refactor(subs): remove podnapisi
ci / web (push) Successful in 30s
ci / rust (push) Failing after 1m58s
e2e / e2e (push) Failing after 2m15s

This commit is contained in:
Miguel Palhas
2026-08-26 22:58:01 +01:00
parent 3f4c1ade3b
commit 4cec2b6ed5
26 changed files with 44 additions and 1405 deletions
Generated
-67
View File
@@ -2,12 +2,6 @@
# It is not intended for manual editing.
version = 4
[[package]]
name = "adler2"
version = "2.0.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "320119579fcad9c21884f5c4861d16174d0e06250625266f50fe6898340abefa"
[[package]]
name = "aho-corasick"
version = "1.1.5"
@@ -228,7 +222,6 @@ dependencies = [
"tokio",
"tracing",
"wiremock",
"zip",
]
[[package]]
@@ -466,15 +459,6 @@ version = "2.5.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "217698eaf96b4a3f0bc4f3662aaa55bdf913cd54d7204591faa790070c6d0853"
[[package]]
name = "crc32fast"
version = "1.5.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8498c871161e1742aaa9d52551b2d6ebdd4c3d45a3be423e3728f33b955be550"
dependencies = [
"cfg-if",
]
[[package]]
name = "crossbeam-queue"
version = "0.3.13"
@@ -640,16 +624,6 @@ version = "0.1.11"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d45db016d36b838f563236e9193d0ee6ce38f3f68b6c94e914b4929c96bbb890"
[[package]]
name = "flate2"
version = "1.1.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "843fba2746e448b37e26a819579957415c8cef339bf08564fe8b7ddbd959573c"
dependencies = [
"miniz_oxide",
"zlib-rs",
]
[[package]]
name = "flume"
version = "0.12.0"
@@ -1299,16 +1273,6 @@ dependencies = [
"unicase",
]
[[package]]
name = "miniz_oxide"
version = "0.8.9"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "1fa76a2c86f704bdb222d66965fb3d63269ce38518b83cb0575fca855ebb6316"
dependencies = [
"adler2",
"simd-adler32",
]
[[package]]
name = "mio"
version = "1.2.2"
@@ -1827,12 +1791,6 @@ dependencies = [
"libc",
]
[[package]]
name = "simd-adler32"
version = "0.3.10"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "3a219298ac11a56ea9a6d2120044824d6f01aeb034955e7af7bc16858527deea"
[[package]]
name = "slab"
version = "0.4.12"
@@ -2392,12 +2350,6 @@ version = "0.2.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "e421abadd41a4225275504ea4d6566923418b7f05506fbc9c0fe86ba7396114b"
[[package]]
name = "typed-path"
version = "0.12.3"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8e28f89b80c87b8fb0cf04ab448d5dd0dd0ade2f8891bae878de66a75a28600e"
[[package]]
name = "typenum"
version = "1.20.1"
@@ -2907,25 +2859,6 @@ dependencies = [
"syn 3.0.3",
]
[[package]]
name = "zip"
version = "8.6.0"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "2d04a6b5381502aa6087c94c669499eb1602eb9c5e8198e534de571f7154809b"
dependencies = [
"crc32fast",
"flate2",
"indexmap",
"memchr",
"typed-path",
]
[[package]]
name = "zlib-rs"
version = "0.6.7"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "34b31d188d9d685a4f9c7b46d6e36631b07058d2cfe190267adce54dc230bf12"
[[package]]
name = "zmij"
version = "1.0.23"
-1
View File
@@ -57,7 +57,6 @@ tracing = "0.1"
tracing-subscriber = { version = "0.3", features = ["env-filter"] }
url = "2"
uuid = { version = "1", features = ["v4", "serde"] }
zip = { version = "8", default-features = false, features = ["deflate-flate2-zlib-rs"] }
# Test-only
tempfile = "3"
+1 -1
View File
@@ -931,7 +931,7 @@ they satisfy viewing but can never feed a translator, and arr does not OCR
them. `arr-probe` already reports subtitle tracks with resolved languages; the
format is the new fact it must carry.
**Providers.** OpenSubtitles.com and Podnapisi, behind one trait.
**Providers.** OpenSubtitles.com, behind one trait.
**Ranking.** A `moviehash` match wins outright. Then an exact release-name
match, then same release group or same source, then uploader rating and
+1 -5
View File
@@ -275,8 +275,6 @@ async fn probe_subtitles(state: &AppState) -> SubtitleHealth {
async fn probe_provider(state: &AppState, id: &str) -> ProviderCheck {
let check = match state.subtitle_provider(id) {
Some(provider) => check_from(provider.probe().await),
// Only reachable for a provider that needs credentials: Podnapisi is
// anonymous (#215) and is always attached, so it never lands here.
None => Check::unconfigured(format!(
"{id} is enabled but has no credentials — they are bootstrap config, not a setting"
)),
@@ -481,15 +479,13 @@ mod tests {
#[tokio::test]
async fn an_enabled_provider_without_credentials_is_unconfigured() {
let (_dir, state) = application().await;
// The seed row enables opensubtitles and podnapisi; none are attached.
// The seed row enables OpenSubtitles; none is attached.
let body = report(state).await;
assert_eq!(
body["subtitles"]["providers"],
serde_json::json!([
{ "id": "opensubtitles", "status": "unconfigured",
"detail": "opensubtitles is enabled but has no credentials — they are bootstrap config, not a setting" },
{ "id": "podnapisi", "status": "unconfigured",
"detail": "podnapisi is enabled but has no credentials — they are bootstrap config, not a setting" },
])
);
assert_eq!(body["status"], "degraded");
+2 -2
View File
@@ -400,7 +400,7 @@ mod tests {
fn valid_input() -> serde_json::Value {
serde_json::json!({
"wanted_languages": ["pt-PT", "en"],
"providers_enabled": ["opensubtitles", "podnapisi"],
"providers_enabled": ["opensubtitles"],
"translation_engine": null,
"provider_daily_budgets": { "opensubtitles": 100 },
"translator_daily_budgets": {},
@@ -423,7 +423,7 @@ mod tests {
);
assert_eq!(
settings["providers_enabled"],
serde_json::json!(["opensubtitles", "podnapisi"])
serde_json::json!(["opensubtitles"])
);
assert!(settings["translation_engine"].is_null());
// `available_engines` is a fact about the build, not about the seed
+1 -1
View File
@@ -2609,7 +2609,7 @@ mod tests {
async fn a_grab_naming_an_unconfigured_provider_is_a_422() {
let fixture = stub_application().await;
let mut body = pt();
body["provider"] = serde_json::json!("podnapisi");
body["provider"] = serde_json::json!("unconfigured");
let (status, _) = grab(&fixture, body).await;
assert_eq!(status, StatusCode::UNPROCESSABLE_ENTITY);
}
+2 -3
View File
@@ -330,8 +330,7 @@ mod tests {
.await
.unwrap();
db.migrate().await.unwrap();
// The seed row already enables opensubtitles and podnapisi; none are
// attached to this action.
// The seed row enables OpenSubtitles, which is not attached to this action.
let prowlarr = MockServer::start().await;
Mock::given(method("GET"))
.and(path("/ping"))
@@ -354,7 +353,7 @@ mod tests {
let first = action.tick(&db).await;
let second = action.tick(&db).await;
assert_eq!(first.len(), 2, "one lamp per missing provider");
assert_eq!(first.len(), 1, "one lamp per missing provider");
assert_eq!(second.len(), 0, "does not repeat while still broken");
// Attaching nothing but disabling them silences the lamps.
+2 -48
View File
@@ -36,8 +36,6 @@ pub const ENV_NTFY_OPERATOR_TOPIC: &str = "ARR_NTFY_OPERATOR_TOPIC";
pub const ENV_OPENSUBTITLES_API_KEY: &str = "ARR_OPENSUBTITLES_API_KEY";
pub const ENV_OPENSUBTITLES_USERNAME: &str = "ARR_OPENSUBTITLES_USERNAME";
pub const ENV_OPENSUBTITLES_PASSWORD: &str = "ARR_OPENSUBTITLES_PASSWORD";
// Podnapisi takes no credentials (#215): its search and download are
// unauthenticated, so there is nothing here for an operator to set.
