6 Commits
Author SHA1 Message Date
Worph bdeecd25a0 align with metadata standard
Publish image / publish (push) Failing after 13s
2026-09-13 15:53:06 +02:00
Worph 5f5ed1f483 rework domain field
Publish image / publish (push) Failing after 13s
2026-09-02 00:02:48 +02:00
Worph 759a336bb8 align to metadata keys doc 2026-08-30 19:50:37 +02:00
Worph c136d8a140 rework plugin 2026-08-24 16:32:11 +02:00
Worph ee4036f93f audio support widening 2026-08-21 18:21:32 +02:00
Worph 631beb1141 genre management
Publish image / publish (push) Failing after 14s
2026-08-01 20:22:06 +02:00
16 changed files with 2644 additions and 44 deletions
+16
View File
@@ -53,6 +53,13 @@ const SUBTITLES_TABLE: TableDefinition<'_, &str, &str> = TableDefinition::new("s
/// OpenSubtitles search results keyed by `"<tmdb_id>\x01<lang3>"`. Value is the
/// resolved subtitle's cid, or the literal `"null"` negative-cache sentinel.
const OPENSUBTITLES_TABLE: TableDefinition<'_, &str, &str> = TableDefinition::new("opensubtitles");
/// Small durable plugin scratch values that must survive a restart but aren't
/// content-derived — today, the torznab plugin's Prowlarr indexer-set
/// fingerprint, so a set that changed *while the feeder was down* is still
/// detected as a change on the next boot. Keys are plugin-namespaced
/// (`prowlarr:indexer-set`). redb creates the table on open, so an older
/// `.redb` file gains it transparently.
const MISC_TABLE: TableDefinition<'_, &str, &str> = TableDefinition::new("misc");
/// Per-plugin midhash cache. Cheap to clone (the inner `Database` is shared
/// via `Arc`).
@@ -105,6 +112,7 @@ impl MidhashCache {
tx.open_table(TMDB_PRINCIPAL_TOPN_TABLE)?;
tx.open_table(SUBTITLES_TABLE)?;
tx.open_table(OPENSUBTITLES_TABLE)?;
tx.open_table(MISC_TABLE)?;
tx.commit()?;
}
Ok(Self { db: Arc::new(db) })
@@ -112,6 +120,14 @@ impl MidhashCache {
str_table_accessors!(get_midhash, put_midhash, MIDHASH_TABLE);
str_table_accessors!(
/// Read a durable plugin scratch value (see [`MISC_TABLE`]).
get_misc,
/// Write a durable plugin scratch value (see [`MISC_TABLE`]).
put_misc,
MISC_TABLE,
);
pub fn entry_count(&self) -> Result<u64, redb::Error> {
let tx = self.db.begin_read()?;
let table = tx.open_table(MIDHASH_TABLE)?;
+549
View File
@@ -0,0 +1,549 @@
//! `contentKind` → `domain` / `workForm` — the routing tables from
//! METADATA_KEYS.md §1.
//!
//! `domain` answers **"which application is this record destined for?"**. It
//! is the coarsest facet in the registry and the key every client filters its
//! catalogue on — **one value per app**, no alternation:
//!
//! ```text
//! meta-watch → domain:screen
//! meta-listen → domain:music
//! meta-read → domain:literature
//! ```
//!
//! `workForm` answers the *other* question the old `film`/`tv` split was
//! secretly encoding: **is this work complete in itself, or one instalment of
//! a run?** (`standalone` | `serial`). It is cross-domain — a comic issue is
//! `serial` in `literature` exactly as an episode is `serial` in `screen` —
//! which is what lets a client ask for it independently of the wall.
//!
//! # Why they are stored and not derived on read
//!
//! `contentKind`'s vocabulary is deliberately open — a new feeder may mint a
//! kind and ship without a fleet-wide update. That only stays safe because the
//! domain travels *with* the record: a peer that has never heard of
//! `podcastEpisode` still routes it correctly, because it never has to
//! interpret the token. Deriving on read would silently drop every record
//! whose kind the reader predates. The same argument applies to `workForm`,
//! plus a second one: the two kinds that most need it (`pack` and `card`) are
//! exactly the two the kind alone cannot decide.
//!
//! # ⚠ The binding is permanent and one-way
//!
//! Add kinds freely; **never re-map an existing kind to a different domain**.
//! Because the value is stored, two peers on different versions that disagree
//! about where `audiobook` belongs produce a split-brain that redeploying
//! cannot heal — the wrong domain is already written into records across the
//! mesh. If a kind genuinely has to move, mint a new kind.
//!
