konservejo/tests/common/mod.rs

//! Common test utilities for integration tests.

#![expect(clippy::expect_used)]

use std::{collections::HashMap, path::Path};

use tempfile::{NamedTempFile, TempDir};
use wiremock::{
  Mock,
  MockServer,
  ResponseTemplate,
  matchers::{header, method, path_regex},
};

/// Configuration for a mock repository.
#[derive(Debug, Clone)]
pub struct MockRepo {
  pub owner:         String,
  pub name:          String,
  pub archive_bytes: Vec<u8>,
}

impl MockRepo {
  pub fn new(owner: &str, name: &str, archive_bytes: Vec<u8>) -> Self {
    Self {
      owner: owner.to_string(),
      name: name.to_string(),
      archive_bytes,
    }
  }
}

/// Sets up a mock Forgejo server with the given repositories.
pub async fn setup_mock_server(repos: Vec<MockRepo>) -> (MockServer, String) {
  let server = MockServer::start().await;

  // Group repos by organization
  let mut org_repos: HashMap<String, Vec<MockRepo>> = HashMap::new();
  for repo in repos {
    org_repos.entry(repo.owner.clone()).or_default().push(repo);
  }

  // Set up organization repo listing endpoints
  for (org, repos) in org_repos {
    let repos_clone = repos.clone();

    Mock::given(method("GET"))
      .and(path_regex(format!(r"/orgs/{}/repos", regex::escape(&org))))
      .and(header("authorization", "token test-token"))
      .and(header("accept", "application/json"))
      .respond_with(move |_req: &wiremock::Request| {
        let api_repos: Vec<serde_json::Value> = repos_clone
          .iter()
          .map(|r| {
            serde_json::json!({
              "name": r.name,
              "default_branch": "main",
              "owner": {"login": r.owner}
            })
          })
          .collect();

        ResponseTemplate::new(200).set_body_json(api_repos)
      })
      .mount(&server)
      .await;

    // Set up archive download endpoints for each repo
    for repo in repos {
      let archive_bytes = repo.archive_bytes.clone();
      Mock::given(method("GET"))
        .and(path_regex(format!(
          r"/repos/{}/{}/archive/.*",
          regex::escape(&repo.owner),
          regex::escape(&repo.name)
        )))
        .and(header("authorization", "token test-token"))
        .respond_with(move |_req: &wiremock::Request| {
          ResponseTemplate::new(200)
            .set_body_bytes(archive_bytes.clone())
            .insert_header("content-type", "application/gzip")
        })
        .mount(&server)
        .await;
    }
  }

  let url = server.uri();
  (server, url)
}

/// Test context holding all temporary resources.
pub struct TestContext {
  temp_dir:    TempDir,
  sink_dir:    TempDir,
  db_file:     NamedTempFile,
  config_file: NamedTempFile,
  mock_url:    String,
}

impl TestContext {
  /// Creates a new test context with the given mock server URL.
  pub fn new(mock_url: String) -> Self {
    let temp_dir = TempDir::new().expect("Failed to create temp dir");
    let sink_dir = TempDir::new().expect("Failed to create sink dir");
    let db_file = NamedTempFile::new().expect("Failed to create temp db file");
    let config_file =
      NamedTempFile::new().expect("Failed to create temp config file");

    Self {
      temp_dir,
      sink_dir,
      db_file,
      config_file,
      mock_url,
    }
  }

  /// Writes a TOML configuration file for the test.
  pub fn write_config(
    &self,
    orgs: &[String],
    excludes: &[String],
    concurrency: usize,
    max_retries: u32,
    backoff_ms: u64,
  ) {
    let orgs_list = orgs
      .iter()
      .map(|o| format!("\"{o}\""))
      .collect::<Vec<_>>()
      .join(", ");
    let excludes_str = excludes
      .iter()
      .map(|e| format!("\"{e}\""))
      .collect::<Vec<_>>()
      .join(", ");

    let config = format!(
      r#"[service]
name = "test"
temp_dir = "{}"
state_db_path = "{}"
concurrency_limit = {concurrency}

[service.retry]
max_retries = {max_retries}
initial_backoff_ms = {backoff_ms}

[[source]]
type = "forgejo"
id = "test-source"
api_url = "{}"
token = "test-token"

[source.scope]
organizations = [{orgs_list}]
exclude_repos = [{excludes_str}]

[[sink]]
type = "filesystem"
id = "test-sink"
path = "{}"
verify_on_write = true
"#,
      self.temp_dir.path().display(),
      self.db_file.path().display(),
      self.mock_url,
      self.sink_dir.path().display()
    );

    std::fs::write(self.config_file.path(), config)
      .expect("Failed to write config file");
  }

  /// Returns the path to the configuration file.
  pub fn config_path(&self) -> &Path {
    self.config_file.path()
  }

  /// Returns the path to the database file.
  pub fn db_path(&self) -> &Path {
    self.db_file.path()
  }
}

/// Creates a minimal valid tar.gz archive containing a single file.
pub fn create_test_archive(seed: &str) -> Vec<u8> {
  use std::io::Write;

  let mut tar_builder = tar::Builder::new(Vec::new());

  let content =
    format!("Test content for seed: {seed}\nThis is deterministic.");
  let mut header = tar::Header::new_gnu();
  header
    .set_path(format!("{seed}/test.txt"))
    .expect("Failed to set tar path");
  header.set_size(content.len() as u64);
  header.set_mode(0o644);
  header.set_cksum();

  tar_builder
    .append(&header, content.as_bytes())
    .expect("Failed to append to tar");

  let tar_data = tar_builder
    .into_inner()
    .expect("Failed to finish tar archive");

  let mut encoder =
    flate2::write::GzEncoder::new(Vec::new(), flate2::Compression::default());
  encoder
    .write_all(&tar_data)
    .expect("Failed to compress tar");
  encoder.finish().expect("Failed to finish gzip")
}

/// Computes the blake3 hash of the given data.
#[allow(dead_code)]
pub fn compute_hash(data: &[u8]) -> String {
  blake3::hash(data).to_hex().to_string()
}

/// Loads the configuration from the test context.
pub fn load_config(
  ctx: &TestContext,
) -> anyhow::Result<konservejo::config::Config> {
  konservejo::config::Config::from_file(ctx.config_path())
}