// The OpenAI-compatible backend's base URL and model are `subtitle_settings`
// rows, not bootstrap keys (#220): that backend is anything speaking the
// shape, and which endpoint is in use is something the operator changes from
@@ -51,11 +49,6 @@ pub const ENV_TRANSLATE_GOOGLE_BASE_URL: &str = "ARR_TRANSLATE_GOOGLE_BASE_URL";
pub const ENV_TRANSLATE_COMMAND_TEMPLATE: &str = "ARR_TRANSLATE_COMMAND_TEMPLATE";
pub const ENV_ALASS_PATH: &str = "ARR_ALASS_PATH";
pub const ENV_FFMPEG_PATH: &str = "ARR_FFMPEG_PATH";
/// E2E seam only, env-only, same shape as [`ENV_TMDB_URL`]: `None` means
/// Podnapisi's real address. Kept out of the config file for the same
/// reason `tmdb_url` is — DESIGN.md §10 has no business exposing a seam that
/// only a test harness uses.
pub const ENV_PODNAPISI_URL: &str = "ARR_PODNAPISI_URL";
pub const DEFAULT_BIND_ADDR: &str = "0.0.0.0:7878";
pub const DEFAULT_DATABASE_PATH: &str = "arr.db";
@@ -166,7 +159,6 @@ pub struct EnvOverrides {
pub tmdb_url: Option<String>,
pub jellyfin_url: Option<String>,
pub jellyfin_api_key: Option<String>,
pub podnapisi_url: Option<String>,
pub ntfy_url: Option<String>,
pub ntfy_operator_topic: Option<String>,
pub opensubtitles_api_key: Option<String>,
@@ -199,7 +191,6 @@ impl EnvOverrides {
tmdb_url: std::env::var(ENV_TMDB_URL).ok(),
jellyfin_url: std::env::var(ENV_JELLYFIN_URL).ok(),
jellyfin_api_key: std::env::var(ENV_JELLYFIN_API_KEY).ok(),
podnapisi_url: std::env::var(ENV_PODNAPISI_URL).ok(),
ntfy_url: std::env::var(ENV_NTFY_URL).ok(),
ntfy_operator_topic: std::env::var(ENV_NTFY_OPERATOR_TOPIC).ok(),
opensubtitles_api_key: std::env::var(ENV_OPENSUBTITLES_API_KEY).ok(),
@@ -238,9 +229,6 @@ pub struct Config {
pub tmdb_url: Option<String>,
pub jellyfin_url: String,
pub jellyfin_api_key: Option<String>,
/// E2E seam only, env-only. `None` means Podnapisi's built-in default
/// address; see [`ENV_PODNAPISI_URL`].
pub podnapisi_url: Option<String>,
pub ntfy_url: String,
/// The operator's ntfy topic (DESIGN.md §9.5) for *needs a decision* and
/// *broken*. `None` means those two notifications are skipped — there is
@@ -401,7 +389,6 @@ impl Config {
.or(file.jellyfin_url)
.unwrap_or_else(|| DEFAULT_JELLYFIN_URL.to_string()),
jellyfin_api_key: env.jellyfin_api_key,
podnapisi_url: env.podnapisi_url,
ntfy_url: env
.ntfy_url
.or(file.ntfy_url)
@@ -457,7 +444,6 @@ mod tests {
DEFAULT_SEED_IDLE_LIMIT_MINUTES
);
assert_eq!(config.jellyfin_url, DEFAULT_JELLYFIN_URL);
assert_eq!(config.podnapisi_url, None);
assert!(config.tracker_seeding.is_empty());
assert_eq!(config.ntfy_url, DEFAULT_NTFY_URL);
assert_eq!(config.ntfy_operator_topic, None);
@@ -536,9 +522,8 @@ prowlarr_url = "http://prowlarr.internal:9696"
assert_eq!(config.jellyfin_api_key.as_deref(), Some("secret-3"));
}
/// §15: `OpenSubtitles` and every translator's credentials are secrets,
/// so this mirrors [`secrets_come_only_from_env`] for them. Podnapisi has
/// no entry: it takes no credentials at all (#215).
/// §15: OpenSubtitles and every translator's credentials are secrets, so
/// this mirrors [`secrets_come_only_from_env`] for them.
#[test]
fn subtitle_secrets_come_only_from_env() {
let env = EnvOverrides {
@@ -656,37 +641,6 @@ alass_path = "/usr/local/bin/alass"
));
}
/// Same env-only seam as `tmdb_url`, for the same reason: `arr-e2e` needs
/// to point Podnapisi at a `wiremock` fake without a live-tracker risk
/// creeping into the config file (DESIGN.md §15, §10).
#[test]
fn podnapisi_url_is_an_env_only_seam() {
let config = Config::resolve(EnvOverrides::default()).unwrap();
assert_eq!(config.podnapisi_url, None);
let env = EnvOverrides {
podnapisi_url: Some("http://127.0.0.1:9/subtitles".into()),
..EnvOverrides::default()
};
let config = Config::resolve(env).unwrap();
assert_eq!(
config.podnapisi_url.as_deref(),
Some("http://127.0.0.1:9/subtitles")
);
let dir = tempfile::tempdir().unwrap();
let path = dir.path().join("arr.toml");
std::fs::write(&path, "podnapisi_url = \"http://127.0.0.1:9/subtitles\"\n").unwrap();
let env = EnvOverrides {
config_file: Some(path.to_string_lossy().into_owned()),
..EnvOverrides::default()
};
assert!(matches!(
Config::resolve(env),
Err(ConfigError::TomlDecode(_))
));
}
#[test]
fn a_secret_in_the_config_file_is_a_parse_error() {
let dir = tempfile::tempdir().unwrap();
+1 -16
View File
@@ -135,7 +135,6 @@ fn api_state(
config.opensubtitles_api_key.clone(),
config.opensubtitles_username.clone(),
config.opensubtitles_password.clone(),
config.podnapisi_url.clone(),
))
.with_translation_backends(translators.backends.clone())
.with_jellyfin(jellyfin)
@@ -460,7 +459,6 @@ fn register_broken(
config.opensubtitles_api_key.clone(),
config.opensubtitles_username.clone(),
config.opensubtitles_password.clone(),
config.podnapisi_url.clone(),
),
backends: translators.backends.clone(),
alass_path: config.alass_path.clone(),
@@ -607,7 +605,6 @@ fn subtitle_action(
config.opensubtitles_api_key.clone(),
config.opensubtitles_username.clone(),
config.opensubtitles_password.clone(),
config.podnapisi_url.clone(),
),
translators.backends.clone(),
arr_subs::Syncer::new().with_binary(config.alass_path.clone()),
@@ -626,26 +623,14 @@ fn subtitle_action(
/// which providers *exist* is decided here, once, at startup; which of them a
/// search *runs* is the `providers_enabled` setting the API reads per
/// request. OpenSubtitles.com needs a registered API key to be called at all,
/// so without one it is not offered; Podnapisi is anonymous and always is.