//! The `film`+`tv` → `screen` merge (METADATA_KEYS.md §14.17) is the one
//! deliberate exception, taken once and early and paid for with a one-shot
//! meta-core corpus sweep. It is not a precedent.
//!
//! # Why `domain` and not one kind-key per medium
//!
//! `record_matches` is AND-across-keys / OR-within-key, and a non-matching key
//! fails the record outright. A per-medium split (`videoKind` / `audioKind` /
//! `documentKind`) would force meta-listen to express "music video **or**
//! audio track" as an OR across two *different* keys — which the query
//! language cannot represent at all. One domain key, OR-ed within itself,
//! always can. `workForm` is one key everywhere for the same reason.
/// The `contentKind` → `domain` table. A kind absent from this list has no
/// domain and is **not independently routable** — correct for sidecars
/// (subtitles, artwork) and for un-anchored hits, a producer bug for anything
/// else.
///
/// ⚠ `pack` is deliberately absent. A season pack and an album release are the
/// same structural thing, so the kind alone cannot tell `screen` from `music`;
/// the writer stamps `pack`'s domain from its own context (the same reason
/// `derive_file_type` cannot have a second `pack` arm).
///
/// ⚠ `podcast` / `podcastEpisode` are absent too: the `spoken` domain was
/// retired unused (METADATA_KEYS.md §14.17), and the PR that first *writes*
/// those kinds picks their domain.
const DOMAIN_BY_CONTENT_KIND: &[(&str, &str)] = &[
// screen — everything meta-watch consumes, films and serials alike. The
// standalone/serial split lives in `workForm`, not here.
("movie", "screen"),
("series", "screen"),
("episode", "screen"),
// music
("track", "music"),
("album", "music"),
("artist", "music"),
("musicVideo", "music"),
("djMix", "music"),
("liveSet", "music"),
// literature — an audiobook's identity is the book (same author, same ISBN
// family, same series as the ebook), so meta-read groups the two editions
// on one page. The player does not decide the domain.
("book", "literature"),
("audiobook", "literature"),
("comic", "literature"),
("manga", "literature"),
("magazine", "literature"),
// science
("paper", "science"),
];
/// The `contentKind` → `workForm` table (METADATA_KEYS.md §1).
///
/// `standalone` = a work complete in itself. `serial` = an instalment of an
/// ongoing run, **or the container or headline for one** — which is why a
/// `series` card and an `episode` are both `serial`.
///
/// ⚠ `pack` is absent for the same reason it is absent above: a season pack is
/// `serial`, an album release is `standalone`, and only the writer knows which.
const WORK_FORM_BY_CONTENT_KIND: &[(&str, &str)] = &[
// screen
("movie", "standalone"),
("series", "serial"),
("episode", "serial"),
// music — an album is a closed work, not an ongoing run, so every rung of
// the music ladder is standalone.
("track", "standalone"),
("album", "standalone"),
("artist", "standalone"),
("musicVideo", "standalone"),
("djMix", "standalone"),
("liveSet", "standalone"),
// literature — the split that proves `workForm` is not a screen-only axis.
("book", "standalone"),
("audiobook", "standalone"),
("comic", "serial"),
("manga", "serial"),
("magazine", "serial"),
// science
("paper", "standalone"),
];
/// Resolve the domain a `contentKind` belongs to.
///
/// Returns `None` for `pack` (not derivable from the kind — the writer supplies
/// it), for format/sidecar kinds that route nowhere, and for any kind this
/// build has never heard of. A caller that already knows the domain from its
/// own context must prefer that over this table.
pub fn domain_for_content_kind(content_kind: &str) -> Option<&'static str> {
lookup(DOMAIN_BY_CONTENT_KIND, content_kind)
}
/// Resolve the work form a `contentKind` sits on. Same contract as
/// [`domain_for_content_kind`]: `None` means "not derivable", never a guess.
pub fn work_form_for_content_kind(content_kind: &str) -> Option<&'static str> {
lookup(WORK_FORM_BY_CONTENT_KIND, content_kind)
}
fn lookup(table: &'static [(&'static str, &'static str)], key: &str) -> Option<&'static str> {
let k = key.trim();
table
.iter()
.find(|(kind, _)| kind.eq_ignore_ascii_case(k))
.map(|(_, value)| *value)
}
/// Kinds whose `domain` the kind→domain table deliberately cannot decide, so
/// the **writer** supplies it (METADATA_KEYS.md §1). Today that is `pack`
/// alone: a season pack and an album release are the same structural thing, and
/// only the producer knows which — indexer-feeder's `bt.rs` reads the domain off
/// the per-file kind *before* overwriting it with `pack` precisely so the value
/// survives.
