repo"
},
"comment": {
"id": 1,
"body": "@other_bot do something",
"user": {
"id": 1,
"login": "test_user"
}
}
});
let result = WebhookType::from_event("pull_request_comment", "test_bot", json);
assert!(matches!(result.unwrap_err(), AppError::UnauthorizedUserErr));
}
#[test]
fn test_from_event_rejects_no_bot_prefix() {
let json = json!({
"action": "created",
"pull_request": {
"id": 1,
"diff_url": "https://mydiff.fr",
"number": 1,
"title": "My PR"
},
"repository": {
"full_name": "owner/repo"
},
"comment": {
"id": 1,
"body": "just a comment without bot mention",
"user": {
"id": 1,
"login": "test_user"
}
}
});
let result = WebhookType::from_event("pull_request_comment", "test_bot", json);
assert!(matches!(result.unwrap_err(), AppError::UnauthorizedUserErr));
}
}
+114
View File
@@ -0,0 +1,114 @@
use crate::{bot::Bot, gitea::{GiteaAPI, WebhookType}, open_router::OpenRouterClient, state::AppState};
use dotenvy::dotenv;
use tokio::signal::unix::{SignalKind, signal};
use tokio_util::sync::CancellationToken;
use tracing::{info, warn};
use tracing_subscriber::{EnvFilter, fmt, layer::SubscriberExt, util::SubscriberInitExt};
mod api;
mod bot;
mod bot_actions;
mod consts;
mod env;
mod errors;
mod gitea;
mod metrics;
mod open_router;
mod state;
fn main() -> anyhow::Result<()> {
dotenv().ok();
tracing_subscriber::registry()
.with(fmt::layer())
.with(
EnvFilter::try_from_default_env() // lit RUST_LOG depuis l'env
.unwrap_or_else(|_| EnvFilter::new("info")),
)
.init();
let _sentry_guard = if let Ok(sentry_dsn) = env::try_get_env("SENTRY_DSN") {
info!("Initialize sentry");
Some(sentry::init((
sentry_dsn,
sentry::ClientOptions {
release: sentry::release_name!(),
send_default_pii: true,
..Default::default()
},
)))
} else {
warn!("SENTRY_DSN not set, sentry will not be initialized");
None
};
tokio::runtime::Runtime::new()?.block_on(run())
}
async fn run() -> anyhow::Result<()> {
let config = env::load_config()?;
if let Some(metric_bind_addr) = &config.metrics_bind_addr {
metrics::install(metric_bind_addr)?;
}
let gitea_api = GiteaAPI::new(&config.gitea_url, &config.gitea_token, config.gitea_timeout)?;
let gitea_user = gitea_api.get_authorized_user().await?;
info!(
port = config.http_port,
model = %config.open_router_model,
gitea_url = %config.gitea_url,
bot_name = %gitea_user.login,
"Starting Herald"
);
let open_router_client = OpenRouterClient::new(
&config.open_router_api_key,
&config.open_router_model,
config.open_router_timeout,
)?;
let shutdown = CancellationToken::new();
let bot = Bot::new(
gitea_user.login,
gitea_api,
open_router_client,
reqwest::Client::new(),
config.bot_max_concurrent,
config.open_router_model.clone(),
);
let (tx, rx) = tokio::sync::mpsc::channel::<WebhookType>(config.bot_max_concurrent * 2);
let app_state = AppState {
bot_tx: tx,
bot: bot.clone(),
config,
};
let signal = async {
let mut sigterm = signal(SignalKind::terminate())?;
let mut sigint = signal(SignalKind::interrupt())?;
tokio::select! {
_ = sigterm.recv() => info!("Received SIGTERM"),
_ = sigint.recv() => info!("Received SIGINT"),
}
info!("Shutting down...");
shutdown.cancel();
anyhow::Ok(())
};
tokio::try_join!(
bot.start(rx, shutdown.clone()),
api::start(app_state, shutdown.clone()),
signal
)?;
info!("Shutdown complete");
Ok(())
}
+62
View File
@@ -0,0 +1,62 @@
use std::{net::SocketAddr, str::FromStr};
use metrics::{Unit, describe_counter, describe_gauge, counter, gauge};