/// so without one it is not offered.
fn subtitle_providers(
opensubtitles_api_key: Option<String>,
username: Option<String>,
password: Option<String>,
podnapisi_url: Option<String>,
) -> Vec<std::sync::Arc<dyn arr_subs::Provider>> {
let mut providers: Vec<std::sync::Arc<dyn arr_subs::Provider>> = Vec::new();
// `podnapisi_url` is the e2e-only seam (`ARR_PODNAPISI_URL`); absent, this
// is the real Podnapisi.net.
let mut builder = arr_subs::Podnapisi::builder();
if let Some(url) = podnapisi_url {
builder = builder.base_url(url);
}
match builder.build() {
Ok(podnapisi) => providers.push(std::sync::Arc::new(podnapisi)),
Err(error) => tracing::warn!(%error, "Podnapisi not available"),
}
let Some(api_key) = opensubtitles_api_key else {
tracing::info!("no OpenSubtitles.com API key configured; that provider is off");
return providers;
+11 -5
View File
@@ -1688,18 +1688,24 @@ mod tests {
async fn a_cheaper_candidate_is_downloaded_when_the_ranked_winner_is_over_budget() {
let fixture = Fixture::new(&no_tracks()).await;
fixture.configure(r#"["pt-PT"]"#, None).await;
sqlx::query(
"UPDATE subtitle_settings SET providers_enabled = '[\"opensubtitles\",\"fallback\"]'",
)
.execute(fixture.database.pool())
.await
.unwrap();
fixture.configure_provider_budget("opensubtitles", 0).await;
let winner = Candidate {
hash_match: true,
..candidate("opensubtitles", "hash", Language::PortuguesePortugal)
};
let opensubtitles = StubProvider::new("opensubtitles", vec![winner]);
let podnapisi = StubProvider::new(
"podnapisi",
vec![candidate("podnapisi", "7", Language::PortuguesePortugal)],
let fallback = StubProvider::new(
"fallback",
vec![candidate("fallback", "7", Language::PortuguesePortugal)],
);
let action = action(
vec![Arc::new(opensubtitles), Arc::new(podnapisi)],
vec![Arc::new(opensubtitles), Arc::new(fallback)],
Vec::new(),
);
@@ -1707,7 +1713,7 @@ mod tests {
let files = fixture.files().await;
assert_eq!(files.len(), 1);
assert_eq!(files[0].provider.as_deref(), Some("podnapisi"));
assert_eq!(files[0].provider.as_deref(), Some("fallback"));
let attempt = fixture.attempt("pt-PT").await.unwrap();
assert_eq!(attempt.state, SubtitleState::Satisfied);
}
@@ -14,7 +14,7 @@ CREATE TABLE subtitle_settings (
-- Provider ids in search order (DESIGN.md §15's ranking runs per
-- provider before it runs per candidate). Absent from the array means
-- disabled.
providers_enabled TEXT NOT NULL DEFAULT '["opensubtitles","podnapisi"]'
providers_enabled TEXT NOT NULL DEFAULT '["opensubtitles"]'
CHECK (json_valid(providers_enabled)),
-- The compiled-in backend in use, or NULL when translation is off. Not
-- checked against the running binary's features here — the API does
@@ -0,0 +1,12 @@
-- Podnapisi.net no longer has a reachable public service. Existing settings
-- must not keep selecting it or preserve a budget for a provider arr no
-- longer ships.
UPDATE subtitle_settings
SET providers_enabled = COALESCE(
(SELECT json_group_array(value)
FROM json_each(providers_enabled)
WHERE value != 'podnapisi'),
'[]'
),
provider_daily_budgets = json_remove(provider_daily_budgets, '$.podnapisi'),
updated_at = strftime('%Y-%m-%dT%H:%M:%fZ', 'now');
+1 -1
View File
@@ -213,7 +213,7 @@ mod tests {
);
// Same kind, other name: untouched.
assert_eq!(
spent_today(db.pool(), BudgetKind::Provider, "podnapisi")
spent_today(db.pool(), BudgetKind::Provider, "another-provider")
.await
.unwrap(),
0
+1 -1
View File
@@ -970,7 +970,7 @@ mod tests {
let file = media_file(db.pool(), "/m/d.mkv", "2026-01-01T00:00:00.000Z").await;
let id = record_file(
db.pool(),
&NewSubtitleFile::fetched(file, "en", "podnapisi", "7", "/m/d.en.srt"),
&NewSubtitleFile::fetched(file, "en", "opensubtitles", "7", "/m/d.en.srt"),
)
.await
.unwrap();
Binary file not shown.
Binary file not shown.
+2 -11
View File
@@ -38,15 +38,6 @@ pub mod fixtures {
pub const PROWLARR_MOVIE_SEARCH: &str = include_str!("../fixtures/prowlarr-movie-search.xml");
pub const TMDB_CONFIGURATION: &str = include_str!("../fixtures/tmdb-configuration.json");
pub const TMDB_MOVIE_DUNE: &str = include_str!("../fixtures/tmdb-movie-dune.json");
/// A zipped SRT with one cue at 00:00:01 — plausibly close to a fetched
/// subtitle's real timing for [`crate::probe_movie_fixture`]'s
/// three-second clip.
pub const PODNAPISI_PLAUSIBLE_ZIP: &[u8] =
include_bytes!("../fixtures/podnapisi-plausible.zip");
/// A zipped SRT timed five minutes in — implausible against the same
/// three-second clip, so `alass` must reject it (DESIGN.md §15).
pub const PODNAPISI_FARFETCHED_ZIP: &[u8] =
include_bytes!("../fixtures/podnapisi-farfetched.zip");
}
/// The clip the subtitle scenarios probe and sync against: three seconds,
@@ -211,8 +202,8 @@ impl Daemon {
}
/// Same as [`Self::spawn`], plus extra environment variables for the
/// child — the subtitle seams (`ARR_PODNAPISI_URL`,
/// `ARR_TRANSLATE_COMMAND_TEMPLATE`, ...) that only a test harness sets.
/// child — configuration such as `ARR_TRANSLATE_COMMAND_TEMPLATE` that
/// only a test harness sets.
///
/// # Panics
///
-430
View File
@@ -1,430 +0,0 @@
//! The subtitle path end to end (DESIGN.md §15, issue #205), cross-process
//! like the rest of this crate: the real `arr` binary, real `ffmpeg` and real
//! `alass`. Only the HTTP boundary — Podnapisi — and the remote-command
//! translation backend are stubbed, the same rule §12 already applies to
//! Prowlarr and TMDB: a live tracker or a live translation API never runs in
//! a test.
//!
//! There is no HTTP route that adopts an already-imported file (that
//! pipeline is its own seam — grab, download, import), so every scenario
//! here seeds a `media_files` row directly against the daemon's own SQLite
//! file, the same fixture shape the daemon's in-process subtitle tests use.
// Same per-target quirk as `e2e.rs`: dev-dependencies used only by this
// integration target still link into the crate's own test build.
use {arr_db as _, arr_dl as _, arr_indexer as _, arr_meta as _, chrono as _};
use std::os::unix::fs::PermissionsExt;
use std::path::{Path, PathBuf};
use std::time::Duration;
use arr_e2e::{fixtures, probe_movie_fixture, Daemon};
use wiremock::matchers::{method, path};
use wiremock::{Mock, MockServer, ResponseTemplate};
/// Neither Prowlarr, TMDB nor Transmission matter to a subtitle scenario;
/// every test points them at an address nothing answers on rather than
/// standing up fakes it never calls.
const UNUSED_UPSTREAM: &str = "http://127.0.0.1:1";
/// Insert a movie, an imported grab and a media file directly into the
/// daemon's database, then copy `clip` into the daemon's media root next to
/// it. Returns the media file id and the video's path on disk.