///
/// They must ride along in every `domain:` → `contentKind:` expansion. The
/// expansion is only ever a **pre-filter** — the real `domain:` term is enforced
/// afterwards by [`crate::query_eval::record_matches`] — so widening it is free,
/// while omitting these kinds would drop exactly the records whose `domain` is
/// the most authoritative one on the wire.
pub const WRITER_SUPPLIED_DOMAIN_KINDS: &[&str] = &["pack"];
/// The `workForm` twin of [`WRITER_SUPPLIED_DOMAIN_KINDS`], and for the same
/// reason: a season pack is `serial`, an album release is `standalone`.
pub const WRITER_SUPPLIED_WORK_FORM_KINDS: &[&str] = &["pack"];
/// The inverse: every `contentKind` that belongs to `domain`.
///
/// Used at the gateway-routing boundary. A gateway declares the *content kinds*
/// it serves (`served_content_kinds`), not domains, so a `domain:` filter is
/// matched by expanding it here rather than by adding a `served_domains` field
/// to the heartbeat — which would be a wire-protocol change requiring every
/// feeder manifest and every peer to move at once.
///
/// ⚠ Since the `film`+`tv` merge this returns **several** kinds for `screen`
/// (`movie`, `series`, `episode`) where the two old values were disjoint. A
/// caller that reads only the first element silently means "movie" — scan the
/// whole list, or narrow with a `workForm:` term first.
pub fn content_kinds_for_domain(domain: &str) -> Vec<&'static str> {
inverse(DOMAIN_BY_CONTENT_KIND, domain)
}
/// The inverse of the `workForm` table. Same routing-boundary role, same
/// scan-the-whole-list warning as [`content_kinds_for_domain`].
pub fn content_kinds_for_work_form(work_form: &str) -> Vec<&'static str> {
inverse(WORK_FORM_BY_CONTENT_KIND, work_form)
}
fn inverse(table: &'static [(&'static str, &'static str)], value: &str) -> Vec<&'static str> {
let v = value.trim();
table
.iter()
.filter(|(_, val)| val.eq_ignore_ascii_case(v))
.map(|(kind, _)| *kind)
.collect()
}
/// Rewrite a query so a `domain:` / `workForm:` filter can reach a plugin that
/// has never heard of either key.
///
/// Returns `None` when the query carries neither — the common path, and cheaper
/// than cloning.
///
/// # Why translate instead of teaching every plugin
///
/// `record_matches` fails a record outright when a filter key is missing from
/// its fields, and `domain` is missing from every record a not-yet-updated
/// feeder emits (`workForm` from every record written before it existed).
/// Most feeders pin the SDK by git tag, so they cannot pick up a new key on our
/// schedule. Translating `domain:literature` into the `contentKind:`
/// alternation it covers means the wire query a plugin sees never mentions
/// either key — old plugins keep working unchanged, while the dispatcher still
/// enforces the real filters afterwards, against records it has just stamped.
///
/// # The two axes intersect
///
/// `domain:screen workForm:serial` covers `series` and `episode` — the
/// **intersection**, not the union. Widening to the union would hand a serial
/// query the whole film corpus as a pre-filter; the post-stamp check would then
/// throw it away, having already paid for it upstream.
///
/// An explicit `contentKind:` filter already in the query is left alone: it is
/// the narrower of the three, and the `domain`/`workForm` terms are still
/// enforced post-stamp, so the AND semantics hold either way.
pub fn expand_routing_filters(
query: &crate::query::GatewayQuery,
) -> Option<crate::query::GatewayQuery> {
let domains = query.filters.get("domain");
let work_forms = query.filters.get("workForm");
if domains.is_none() && work_forms.is_none() {
return None;
}
let by_domain = domains.map(|vs| expand_values(vs, content_kinds_for_domain));
let by_work_form = work_forms.map(|vs| expand_values(vs, content_kinds_for_work_form));
let mut out = query.clone();
out.filters.remove("domain");
out.filters.remove("workForm");
if out.filters.contains_key("contentKind") {
return Some(out);
}
let mut kinds: Vec<&'static str> = match (by_domain, by_work_form) {
(Some(d), Some(w)) => d.into_iter().filter(|k| w.contains(k)).collect(),
(Some(d), None) => d,
(None, Some(w)) => w,
(None, None) => unreachable!("guarded above"),
};
// Domain-ambiguous kinds come along whenever the expansion is non-empty
// — see [`WRITER_SUPPLIED_DOMAIN_KINDS`]. Skipped when nothing mapped,
// so an unknown value still expands to nothing and leaves the filter
// out entirely (below) rather than narrowing to `contentKind:pack`.
