doctate/server/tests/transcribe_test.rs

use std::path::PathBuf;
use std::time::Duration;

use doctate_server::transcribe;
use doctate_server::transcribe::ffmpeg::remux_faststart;
use doctate_server::transcribe::recovery::scan_and_enqueue;
use doctate_server::transcribe::whisper::{transcribe, WhisperError};
use wiremock::matchers::{method, path, query_param};
use wiremock::{Mock, MockServer, ResponseTemplate};

fn fixture(name: &str) -> PathBuf {
    PathBuf::from(env!("CARGO_MANIFEST_DIR"))
        .join("tests/fixtures")
        .join(name)
}

#[tokio::test]
async fn ffmpeg_remux_produces_valid_output() {
    let input = fixture("sample.m4a");
    assert!(input.exists(), "fixture missing: {}", input.display());

    let output = remux_faststart(&input)
        .await
        .expect("remux failed");

    let meta = std::fs::metadata(output.path()).expect("output missing");
    assert!(meta.len() > 0, "output file is empty");

    // Sanity check: ffprobe should still recognize it as an m4a/mp4 container.
    let probe = std::process::Command::new("ffprobe")
        .arg("-hide_banner")
        .arg("-loglevel")
        .arg("error")
        .arg("-show_entries")
        .arg("format=format_name")
        .arg("-of")
        .arg("default=nw=1:nk=1")
        .arg(output.path())
        .output()
        .expect("ffprobe spawn failed");
    assert!(probe.status.success(), "ffprobe failed");
    let fmt = String::from_utf8_lossy(&probe.stdout);
    assert!(fmt.contains("mp4") || fmt.contains("m4a"), "unexpected format: {fmt}");
}

#[tokio::test]
async fn ffmpeg_remux_fails_on_missing_input() {
    let input = PathBuf::from("/nonexistent/does-not-exist.m4a");
    let result = remux_faststart(&input).await;
    assert!(result.is_err(), "expected error for missing input");
}

#[tokio::test]
async fn whisper_client_posts_multipart_and_returns_text() {
    let server = MockServer::start().await;
    let expected = "Hallo Welt, das ist ein Test.";

    Mock::given(method("POST"))
        .and(path("/asr"))
        .and(query_param("output", "txt"))
        .and(query_param("language", "de"))
        .respond_with(ResponseTemplate::new(200).set_body_string(expected))
        .expect(1)
        .mount(&server)
        .await;

    let client = reqwest::Client::new();
    let text = transcribe(&client, &server.uri(), &fixture("sample.m4a"), Duration::from_secs(10))
        .await
        .expect("transcribe failed");

    assert_eq!(text, expected);
}

#[tokio::test]
async fn whisper_client_returns_status_error_on_500() {
    let server = MockServer::start().await;

    Mock::given(method("POST"))
        .and(path("/asr"))
        .respond_with(ResponseTemplate::new(500).set_body_string("boom"))
        .mount(&server)
        .await;

    let client = reqwest::Client::new();
    let err = transcribe(&client, &server.uri(), &fixture("sample.m4a"), Duration::from_secs(10))
        .await
        .expect_err("expected error");

    match err {
        WhisperError::Status { status, body } => {
            assert_eq!(status, 500);
            assert_eq!(body, "boom");
        }
        other => panic!("unexpected error variant: {other:?}"),
    }
}

#[tokio::test]
async fn whisper_client_times_out() {
    let server = MockServer::start().await;

    Mock::given(method("POST"))
        .and(path("/asr"))
        .respond_with(ResponseTemplate::new(200).set_delay(Duration::from_secs(2)))
        .mount(&server)
        .await;

    let client = reqwest::Client::new();
    let err = transcribe(
        &client,
        &server.uri(),
        &fixture("sample.m4a"),
        Duration::from_millis(100),
    )
    .await
    .expect_err("expected timeout");

    assert!(matches!(err, WhisperError::Http(_)), "expected Http error, got {err:?}");
}

#[tokio::test]
async fn recovery_enqueues_only_pending_recordings() {
    let tmp = tempfile::tempdir().unwrap();
    let root = tmp.path();

    // User 1, open case with two .m4a — one pending, one already transcribed.
    let case_a = root.join("dr_a/open/aaaa");
    std::fs::create_dir_all(&case_a).unwrap();
    std::fs::write(case_a.join("2026-04-10T10-00-00Z.m4a"), b"x").unwrap();
    std::fs::write(case_a.join("2026-04-11T10-00-00Z.m4a"), b"x").unwrap();
    std::fs::write(case_a.join("2026-04-11T10-00-00Z.transcript.txt"), b"done").unwrap();

    // User 2, open case with one pending .m4a.
    let case_b = root.join("dr_b/open/bbbb");
    std::fs::create_dir_all(&case_b).unwrap();
    std::fs::write(case_b.join("2026-04-12T10-00-00Z.m4a"), b"x").unwrap();

    // done/ cases must be ignored (not re-transcribed).
    let case_c = root.join("dr_a/done/cccc");
    std::fs::create_dir_all(&case_c).unwrap();
    std::fs::write(case_c.join("2026-04-09T10-00-00Z.m4a"), b"x").unwrap();

    let (tx, mut rx) = transcribe::channel();
    scan_and_enqueue(root, &tx).await;
    drop(tx); // close channel so the loop below terminates

    let mut received = Vec::new();
    while let Some(job) = rx.recv().await {
        received.push(job.audio_path);
    }

    assert_eq!(received.len(), 2, "got: {received:?}");
    // Oldest first:
    assert!(received[0].ends_with("2026-04-10T10-00-00Z.m4a"));
    assert!(received[1].ends_with("2026-04-12T10-00-00Z.m4a"));
}