async fn seed_movie_file(
database: &arr_db::Db,
media_root: &Path,
tmdb_id: i64,
release_name: &str,
probed: Option<&serde_json::Value>,
) -> (i64, PathBuf) {
let pool = database.pool();
let root_id: i64 =
sqlx::query_scalar("SELECT id FROM roots WHERE kind = 'movie' AND audience = 'main'")
.fetch_one(pool)
.await
.expect("migrations seed the main movie root");
let movie_id: i64 = sqlx::query_scalar(
"INSERT INTO movies (tmdb_id, title, root_id) VALUES (?, ?, ?) RETURNING id",
)
.bind(tmdb_id)
.bind(format!("Movie {tmdb_id}"))
.bind(root_id)
.fetch_one(pool)
.await
.expect("insert movie");
let folder = media_root.join(format!("Movie {tmdb_id} (2024) - [1080p]"));
std::fs::create_dir_all(&folder).expect("create title folder");
let video = folder.join(format!("Movie {tmdb_id} (2024) - [1080p].mkv"));
std::fs::copy(probe_movie_fixture(), &video).expect("copy the real probe clip");
let size = i64::try_from(std::fs::metadata(&video).expect("clip metadata").len())
.expect("clip size fits in i64");
let media_file_id: i64 = sqlx::query_scalar(
"INSERT INTO media_files (owner_kind, owner_id, path, size, probed)
VALUES ('movie', ?, ?, ?, ?) RETURNING id",
)
.bind(movie_id)
.bind(video.to_string_lossy().into_owned())
.bind(size)
.bind(probed.map(ToString::to_string))
.fetch_one(pool)
.await
.expect("insert media file");
// A release and an imported grab, so `Target::release_name` resolves to
// `release_name` — Podnapisi searches nothing without one.
let release_id: i64 = sqlx::query_scalar(
"INSERT INTO releases (indexer_id, guid, name, size, download_url, parsed, verdict)
VALUES (1, ?, ?, 1000000, 'http://example.invalid/download', '{}', 'eligible')
RETURNING id",
)
.bind(uuid::Uuid::new_v4().to_string())
.bind(release_name)
.fetch_one(pool)
.await
.expect("insert release");
sqlx::query(
"INSERT INTO grabs (release_id, target_kind, target_id, infohash, state, imported_at)
VALUES (?, 'movie', ?, ?, 'imported', strftime('%Y-%m-%dT%H:%M:%fZ', 'now'))",
)
.bind(release_id)
.bind(movie_id)
.bind(uuid::Uuid::new_v4().to_string())
.execute(pool)
.await
.expect("insert grab");
(media_file_id, video)
}
/// `PUT /api/settings/subtitles`, the operator surface the reconcile loop
/// reads every tick (DESIGN.md §15, #198).
async fn configure_subtitles(
base: &str,
wanted_languages: &[&str],
providers_enabled: &[&str],
translation_engine: Option<&str>,
) {
let response = reqwest::Client::new()
.put(format!("{base}/api/settings/subtitles"))
.json(&serde_json::json!({
"wanted_languages": wanted_languages,
"providers_enabled": providers_enabled,
"translation_engine": translation_engine,
"provider_daily_budgets": {},
"translator_daily_budgets": {},
"remote_command_timeout_seconds": 30,
}))
.send()
.await
.expect("put subtitle settings");
assert_eq!(
response.status(),
reqwest::StatusCode::OK,
"subtitle settings accepted: {}",
response.text().await.unwrap_or_default()
);
}
/// Poll `/api/media-files/{id}/subtitles` until a row of the given `origin`
/// shows up — the reconcile lane closes a gap from a detached task, so its
/// effects land some time after the tick that started it (§8, §15).
async fn wait_for_subtitle(base: &str, media_file_id: i64, origin: &str) -> serde_json::Value {
for _ in 0..200 {
let subtitles: Vec<serde_json::Value> =
reqwest::get(format!("{base}/api/media-files/{media_file_id}/subtitles"))
.await
.expect("list subtitles")
.json()
.await
.expect("subtitles json");
if let Some(found) = subtitles
.iter()
.find(|subtitle| subtitle["origin"] == origin)
{
return found.clone();
}
tokio::time::sleep(Duration::from_millis(300)).await;
}
panic!("no {origin} subtitle appeared for media file {media_file_id} in time");
}
/// A remote-command translation backend stub (DESIGN.md §15, issue #193):
/// discards whatever batch it is fed and always answers with the same
/// translated cue. Every scenario here feeds it exactly one embedded English
/// cue ("hello", numbered 1, from the committed probe fixture), so a fixed
/// reply exercises the real IPC without needing a JSON parser in `sh`.
fn write_stub_translator(dir: &Path) -> PathBuf {
let script = dir.join("stub-translate.sh");
std::fs::write(
&script,
"#!/bin/sh\ncat >/dev/null\nprintf '[{\"number\":1,\"text\":\"Ola\"}]'\n",
)
.expect("write stub translator");
let mut permissions = std::fs::metadata(&script)
.expect("stub translator metadata")
.permissions();
permissions.set_mode(0o755);
std::fs::set_permissions(&script, permissions).expect("make stub translator executable");
script
}
async fn connect(daemon: &Daemon) -> arr_db::Db {
arr_db::Db::connect(daemon.database_path())
.await
.expect("connect to the daemon's own database")
}
/// Where DESIGN.md §15 puts a sidecar for `video`, e.g. `<stem>.pt-PT.srt`.
fn sidecar_path(video: &Path, suffix: &str) -> PathBuf {
let stem = video
.file_stem()
.and_then(|stem| stem.to_str())
.expect("the seeded video path has a UTF-8 stem");
video.with_file_name(format!("{stem}.{suffix}"))
}
/// A media file with no subtitles gets one fetched from a provider, synced
/// by real `alass`, and written under the §15 sidecar name — and once
/// satisfied, the loop never asks the provider again.
#[tokio::test]
async fn a_missing_subtitle_is_fetched_synced_and_named() {
let podnapisi = MockServer::start().await;
let pid = "movie-one";
Mock::given(method("GET"))
.and(path("/subtitles/search/advanced"))
.respond_with(
ResponseTemplate::new(200).set_body_string(
serde_json::json!({
"data": [{
"id": pid,
"language": "pt-PT",
"flags": [],
"releases": ["Movie.One.2024.1080p.WEB-DL-GROUP"],
"custom_releases": []
}],
"page": 1,
"all_pages": 1
})
.to_string(),
),
)
.mount(&podnapisi)
.await;
Mock::given(method("GET"))
.and(path(format!("/subtitles/{pid}/download")))
.respond_with(
ResponseTemplate::new(200)
.set_body_raw(fixtures::PODNAPISI_PLAUSIBLE_ZIP, "application/zip"),
)
.mount(&podnapisi)
.await;
let podnapisi_url = format!("{}/subtitles", podnapisi.uri());
let daemon = Daemon::spawn_with_env(
UNUSED_UPSTREAM,
UNUSED_UPSTREAM,
UNUSED_UPSTREAM,
&[("ARR_PODNAPISI_URL", &podnapisi_url)],
)
.await;
let database = connect(&daemon).await;
let (media_file_id, video) = seed_movie_file(
&database,
daemon.media_root(),
900_001,
"Movie.One.2024.1080p.WEB-DL-GROUP",
Some(&serde_json::json!({ "sub_tracks": [] })),
)
.await;
configure_subtitles(daemon.base_url(), &["pt-PT"], &["podnapisi"], None).await;
let subtitle = wait_for_subtitle(daemon.base_url(), media_file_id, "provider").await;
assert_eq!(subtitle["language"], "pt-PT");
assert_eq!(subtitle["provider"], "podnapisi");
assert_eq!(
subtitle["sync"], "synced",
"real alass produced a plausible shift"
);
let expected_path = sidecar_path(&video, "pt-PT.srt");
assert_eq!(
subtitle["path"].as_str(),
Some(expected_path.to_string_lossy()).as_deref()
);
assert!(
expected_path.exists(),
"sidecar written at the §15 name: {}",
expected_path.display()
);
// §15: once satisfied, the loop never searches this language again.
// Wait past one more 30-second reconcile tick and confirm nothing moved.
tokio::time::sleep(Duration::from_secs(35)).await;
let requests = podnapisi
.received_requests()
.await
.expect("request recording is on by default");
assert_eq!(
requests.len(),
3,
"a pt-PT want searches both pt-PT and pt-BR (§15's substitution), plus one \
download, and never again once satisfied: {requests:?}"
);
}
/// A file whose only English text lives in an embedded track gets that track
/// extracted by real `ffmpeg` and translated into a `.mt` sidecar — the
/// improvement over Bazarr §15 calls out explicitly.