pub fn webhook_received(event_type: &str) {
counter!("herald_webhooks_received_total", "event_type" => event_type.to_string())
.increment(1);
}
pub fn webhook_duplicate(event_type: &str) {
counter!("herald_webhooks_duplicate_total", "event_type" => event_type.to_string())
.increment(1);
}
pub fn webhook_channel_full(event_type: &str) {
counter!("herald_webhooks_channel_full_total", "event_type" => event_type.to_string())
.increment(1);
}
pub fn increment_task_active() {
gauge!("herald_bot_tasks_active").increment(1.0);
}
pub fn decrement_task_active() {
gauge!("herald_bot_tasks_active").decrement(1.0);
}
pub fn task_completed(event_type: &str) {
counter!("herald_bot_tasks_completed_total", "event_type" => event_type.to_string())
.increment(1);
}
pub fn task_failed(event_type: &str) {
counter!("herald_bot_tasks_failed_total", "event_type" => event_type.to_string())
.increment(1);
}
pub fn openrouter_cost_usd(cost: f64) {
counter!("herald_openrouter_cost_cents_total").increment((cost * 100.0).round() as u64);
}
pub fn describe() {
describe_counter!("herald_webhooks_received_total", Unit::Count, "Total webhooks received");
describe_counter!("herald_webhooks_duplicate_total", Unit::Count, "Webhooks rejected as duplicates");
describe_counter!("herald_webhooks_channel_full_total", Unit::Count, "Webhooks dropped because the bot channel was full");
describe_gauge!("herald_bot_tasks_active", Unit::Count, "Bot tasks currently in progress");
describe_counter!("herald_bot_tasks_completed_total", Unit::Count, "Bot tasks completed successfully");
describe_counter!("herald_bot_tasks_failed_total", Unit::Count, "Bot tasks that failed");
describe_counter!("herald_openrouter_cost_cents_total", Unit::Count, "Total OpenRouter cost in cents (divide by 100 for USD)");
}
pub fn install(bind_addr: &str) -> anyhow::Result<()> {
describe();
let builder = metrics_exporter_prometheus::PrometheusBuilder::new();
builder
.with_http_listener(SocketAddr::from_str(bind_addr)?)
.install()?;
tracing::info!(bind_addr, "Prometheus metrics exporter installed");
Ok(())
}
+50
View File
@@ -0,0 +1,50 @@
use std::time::Duration;
use openrouter_rs::{Message, api::chat::ChatCompletionRequest};
use tracing::instrument;
pub struct ChatResult {
pub message: String,
pub cost: Option<f64>,
}
#[derive(Clone)]
pub struct OpenRouterClient {
client: openrouter_rs::OpenRouterClient,
model: String,
}
impl OpenRouterClient {
pub fn new(token: &str, model: &str, timeout: u64) -> anyhow::Result<Self> {
Ok(Self {
client: openrouter_rs::OpenRouterClient::builder()
.api_key(token)
.http_client(
reqwest::Client::builder()
.timeout(Duration::from_secs(timeout))
.build()?,
)
.build()?,
model: String::from(model),
})
}
#[instrument(skip(self), err)]
pub async fn chat(&self, msg: &str) -> anyhow::Result<ChatResult> {
let request = ChatCompletionRequest::builder()
.model(&self.model)
.enable_reasoning()
.messages(vec![Message::new(openrouter_rs::types::Role::User, msg)])
.build()?;
let response = self.client.chat().create(&request).await?;
Ok(ChatResult {
message: response.choices[0]
.content()
.map(String::from)
.ok_or(anyhow::anyhow!("No content"))?,
cost: response.usage.and_then(|u| u.cost),
})
}
}
+8
View File
@@ -0,0 +1,8 @@
use crate::{bot::Bot, env::EnvConfig, gitea::WebhookType};
#[derive(Clone)]
pub struct AppState {
pub bot_tx: tokio::sync::mpsc::Sender<WebhookType>,
pub bot: Bot,
pub config: EnvConfig,
}