if !kinds.is_empty() {
for k in WRITER_SUPPLIED_DOMAIN_KINDS
.iter()
.chain(WRITER_SUPPLIED_WORK_FORM_KINDS.iter())
{
if !kinds.contains(k) {
kinds.push(k);
}
}
}
// An unknown value — or two axes with no kind in common — expands to
// nothing. Inserting an empty alternation would drop every record at the
// plugin; leaving the filter out lets the plugin answer and the post-stamp
// check do the (correct, empty) filtering.
if !kinds.is_empty() {
out.filters.insert(
"contentKind".to_string(),
kinds.into_iter().map(|k| k.to_string()).collect(),
);
}
Some(out)
}
/// Deprecated alias for [`expand_routing_filters`], kept so a caller pinned to
/// an older SDK keeps compiling. It expands `workForm:` too — the name is the
/// only thing that is stale.
#[deprecated(note = "renamed to expand_routing_filters; it now expands workForm: as well")]
pub fn expand_domain_filter(
query: &crate::query::GatewayQuery,
) -> Option<crate::query::GatewayQuery> {
expand_routing_filters(query)
}
fn expand_values(
values: &[String],
expand: fn(&str) -> Vec<&'static str>,
) -> Vec<&'static str> {
let mut out: Vec<&'static str> = Vec::new();
for v in values {
for k in expand(v) {
if !out.contains(&k) {
out.push(k);
}
}
}
out
}
/// Stamp `domain` on a field map that already carries a `contentKind`.
///
/// The convention from METADATA_KEYS.md §1: **whoever writes `contentKind`
/// writes `domain` in the same breath**. An existing `domain` is never
/// overwritten — a writer that knows better than the table (a `pack`, say) has
/// already set it.
pub fn stamp_domain(fields: &mut std::collections::BTreeMap<String, String>) {
stamp(fields, "domain", domain_for_content_kind);
}
/// Stamp `workForm` the same way, from the same `contentKind`. Call it
/// wherever [`stamp_domain`] is called — the registry treats the two as one
/// breath.
pub fn stamp_work_form(fields: &mut std::collections::BTreeMap<String, String>) {
stamp(fields, "workForm", work_form_for_content_kind);
}
fn stamp(
fields: &mut std::collections::BTreeMap<String, String>,
key: &str,
derive: fn(&str) -> Option<&'static str>,
) {
if fields.contains_key(key) {
return;
}
let Some(kind) = fields.get("contentKind").cloned() else {
return;
};
if let Some(value) = derive(&kind) {
fields.insert(key.to_string(), value.to_string());
}
}
#[cfg(test)]
mod tests {
use super::*;
use std::collections::BTreeMap;
#[test]
fn kinds_map_to_their_domain() {
assert_eq!(domain_for_content_kind("movie"), Some("screen"));
assert_eq!(domain_for_content_kind("episode"), Some("screen"));
assert_eq!(domain_for_content_kind("track"), Some("music"));
// The case that proves medium and domain are different axes.
assert_eq!(domain_for_content_kind("musicVideo"), Some("music"));
// The case that proves the player does not decide the domain.
assert_eq!(domain_for_content_kind("audiobook"), Some("literature"));
}
#[test]
fn film_and_tv_are_one_domain_now() {
// The merge (METADATA_KEYS.md §14.17): meta-watch's wall is one term.
assert_eq!(
domain_for_content_kind("movie"),
domain_for_content_kind("series")
);
// ...and the split it used to carry lives here instead.
assert_eq!(work_form_for_content_kind("movie"), Some("standalone"));
assert_eq!(work_form_for_content_kind("series"), Some("serial"));
assert_eq!(work_form_for_content_kind("episode"), Some("serial"));
}
#[test]
fn work_form_is_not_a_screen_only_axis() {
// The point of the key: it means the same thing in every domain.
assert_eq!(work_form_for_content_kind("comic"), Some("serial"));
assert_eq!(work_form_for_content_kind("book"), Some("standalone"));
assert_eq!(work_form_for_content_kind("album"), Some("standalone"));
}
#[test]
fn the_spoken_domain_is_gone() {
// Retired unused. The kinds stay reserved with no domain until the PR
// that writes them picks one.