#[tokio::test]
async fn an_embedded_track_is_extracted_and_translated() {
let scripts = tempfile::tempdir().expect("tempdir for the stub translator");
let template = write_stub_translator(scripts.path());
let daemon = Daemon::spawn_with_env(
UNUSED_UPSTREAM,
UNUSED_UPSTREAM,
UNUSED_UPSTREAM,
&[(
"ARR_TRANSLATE_COMMAND_TEMPLATE",
template.to_str().expect("utf8 path"),
)],
)
.await;
let database = connect(&daemon).await;
let (media_file_id, video) = seed_movie_file(
&database,
daemon.media_root(),
900_002,
"Movie.Two.2024.1080p.WEB-DL-GROUP",
Some(&serde_json::json!({ "sub_tracks": [
{ "language": "en", "codec": "subrip", "forced": false, "sdh": false }
] })),
)
.await;
// No provider enabled: §15 translates immediately once nothing else has
// the language, no waiting window.
configure_subtitles(daemon.base_url(), &["pt-PT"], &[], Some("command")).await;
let translated = wait_for_subtitle(daemon.base_url(), media_file_id, "translated").await;
assert_eq!(translated["language"], "pt-PT");
assert_eq!(translated["engine"], "command");
let subtitles: Vec<serde_json::Value> = reqwest::get(format!(
"{}/api/media-files/{media_file_id}/subtitles",
daemon.base_url()
))
.await
.expect("list subtitles")
.json()
.await
.expect("subtitles json");
let extracted = subtitles
.iter()
.find(|subtitle| subtitle["origin"] == "extracted")
.expect("the embedded English track was extracted as the translation source");
assert_eq!(extracted["language"], "en");
let sidecar = sidecar_path(&video, "pt-PT.mt.srt");
let content = tokio::fs::read_to_string(&sidecar)
.await
.expect("read the translated sidecar");
assert!(
content.contains("Ola"),
"the stub's translation reached the sidecar: {content}"
);
}
/// A subtitle whose sync is implausible — real `alass` correlating a
/// five-minutes-off cue against a three-second clip — keeps the unsynced
/// original and is flagged, rather than silently writing a bad shift.
#[tokio::test]
async fn an_implausible_sync_keeps_the_original_and_is_flagged() {
let podnapisi = MockServer::start().await;
let pid = "movie-three";
Mock::given(method("GET"))
.and(path("/subtitles/search/advanced"))
.respond_with(
ResponseTemplate::new(200).set_body_string(
serde_json::json!({
"data": [{
"id": pid,
"language": "pt-PT",
"flags": [],
"releases": ["Movie.Three.2024.1080p.WEB-DL-GROUP"],
"custom_releases": []
}],
"page": 1,
"all_pages": 1
})
.to_string(),
),
)
.mount(&podnapisi)
.await;
Mock::given(method("GET"))
.and(path(format!("/subtitles/{pid}/download")))
.respond_with(
ResponseTemplate::new(200)
.set_body_raw(fixtures::PODNAPISI_FARFETCHED_ZIP, "application/zip"),
)
.mount(&podnapisi)
.await;
let podnapisi_url = format!("{}/subtitles", podnapisi.uri());
let daemon = Daemon::spawn_with_env(
UNUSED_UPSTREAM,
UNUSED_UPSTREAM,
UNUSED_UPSTREAM,
&[("ARR_PODNAPISI_URL", &podnapisi_url)],
)
.await;
let database = connect(&daemon).await;
let (media_file_id, video) = seed_movie_file(
&database,
daemon.media_root(),
900_003,
"Movie.Three.2024.1080p.WEB-DL-GROUP",
Some(&serde_json::json!({ "sub_tracks": [] })),
)
.await;
configure_subtitles(daemon.base_url(), &["pt-PT"], &["podnapisi"], None).await;
let subtitle = wait_for_subtitle(daemon.base_url(), media_file_id, "provider").await;
assert_eq!(subtitle["sync"], "rejected");
let sidecar = sidecar_path(&video, "pt-PT.srt");
let content = tokio::fs::read_to_string(&sidecar)
.await
.expect("read the unsynced sidecar");
assert!(
content.contains("completely unrelated dialogue") && content.contains("00:05:00"),
"the unsynced original is kept verbatim, timings untouched: {content}"
);
let queue: serde_json::Value =
reqwest::get(format!("{}/api/queues/subtitles", daemon.base_url()))
.await
.expect("get subtitle queue")
.json()
.await
.expect("queue json");
let movie_gaps = queue["movies"]
.as_array()
.expect("movies array")
.iter()
.find(|movie| movie["media_file_id"] == media_file_id)
.expect("the rejected sync surfaces in the missing-subtitles queue (#202)");
assert!(
movie_gaps["gaps"]
.as_array()
.expect("gaps array")
.iter()
.any(|gap| gap["reason"] == "sync_rejected"),
"{movie_gaps}"
);
}
-1
View File
@@ -29,7 +29,6 @@ tempfile.workspace = true
thiserror.workspace = true
tokio.workspace = true
tracing.workspace = true
zip.workspace = true
[dev-dependencies]
tokio = { workspace = true, features = ["macros", "rt-multi-thread"] }
+1 -3
View File
@@ -38,7 +38,6 @@ pub mod openai;
// `subtitle_settings` columns exist either way (#220).
pub mod openai_endpoint;
pub mod opensubtitles;
pub mod podnapisi;
pub mod srt;
pub mod sync;
pub mod translate;
@@ -63,7 +62,6 @@ pub use openai_endpoint::{
DEFAULT_MODEL as OPENAI_DEFAULT_MODEL,
};
pub use opensubtitles::{moviehash, OpenSubtitles, OpenSubtitlesConfig};
pub use podnapisi::{Podnapisi, PodnapisiBuilder, DEFAULT_BASE_URL as PODNAPISI_DEFAULT_BASE_URL};
pub use srt::Cue;
pub use sync::{
binary_present, Outcome, Rejection, Settled, SyncState, Syncer,
@@ -113,7 +111,7 @@ pub fn compiled_engines() -> Vec<&'static str> {
engines
}
/// One subtitle source: OpenSubtitles.com, Podnapisi (DESIGN.md §15).
/// One subtitle source: OpenSubtitles.com (DESIGN.md §15).
///
/// Boxed futures rather than `async fn`, so that the trait stays object safe
/// and the reconcile loop can hold the enabled providers as
-496
View File
@@ -1,496 +0,0 @@
//! The Podnapisi.net provider (`DESIGN.md` §15).
//!
//! Podnapisi has no `moviehash` lane: it matches on a text search over the
//! release name it was imported under, nothing else. Every candidate this
//! provider returns therefore carries `hash_match: false`, which is exactly
//! what ranking (#185) already expresses without a special case — a plain
//! release-name match and below.
//!
//! The search API is unauthenticated JSON, and download hands back a zip
//! containing exactly one subtitle file. Both are undocumented officially;
//! this follows the shape `subliminal`'s maintained provider uses against the
//! live service.
use std::{
collections::HashSet,
io::{Cursor, Read as _},
time::Duration,
};
use arr_core::{
lang::{resolve_portuguese, PortugueseEvidence},
Language,
};
use reqwest::{StatusCode, Url};
use serde::Deserialize;
use crate::{
error::{Error, Result},
model::{
Candidate, CandidateId, Fetched, MediaFile, MediaRef, ProviderId, SearchRequest,
SubtitleFormat,
},
DownloadFuture, Provider, SearchFuture,
};
/// Podnapisi's public search and download root.
pub const DEFAULT_BASE_URL: &str = "https://www.podnapisi.net/subtitles/";
const DEFAULT_TIMEOUT: Duration = Duration::from_secs(10);
/// A results page claiming more pages than this is treated as malformed
/// rather than looped over forever.
const MAX_PAGES: u32 = 50;
fn provider_id() -> ProviderId {
ProviderId::new("podnapisi")
}
/// The Podnapisi.net subtitle provider.
#[derive(Debug)]
pub struct Podnapisi {
http: reqwest::Client,
base_url: Url,
}
impl Podnapisi {
/// A provider against the real Podnapisi.net, with default timeouts.
///
/// # Errors
///
/// [`Error::Config`] if the HTTP client cannot be built.
pub fn new() -> Result<Self> {
Self::builder().build()
}
/// Start configuring a provider.
#[must_use]
pub fn builder() -> PodnapisiBuilder {
PodnapisiBuilder {
base_url: DEFAULT_BASE_URL.to_owned(),
timeout: DEFAULT_TIMEOUT,
}
}
async fn search_language(
&self,
file: &MediaFile,
language: &Language,
) -> Result<Vec<Candidate>> {
let Some(keywords) = file.release_name.as_deref() else {
// No release name, nothing to search Podnapisi's keyword index
// with. Not an error: the provider simply has nothing to offer.
return Ok(Vec::new());
};
let mut candidates = Vec::new();
let mut seen_pids = HashSet::new();
let mut page = 1;
loop {
let response = self
.fetch_page(keywords, language, &file.media, page)
.await?;
let all_pages = response.all_pages.max(1);
for item in response.data {
if !seen_pids.insert(item.id.clone()) {
continue;
}
candidates.push(item.into_candidate(language));
}
if response.page >= all_pages || page >= MAX_PAGES {
break;
}
page += 1;
}
Ok(candidates)
}
async fn fetch_page(
&self,
keywords: &str,
language: &Language,
media: &MediaRef,
page: u32,
) -> Result<SearchResponse> {
let mut url = self
.base_url
.join("search/advanced")
.map_err(|err| config_error(format!("invalid search URL: {err}")))?;
{
let mut params = url.query_pairs_mut();
params.append_pair("keywords", keywords);
params.append_pair("language", &language_param(language));
match media {
MediaRef::Movie { .. } => {
params.append_pair("movie_type", "movie");
}
MediaRef::Episode {
season, episode, ..
} => {
params.append_pair("movie_type", "tv-series");
params.append_pair("movie_type", "mini-series");
params.append_pair("seasons", &season.to_string());
params.append_pair("episodes", &episode.to_string());
}
}
if page > 1 {
params.append_pair("page", &page.to_string());
}
}
let body = self.get(url).await?;
serde_json::from_str(&body).map_err(|err| Error::Malformed {
provider: provider_id(),
detail: format!("search response did not parse: {err}"),
})
}
/// Download one candidate this provider offered.
async fn download_candidate(&self, id: &CandidateId) -> Result<Fetched> {
let (pid, language) = split_candidate_id(id)?;
let mut url = self
.base_url
.join(&format!("{pid}/download"))
.map_err(|err| config_error(format!("invalid download URL: {err}")))?;
url.query_pairs_mut().append_pair("container", "zip");
let bytes = self.get_bytes(url, id).await?;
let (content, format) = extract_single_file(&bytes)?;
Ok(Fetched {
id: id.clone(),
language,
format,
content,
})
}
async fn get(&self, url: Url) -> Result<String> {
let response = self.send(url, None).await?;
response.text().await.map_err(|err| Error::Transport {
provider: provider_id(),
source: Box::new(err),
})
}
async fn get_bytes(&self, url: Url, candidate: &CandidateId) -> Result<Vec<u8>> {
let response = self.send(url, Some(candidate)).await?;
response
.bytes()
.await
.map(|bytes| bytes.to_vec())
.map_err(|err| Error::Transport {
provider: provider_id(),
source: Box::new(err),
})
}
/// `candidate` is only known when the request is a download — it names
/// the [`Error::NotFound`] a 404 becomes.
async fn send(&self, url: Url, candidate: Option<&CandidateId>) -> Result<reqwest::Response> {
tracing::debug!(url = %url, "Podnapisi request");
let response = self
.http
.get(url)
.send()
.await
.map_err(|err| Error::Transport {
provider: provider_id(),
source: Box::new(err),
})?;
let status = response.status();
if status.is_success() {
return Ok(response);
}
Err(match (status, candidate) {
(StatusCode::UNAUTHORIZED | StatusCode::FORBIDDEN, _) => Error::Unauthorized {
provider: provider_id(),
},
(StatusCode::TOO_MANY_REQUESTS, _) => Error::RateLimited {
provider: provider_id(),
retry_after: retry_after(&response),
},
(StatusCode::NOT_FOUND, Some(candidate)) => Error::NotFound {
provider: provider_id(),
candidate: candidate.clone(),
},
(other, _) => Error::Malformed {
provider: provider_id(),
detail: format!("HTTP {other}"),
},
})
}
}
fn config_error(detail: String) -> Error {
Error::Config {
provider: provider_id(),
detail,
}
}
impl Provider for Podnapisi {
fn id(&self) -> ProviderId {
provider_id()
}
fn search<'a>(&'a self, request: &'a SearchRequest) -> SearchFuture<'a> {
Box::pin(async move {
let mut candidates = Vec::new();
for language in &request.languages {
candidates.extend(self.search_language(&request.file, language).await?);
}
Ok(candidates)
})
}
fn download<'a>(&'a self, id: &'a CandidateId) -> DownloadFuture<'a> {
Box::pin(async move { self.download_candidate(id).await })
}
/// Anonymous, so the lamp is reachability alone (#200): the site root
/// answers, no quota spent.
fn probe(&self) -> crate::ProbeFuture<'_> {
Box::pin(async move { self.send(self.base_url.clone(), None).await.map(|_| ()) })
}
}
/// Configuration for a [`Podnapisi`].
#[derive(Debug, Clone)]
pub struct PodnapisiBuilder {
base_url: String,
timeout: Duration,
}
impl PodnapisiBuilder {
/// Point the provider somewhere other than Podnapisi.net. Tests use
/// this; nothing else should.
#[must_use]
pub fn base_url(mut self, base_url: impl Into<String>) -> Self {
self.base_url = base_url.into();
self
}
/// Per-request timeout.
#[must_use]
pub fn timeout(mut self, timeout: Duration) -> Self {
self.timeout = timeout;
self
}
/// Build the provider.
///
/// # Errors
///
/// [`Error::Config`] if the base URL will not parse or the HTTP client
/// cannot be built.
pub fn build(self) -> Result<Podnapisi> {
let mut base_url = self.base_url;
if !base_url.ends_with('/') {
base_url.push('/');
}
let base_url = Url::parse(&base_url).map_err(|err| Error::Config {
provider: provider_id(),
detail: format!("invalid base URL: {err}"),
})?;
let http = reqwest::Client::builder()
.timeout(self.timeout)
.user_agent(concat!("arr/", env!("CARGO_PKG_VERSION")))
.build()
.map_err(|err| Error::Config {
provider: provider_id(),
detail: err.to_string(),
})?;
Ok(Podnapisi { http, base_url })
}
}
/// One page of Podnapisi's `search/advanced` response.