assert_eq!(domain_for_content_kind("podcast"), None);
assert_eq!(domain_for_content_kind("podcastEpisode"), None);
assert!(content_kinds_for_domain("spoken").is_empty());
}
#[test]
fn pack_has_no_derivable_domain_or_work_form() {
assert_eq!(domain_for_content_kind("pack"), None);
assert_eq!(work_form_for_content_kind("pack"), None);
}
#[test]
fn unknown_kind_is_none_not_a_guess() {
assert_eq!(domain_for_content_kind("hologram"), None);
assert_eq!(work_form_for_content_kind("hologram"), None);
}
#[test]
fn expansion_is_the_inverse() {
let music = content_kinds_for_domain("music");
assert!(music.contains(&"track"));
assert!(music.contains(&"musicVideo"));
assert!(!music.contains(&"audiobook"));
for kind in content_kinds_for_domain("literature") {
assert_eq!(domain_for_content_kind(kind), Some("literature"));
}
for kind in content_kinds_for_work_form("serial") {
assert_eq!(work_form_for_content_kind(kind), Some("serial"));
}
}
#[test]
fn screen_expands_to_every_rung_not_just_the_first() {
// The regression the merge creates for any caller reading `.first()`.
let screen = content_kinds_for_domain("screen");
assert!(screen.contains(&"movie"));
assert!(screen.contains(&"series"));
assert!(screen.contains(&"episode"));
}
fn q(filters: &[(&str, &[&str])]) -> crate::query::GatewayQuery {
crate::query::GatewayQuery {
raw_text: String::new(),
free_text: String::new(),
filters: filters
.iter()
.map(|(k, vs)| {
(
k.to_string(),
vs.iter().map(|v| v.to_string()).collect::<Vec<_>>(),
)
})
.collect(),
ranges: Vec::new(),
negations: Vec::new(),
}
}
#[test]
fn no_routing_filter_is_left_untouched() {
assert!(expand_routing_filters(&q(&[("fileType", &["video"])])).is_none());
}
#[test]
fn domain_becomes_the_content_kinds_it_covers() {
let out = expand_routing_filters(&q(&[("domain", &["literature"])])).unwrap();
// The plugin never sees `domain:` — that is the whole point.
assert!(!out.filters.contains_key("domain"));
let kinds = out.filters.get("contentKind").unwrap();
assert!(kinds.contains(&"book".to_string()));
assert!(kinds.contains(&"audiobook".to_string()));
assert!(!kinds.contains(&"track".to_string()));
}
#[test]
fn work_form_alone_also_expands() {
let out = expand_routing_filters(&q(&[("workForm", &["serial"])])).unwrap();
assert!(!out.filters.contains_key("workForm"));
let kinds = out.filters.get("contentKind").unwrap();
assert!(kinds.contains(&"episode".to_string()));
assert!(kinds.contains(&"comic".to_string()));
assert!(!kinds.contains(&"movie".to_string()));
}
#[test]
fn the_two_axes_intersect_they_do_not_union() {
let out = expand_routing_filters(&q(&[
("domain", &["screen"][..]),
("workForm", &["serial"][..]),
]))
.unwrap();
let kinds = out.filters.get("contentKind").unwrap();
assert!(kinds.contains(&"series".to_string()));
assert!(kinds.contains(&"episode".to_string()));
// The union would have dragged the whole film corpus in.
assert!(!kinds.contains(&"movie".to_string()));
// A `serial` comic is in `literature`, not on meta-watch's wall.
assert!(!kinds.contains(&"comic".to_string()));
}
#[test]
fn an_explicit_content_kind_filter_wins() {
let out = expand_routing_filters(&q(&[
("domain", &["music"][..]),
("contentKind", &["track"][..]),
]))
.unwrap();
assert_eq!(
out.filters.get("contentKind").map(Vec::as_slice),
Some(&["track".to_string()][..])
);
}
#[test]
fn expansion_carries_the_writer_supplied_kinds() {
// `pack` has no domain in the table (a season pack and an album are the
// same structural thing), so the writer stamps one. Leaving it out of
// the alternation would make a `domain:screen` pre-filter drop the whole
// torrent-pack corpus, which `record_matches` would then never see.
let out = expand_routing_filters(&q(&[("domain", &["screen"])])).unwrap();
let kinds = out.filters.get("contentKind").unwrap();
assert!(kinds.contains(&"episode".to_string()));
assert!(kinds.contains(&"pack".to_string()));
// ...and it rides along on the `workForm` axis too, exactly once.
let out = expand_routing_filters(&q(&[("workForm", &["serial"])])).unwrap();
let kinds = out.filters.get("contentKind").unwrap();
assert_eq!(kinds.iter().filter(|k| *k == "pack").count(), 1);
}
#[test]
fn unknown_value_adds_no_alternation() {
// An empty alternation would drop every record at the plugin; the
// post-stamp check does the (correct, empty) filtering.