#[derive(Debug, Deserialize)]
struct SearchResponse {
data: Vec<RawCandidate>,
page: u32,
all_pages: u32,
}
#[derive(Debug, Deserialize)]
struct RawCandidate {
id: String,
language: String,
#[serde(default)]
flags: Vec<String>,
#[serde(default)]
releases: Vec<String>,
#[serde(default)]
custom_releases: Vec<String>,
}
impl RawCandidate {
fn into_candidate(self, requested: &Language) -> Candidate {
let release_name = self.releases.into_iter().chain(self.custom_releases).next();
let claims = release_name.as_deref().map(arr_parse::parse);
let language = resolve_language(
&self.language,
claims.as_ref().map_or(&[], |claims| &claims.languages[..]),
requested,
);
Candidate {
provider: provider_id(),
id: CandidateId::new(format!("{}|{language}", self.id)),
language,
hash_match: false,
group: claims.as_ref().and_then(|claims| claims.group.clone()),
source: claims.and_then(|claims| claims.source).map(Into::into),
release_name,
rating: None,
download_count: None,
forced: false,
sdh: self.flags.iter().any(|flag| flag == "hearing_impaired"),
}
}
}
/// Turn a wanted [`Language`] into the string Podnapisi's `language` query
/// parameter expects — the same BCP-47-shaped spelling [`Language::Display`]
/// already produces for the pt-PT/pt-BR split, lower-cased where this crate
/// has no opinion.
fn language_param(language: &Language) -> String {
match language {
Language::PortuguesePortugal => "pt-PT".to_owned(),
Language::PortugueseBrazil => "pt-BR".to_owned(),
Language::PortugueseUnverified => "pt".to_owned(),
Language::Other(tag) => tag.clone(),
}
}
/// Resolve Podnapisi's own `language` tag on a result, using the release
/// name's claimed language markers as the first signal — the same signal
/// order `arr-core::lang` documents, with the container tag Podnapisi sends
/// standing in for a BCP-47 tag.
fn resolve_language(
raw_tag: &str,
name_markers: &[arr_parse::LanguageMarker],
requested: &Language,
) -> Language {
let primary = raw_tag
.split(['-', '_'])
.next()
.unwrap_or(raw_tag)
.to_ascii_lowercase();
if primary == "pt" || primary == "por" {
return resolve_portuguese(PortugueseEvidence {
name_markers,
stream_title: None,
handler_name: None,
container_tag: Some(raw_tag),
});
}
// Podnapisi echoes back whatever it matched the search on; when the tag
// carries no useful primary subtag, the language we asked for is a
// better answer than a guess.
if primary.is_empty() {
return requested.clone();
}
Language::Other(primary)
}
/// The `CandidateId` this provider issues is `{pid}|{language}`, the
/// language spelled exactly as [`Language::Display`] renders it — so
/// `download` recovers it without a second round trip to disambiguate
/// pt-PT from pt-BR.
fn split_candidate_id(id: &CandidateId) -> Result<(&str, Language)> {
id.as_str()
.rsplit_once('|')
.map(|(pid, language)| (pid, parse_language_tag(language)))
.ok_or_else(|| Error::NotFound {
provider: provider_id(),
candidate: id.clone(),
})
}
/// The inverse of [`Language::Display`], for the tag this module itself
/// wrote into a [`CandidateId`]. Not a general parser: it only has to
/// round-trip the four spellings `Display` can produce.
fn parse_language_tag(tag: &str) -> Language {
match tag {
"pt-PT" => Language::PortuguesePortugal,
"pt-BR" => Language::PortugueseBrazil,
"por-unverified" => Language::PortugueseUnverified,
other => Language::Other(other.to_owned()),
}
}
fn retry_after(response: &reqwest::Response) -> Option<Duration> {
response
.headers()
.get(reqwest::header::RETRY_AFTER)?
.to_str()
.ok()?
.trim()
.parse::<u64>()
.ok()
.map(Duration::from_secs)
}
/// Unzip a Podnapisi download, which is always a single subtitle file.
///
/// # Errors
///
/// [`Error::Malformed`] if the body is not a zip, or does not contain
/// exactly one file.
fn extract_single_file(bytes: &[u8]) -> Result<(Vec<u8>, SubtitleFormat)> {
let mut archive = zip::ZipArchive::new(Cursor::new(bytes)).map_err(|err| Error::Malformed {
provider: provider_id(),
detail: format!("download was not a zip: {err}"),
})?;
if archive.len() != 1 {
return Err(Error::Malformed {
provider: provider_id(),
detail: format!("zip contained {} files, expected 1", archive.len()),
});
}
let mut entry = archive.by_index(0).map_err(|err| Error::Malformed {
provider: provider_id(),
detail: format!("could not read the zipped subtitle: {err}"),
})?;
let format =
entry
.name()
.rsplit_once('.')
.map_or(SubtitleFormat::Other(String::new()), |(_, ext)| {
match ext.to_ascii_lowercase().as_str() {
"srt" => SubtitleFormat::Srt,
"ass" | "ssa" => SubtitleFormat::Ass,
"vtt" => SubtitleFormat::Vtt,
other => SubtitleFormat::Other(other.to_owned()),
}
});
let mut content = Vec::new();
entry
.read_to_end(&mut content)
.map_err(|err| Error::Malformed {
provider: provider_id(),
detail: format!("could not read the zipped subtitle: {err}"),
})?;
Ok((content, format))
}
+1 -1
View File
@@ -9,7 +9,7 @@
// the library's dependencies without using them all directly.
use {
arr_parse as _, chardetng as _, encoding_rs as _, reqwest as _, serde as _, serde_json as _,
thiserror as _, tracing as _, wiremock as _, zip as _,
thiserror as _, tracing as _, wiremock as _,
};
use std::{
+1 -1
View File
@@ -9,7 +9,7 @@
// the library's dependencies without using them all directly.
use {
arr_parse as _, chardetng as _, encoding_rs as _, reqwest as _, serde as _, serde_json as _,
tempfile as _, thiserror as _, tracing as _, zip as _,
tempfile as _, thiserror as _, tracing as _,
};
use std::time::Duration;
+1 -1
View File
@@ -6,7 +6,7 @@
// dependencies without using them directly.
use {
arr_parse as _, chardetng as _, encoding_rs as _, reqwest as _, serde as _, thiserror as _,
tracing as _, zip as _,
tracing as _,
};
use std::time::Duration;
-307
View File
@@ -1,307 +0,0 @@
//! Podnapisi provider tests. Everything runs against `wiremock` — the live
//! service is never touched (DESIGN.md §15).
// Same per-target quirk as the crate's own tests: an integration test links
// the library's dependencies without using them all directly.
use {
arr_parse as _, chardetng as _, encoding_rs as _, reqwest as _, serde as _, serde_json as _,
tempfile as _, thiserror as _, tracing as _, zip as _,
};
use std::time::Duration;
use arr_core::Language;
use arr_subs::{
Candidate, Error, Fetched, MediaFile, MediaRef, Podnapisi, Provider, SearchRequest,
SubtitleFormat,
};
use wiremock::matchers::{method, path, query_param, query_param_is_missing};
use wiremock::{Mock, MockServer, ResponseTemplate};
const SEARCH_MOVIE_PAGE1: &str = include_str!("fixtures/search_movie_page1.json");
const SEARCH_MOVIE_PAGE2: &str = include_str!("fixtures/search_movie_page2.json");
const SEARCH_EPISODE: &str = include_str!("fixtures/search_episode.json");
const SEARCH_MALFORMED: &str = include_str!("fixtures/search_malformed.json");
const DOWNLOAD_ZIP: &[u8] = include_bytes!("fixtures/download.zip");
const DOWNLOAD_MULTI_ZIP: &[u8] = include_bytes!("fixtures/download_multi.zip");
fn provider(server: &MockServer) -> Podnapisi {
Podnapisi::builder()
.base_url(format!("{}/subtitles", server.uri()))
.build()
.expect("provider builds")
}
fn movie_file(release_name: Option<&str>) -> MediaFile {
MediaFile {
path: "/mnt/media/film.mkv".into(),
size: 1_234,
release_name: release_name.map(str::to_owned),
media: MediaRef::Movie { tmdb_id: 42 },
}
}
fn request(file: MediaFile, languages: Vec<Language>) -> SearchRequest {
SearchRequest { file, languages }
}
async fn mount_search(server: &MockServer, page: Option<&str>, body: &str) {
let given = Mock::given(method("GET")).and(path("/subtitles/search/advanced"));
let given = match page {
Some(page) => given.and(query_param("page", page)),
None => given.and(query_param_is_missing("page")),
};
given
.respond_with(ResponseTemplate::new(200).set_body_string(body))
.mount(server)
.await;
}
#[tokio::test]
async fn a_provider_with_no_release_name_searches_nothing() {
// No mock is mounted: a request would fail the test if one were sent.
let server = MockServer::start().await;
let request = request(movie_file(None), vec![Language::Other("en".to_owned())]);
let candidates = provider(&server)
.search(&request)
.await
.expect("nothing to search on is not an error");
assert!(candidates.is_empty());
}
#[tokio::test]
async fn search_paginates_and_dedupes_across_pages() {
let server = MockServer::start().await;
mount_search(&server, None, SEARCH_MOVIE_PAGE1).await;
mount_search(&server, Some("2"), SEARCH_MOVIE_PAGE2).await;
let request = request(
movie_file(Some("Movie.2024.1080p.WEB-DL-GROUP")),
vec![Language::PortugueseBrazil],
);
let candidates = provider(&server)
.search(&request)
.await
.expect("search succeeds");
// Three distinct pids across the two pages; "aaa111" repeats on both and
// must survive only once (DESIGN.md's provider forum note on duplicates).
assert_eq!(candidates.len(), 3);
assert!(candidates.iter().all(|candidate| !candidate.hash_match));
}
#[tokio::test]
async fn a_candidate_carries_the_release_name_group_and_source() {
let server = MockServer::start().await;
mount_search(&server, None, SEARCH_MOVIE_PAGE1).await;
mount_search(&server, Some("2"), SEARCH_MOVIE_PAGE2).await;
let request = request(
movie_file(Some("Movie.2024.1080p.WEB-DL-GROUP")),
vec![Language::PortugueseBrazil],
);
let candidates = provider(&server)
.search(&request)
.await
.expect("search succeeds");
let hearing_impaired = candidates
.iter()
.find(|candidate| candidate.sdh)
.expect("one candidate is flagged hearing_impaired");
assert_eq!(
hearing_impaired.release_name.as_deref(),
Some("Movie.2024.1080p.WEB-DL-GROUP")
);
assert_eq!(hearing_impaired.group.as_deref(), Some("GROUP"));
assert_eq!(hearing_impaired.language, Language::PortugueseBrazil);
assert!(!hearing_impaired.forced);
let plain = candidates
.iter()
.find(|candidate: &&Candidate| {
!candidate.sdh && candidate.language == Language::PortuguesePortugal
})
.expect("the pt-PT candidate from page two");
assert_eq!(plain.group.as_deref(), Some("OTHERGRP"));
}
#[tokio::test]
async fn episode_media_sends_season_and_episode_params() {
let server = MockServer::start().await;
Mock::given(method("GET"))
.and(path("/subtitles/search/advanced"))
.and(query_param("seasons", "1"))
.and(query_param("episodes", "2"))
.respond_with(ResponseTemplate::new(200).set_body_string(SEARCH_EPISODE))
.mount(&server)
.await;
let file = MediaFile {
path: "/mnt/media/show/s01e02.mkv".into(),
size: 999,
release_name: Some("Show.S01E02.1080p.WEB-DL-TEAM".to_owned()),
media: MediaRef::Episode {
tmdb_id: 7,
season: 1,
episode: 2,
},
};
let request = request(file, vec![Language::Other("en".to_owned())]);
let candidates = provider(&server)
.search(&request)
.await
.expect("search succeeds");
assert_eq!(candidates.len(), 1);
assert_eq!(candidates[0].language, Language::Other("en".to_owned()));
}
#[tokio::test]
async fn a_401_is_unauthorized() {
let server = MockServer::start().await;
Mock::given(method("GET"))
.and(path("/subtitles/search/advanced"))
.respond_with(ResponseTemplate::new(401))
.mount(&server)
.await;
let request = request(
movie_file(Some("Movie.2024")),
vec![Language::Other("en".to_owned())],
);
let error = provider(&server)
.search(&request)
.await
.expect_err("401 is an error");
assert!(matches!(error, Error::Unauthorized { .. }), "got {error:?}");
}
#[tokio::test]
async fn a_429_surfaces_retry_after() {
let server = MockServer::start().await;
Mock::given(method("GET"))
.and(path("/subtitles/search/advanced"))
.respond_with(ResponseTemplate::new(429).insert_header("retry-after", "13"))
.mount(&server)
.await;
let request = request(
movie_file(Some("Movie.2024")),
vec![Language::Other("en".to_owned())],
);
let error = provider(&server)
.search(&request)
.await
.expect_err("429 is an error");
match error {
Error::RateLimited { retry_after, .. } => {
assert_eq!(retry_after, Some(Duration::from_secs(13)));
}
other => panic!("expected RateLimited, got {other:?}"),
}
}
#[tokio::test]
async fn a_response_that_does_not_parse_is_malformed() {
let server = MockServer::start().await;
mount_search(&server, None, SEARCH_MALFORMED).await;
let request = request(
movie_file(Some("Movie.2024")),
vec![Language::Other("en".to_owned())],
);
let error = provider(&server)
.search(&request)
.await
.expect_err("an unparseable body is an error");
assert!(matches!(error, Error::Malformed { .. }), "got {error:?}");
}
#[tokio::test]
async fn download_extracts_the_single_zipped_file() {
let server = MockServer::start().await;
Mock::given(method("GET"))
.and(path("/subtitles/aaa111/download"))
.and(query_param("container", "zip"))
.respond_with(ResponseTemplate::new(200).set_body_bytes(DOWNLOAD_ZIP))
.mount(&server)
.await;
let id = arr_subs::CandidateId::new("aaa111|pt-BR");
let fetched: Fetched = provider(&server)
.download(&id)
.await
.expect("download succeeds");
assert_eq!(fetched.id, id);
assert_eq!(fetched.language, Language::PortugueseBrazil);
assert_eq!(fetched.format, SubtitleFormat::Srt);
assert!(fetched.content.starts_with(b"1\n00:00:01"));
}
#[tokio::test]
async fn a_zip_with_more_than_one_file_is_malformed() {
let server = MockServer::start().await;
Mock::given(method("GET"))
.and(path("/subtitles/bbb222/download"))
.respond_with(ResponseTemplate::new(200).set_body_bytes(DOWNLOAD_MULTI_ZIP))
.mount(&server)
.await;
let id = arr_subs::CandidateId::new("bbb222|en");
let error = provider(&server)
.download(&id)
.await
.expect_err("more than one file in the zip is an error");
assert!(matches!(error, Error::Malformed { .. }), "got {error:?}");
}
#[tokio::test]
async fn an_id_this_provider_never_issued_is_not_found() {
let server = MockServer::start().await;
// No mock mounted: a malformed id must fail before any request is sent.
let id = arr_subs::CandidateId::new("no-language-suffix");
let error = provider(&server)
.download(&id)
.await
.expect_err("an id without a language suffix is not this provider's");
assert!(matches!(error, Error::NotFound { .. }), "got {error:?}");
}
#[tokio::test]
async fn a_probe_is_reachability_alone() {
let server = MockServer::start().await;
Mock::given(method("GET"))
.and(path("/subtitles/"))
.respond_with(ResponseTemplate::new(200))
.mount(&server)
.await;
provider(&server)
.probe()
.await
.expect("the anonymous provider is up when the site answers");
server.reset().await;
Mock::given(method("GET"))
.and(path("/subtitles/"))
.respond_with(ResponseTemplate::new(500))
.mount(&server)
.await;
let error = provider(&server).probe().await.expect_err("site down");
assert!(matches!(error, Error::Malformed { .. }), "got {error:?}");
}
+2 -2
View File
@@ -37,8 +37,8 @@ export interface RootInput {
const RESOLUTIONS = ["2160p", "1080p", "720p"] as const;
const SOURCES = ["Remux", "BluRay", "WEB-DL", "WEBRip", "HDTV"] as const;
/** DESIGN.md §15: OpenSubtitles and Podnapisi, no cargo feature gate. */
const PROVIDERS = ["opensubtitles", "podnapisi"] as const;
/** DESIGN.md §15: OpenSubtitles is the available subtitle provider. */
const PROVIDERS = ["opensubtitles"] as const;
/** The runtime-editable half of subtitle config (§15) — issue #198's shape. */
export interface SubtitleSettings {