let out = expand_routing_filters(&q(&[("domain", &["hologram"])])).unwrap();
// Not even `pack`: a lone `contentKind:pack` would be a NARROWING of a
// filter that should widen to nothing.
assert!(!out.filters.contains_key("contentKind"));
}
#[test]
fn two_axes_with_nothing_in_common_widen_rather_than_narrow() {
// No music kind is `serial`. The pre-filter drops out; the post-stamp
// check still enforces both terms.
let out = expand_routing_filters(&q(&[
("domain", &["music"][..]),
("workForm", &["serial"][..]),
]))
.unwrap();
assert!(!out.filters.contains_key("contentKind"));
}
#[test]
fn stamp_fills_in_and_never_overwrites() {
let mut f: BTreeMap<String, String> = BTreeMap::new();
f.insert("contentKind".into(), "episode".into());
stamp_domain(&mut f);
stamp_work_form(&mut f);
assert_eq!(f.get("domain").map(String::as_str), Some("screen"));
assert_eq!(f.get("workForm").map(String::as_str), Some("serial"));
// A pack writer that already knows its context keeps it.
let mut p: BTreeMap<String, String> = BTreeMap::new();
p.insert("contentKind".into(), "pack".into());
p.insert("domain".into(), "music".into());
p.insert("workForm".into(), "standalone".into());
stamp_domain(&mut p);
stamp_work_form(&mut p);
assert_eq!(p.get("domain").map(String::as_str), Some("music"));
assert_eq!(p.get("workForm").map(String::as_str), Some("standalone"));
// No contentKind → nothing invented.
let mut n: BTreeMap<String, String> = BTreeMap::new();
stamp_domain(&mut n);
stamp_work_form(&mut n);
assert!(n.get("domain").is_none());
assert!(n.get("workForm").is_none());
}
}
+230 -2
View File
@@ -183,9 +183,9 @@ const NZB_RELEASE_MAX_DIGEST: usize = 64;
///
/// Returns `None` when the id isn't hex or the digest would overflow the
/// multihash budget (an oversized host) — the caller drops that row.
pub fn compute_nzb_release_cid(host: &str, release_id: &str) -> Option<String> {
pub fn compute_nzb_release_cid(api_base: &str, release_id: &str) -> Option<String> {
let id_bytes = decode_hex_id(release_id)?;
let host_b = host.as_bytes();
let host_b = api_base.as_bytes();
// identity-multihash digest = varint(host_len) ‖ host ‖ id_bytes
let mut digest = Vec::with_capacity(2 + host_b.len() + id_bytes.len());
@@ -298,6 +298,138 @@ pub fn compute_card_cid(source: &str, id: &str) -> Option<String> {
Some(format!("b{}", base32_lower_no_padding(&wire)))
}
// ---------------------------------------------------------------------------
// Delegated-playback locators — `yt-video` (0x1008) and `ext-play` (0x1009)
// ---------------------------------------------------------------------------
//
// These two are the limit case *beyond* `card` (0x1007). `card` addresses a
// work with no bytes anywhere; these address a **rendition whose bytes exist
// but are permanently someone else's**. An external player renders the media,
// nothing is transported, nothing is content-addressed, nothing is re-seedable.
//
// See `docs/cid-formats.md` §7 and
// `docs/study/listenbrainz-youtube-playback-tier-2026-08-21.md`.
/// Custom multicodec for a **YouTube rendition** — delegated playback.
///
/// Its own codec rather than a `url` (`0x1006`) for three reasons, in
/// descending order of how much trouble the alternative causes:
///
/// * **`0x1006` means "fetch these bytes once and seed them".** Point a `url`
/// locator at a `watch` page and meta-share does exactly that: fetches the
/// HTML, chunks it, seeds it, and links the resulting content CID back onto
/// the record *as the file*. That is the MSR1 poisoning shape, except
/// permanent and replicated across peers. A resolved `googlevideo` URL is no
/// better — time-limited and IP-bound, so it is not a locator at all.
/// * **The id is canonical, a URL is not.** `youtube.com` / `music.youtube.com`
/// / `youtu.be` all address the same video, and query parameters are tracking
/// noise, so the same rendition would mint several different CIDs. Being a
/// pure function of `(kind, id)`, this codec makes two peers that resolve the
/// same track converge on one meta-core record for free, via the
/// `cids/<bareCid>` reverse index — the property that makes [`compute_card_cid`]
/// useful, for the same reason.
/// * **It is tiny** — 11 bytes of id plus a short kind prefix.
///
/// ⚠ **An immutable id is not durable playback.** Takedowns, geo-blocks and
/// embed-disabled uploads all kill a reference whose CID stays valid forever —
/// the same rot class as [`NZB_RELEASE_CODEC`]. The mitigation is at the
/// *record* level (carry several ranked references per recording so a client
/// can fall through), never at the CID level.
pub const YT_VIDEO_CODEC: u64 = 0x1008;
/// Custom multicodec for an **external page to open, never fetch**.
///
/// Wire-identical to `url` (`0x1006`) apart from the codec, and that difference
/// is the entire contract: `0x1006` resolves cache-through and lands in the
/// blockstore, `0x1009` **must be refused by the byte path** and only ever
/// surfaces as an "Open on …↗" affordance. Use it for anything worth linking
/// but not worth a per-provider codec (Bandcamp, Jamendo, an arbitrary stream
/// page).
pub const EXT_PLAY_CODEC: u64 = 0x1009;
/// Encode a [`YT_VIDEO_CODEC`] locator from a `kind` and a bare YouTube id.
///
/// `kind` is `"video"` | `"playlist"` | `"channel"`; `id` is the bare
/// identifier (`4D7u5KF7SP8`, `OLAK5uy_…`, `UCRr1xG_2WIDs18a6cIiCxeA`) — **not**
/// a URL, and not a `watch?v=` fragment. The framing is byte-identical to
/// [`compute_card_cid`] on purpose, so each of the seven rank/decode
/// implementations extends a parser it already wrote rather than learning a
/// novel shape.
///
/// Returns `None` on an empty field or a digest past the multihash budget —
/// the caller drops that reference.
pub fn compute_yt_video_cid(kind: &str, id: &str) -> Option<String> {
kinded_locator_cid(YT_VIDEO_CODEC, kind, id)
}
/// Encode an [`EXT_PLAY_CODEC`] locator from an `http(s)` URL.
///
/// The scheme check is not cosmetic: it is what keeps a `javascript:` or
/// `data:` payload from ever reaching a client that will hand this string to
/// an anchor or a new tab.
pub fn compute_ext_play_cid(url: &str) -> Option<String> {
if !(url.starts_with("http://") || url.starts_with("https://")) {
return None;
}
let digest = url.as_bytes();
if digest.is_empty() || digest.len() > CARD_LOCATOR_MAX_DIGEST {
return None;
}
Some(identity_locator_cid(EXT_PLAY_CODEC, digest))
}
/// Shared encoder for the `varint(len(kind)) ‖ kind ‖ id` locator framing used
/// by [`compute_card_cid`] and [`compute_yt_video_cid`].
fn kinded_locator_cid(codec: u64, kind: &str, id: &str) -> Option<String> {
if kind.is_empty() || id.is_empty() {
return None;
}
let kind_b = kind.as_bytes();
let id_b = id.as_bytes();
let mut digest = Vec::with_capacity(2 + kind_b.len() + id_b.len());
write_pb_varint(kind_b.len() as u64, &mut digest);
digest.extend_from_slice(kind_b);
digest.extend_from_slice(id_b);
if digest.len() > CARD_LOCATOR_MAX_DIGEST {
return None;
}
Some(identity_locator_cid(codec, &digest))
}
/// CIDv1 with an *identity* multihash:
/// `[version=0x01][codec varint][mh=0x00][len varint][digest]`, base32lower.
fn identity_locator_cid(codec: u64, digest: &[u8]) -> String {
let mut wire = Vec::with_capacity(1 + 3 + 2 + digest.len());
wire.push(0x01);
write_pb_varint(codec, &mut wire);
wire.push(0x00); // multihash code: identity
write_pb_varint(digest.len() as u64, &mut wire);
wire.extend_from_slice(digest);
format!("b{}", base32_lower_no_padding(&wire))
}
/// Custom multicodec for the **url-locator** family: a CID whose bytes are an
/// `http(s)` URL rather than a hash of the file (METADATA_KEYS.md §2.1), so a
/// record references external bytes without carrying a raw URL field.
///
/// Copied from the git `meta-feeder-sdk` (`hash.rs`, present since v1.1.0),
/// which this vendored copy predates. Keep the two byte-identical.
pub const URL_LOCATOR_CODEC: u64 = 0x1006;
/// Encode an `http(s)` URL as a `0x1006` identity-multihash CIDv1. The digest
/// *is* the URL, so minting one costs no I/O.
///
/// `None` for anything that is not an absolute http(s) URL — a relative value
/// would encode fine and then fail unresolvably at render time.
pub fn compute_url_cid(url: &str) -> Option<String> {
let url = url.trim();
if !(url.starts_with("http://") || url.starts_with("https://")) {
return None;
}
Some(identity_locator_cid(URL_LOCATOR_CODEC, url.as_bytes()))
}
/// Custom multicodec for a **Usenet posting** identity (`nzb-posting`),
/// minted from the article Message-IDs of a release we scanned ourselves.
///
@@ -535,6 +667,26 @@ pub(crate) fn base32_lower_no_padding(input: &[u8]) -> String {
mod tests {
use super::*;
/// Pinned against an independent encoder. The tmdb plugin's TS
/// `encodeUrlCid` asserts the same strings, so a drift on either side fails.
/// The long URL exercises the two-byte length varint (> 127 bytes).
#[test]
fn url_cid_matches_pinned_vectors() {
assert_eq!(
compute_url_cid("https://image.tmdb.org/t/p/w500/dqZENchTd7lp5zit1Q7Bkjzcxpi.jpg")
.as_deref(),
Some("bagdcaab7nb2hi4dthixs62lnmftwkltunvsgeltpojts65bpoaxxonjqgaxwi4k2ivhgg2cumq3wy4bvpjuximkrg5bgw2t2mn4ha2jonjygo")
);
let long = format!("https://image.tmdb.org/t/p/w500/{}.jpg", "a".repeat(100));
assert_eq!(long.len(), 136);
assert_eq!(
compute_url_cid(&long).as_deref(),
Some("bagdcaaeiafuhi5dqom5c6l3jnvqwozjoorwwiyron5zgol3uf5yc65zvgayc6ylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcylbmfqwcltkobtq")
);
assert!(compute_url_cid("/relative.jpg").is_none());
assert!(compute_url_cid("ftp://example.com/x.jpg").is_none());
}
#[test]
fn midhash256_matches_ts_fixtures() {
assert_eq!(
@@ -672,6 +824,82 @@ mod tests {
assert_eq!(base32_lower_no_padding(b"foobar"), "mzxw6ytboi");
}
#[test]
fn yt_video_cid_matches_golden_vector() {
// Pinned by `/cid-rank-vectors.json`'s `yt-video-locator` entry. The id
// is the real resolver output for Daft Punk / "Get Lucky" measured in
// the delegated-playback study §4.3.
assert_eq!(
compute_yt_video_cid("video", "4D7u5KF7SP8").unwrap(),
"bagecaaarav3gszdfn42ein3vgvfumn2tka4a"
);
}
#[test]
fn yt_video_cid_framing_matches_card_locator() {
// The two codecs share the `varint(len(kind)) || kind || id` framing on
// purpose, so every decoder extends a parser it already wrote. If this
// ever fails, the two shapes have drifted and §7.2 of docs/cid-formats.md
// is no longer true.
let yt = compute_yt_video_cid("tmdb", "tv:95479").unwrap();
let card = compute_card_cid("tmdb", "tv:95479").unwrap();
// Same digest, different codec => same length, differing only in the
// codec varint region.
assert_ne!(yt, card);
assert_eq!(yt.len(), card.len());
}
#[test]
fn yt_video_cid_is_a_pure_function_of_kind_and_id() {
// Convergence is the reason this is a codec and not a URL: two peers
// that resolve the same video must derive the same address offline.
assert_eq!(
compute_yt_video_cid("video", "4D7u5KF7SP8"),
compute_yt_video_cid("video", "4D7u5KF7SP8")
);
// ...and the kind is part of the identity, not decoration.
assert_ne!(
compute_yt_video_cid("video", "OLAK5uy_kZ8Xq"),
compute_yt_video_cid("playlist", "OLAK5uy_kZ8Xq")
);
}
#[test]
fn yt_video_cid_rejects_empty_fields() {
assert!(compute_yt_video_cid("", "4D7u5KF7SP8").is_none());
assert!(compute_yt_video_cid("video", "").is_none());
}
#[test]
fn ext_play_cid_matches_golden_vector() {
assert_eq!(
compute_ext_play_cid("https://example.bandcamp.com/album/x").unwrap(),
"bagesaabenb2hi4dthixs6zlymfwxa3dffzrgc3temnqw24bomnxw2l3bnrrhk3jppa"
);
}
#[test]
fn ext_play_cid_rejects_non_http_schemes() {
// Load-bearing: a client hands this string to an anchor or a new tab,
// so a `javascript:` or `data:` payload must never mint a locator.
assert!(compute_ext_play_cid("javascript:alert(1)").is_none());
assert!(compute_ext_play_cid("data:text/html,<script>").is_none());
assert!(compute_ext_play_cid("ftp://example.com/x").is_none());
assert!(compute_ext_play_cid("").is_none());
}
#[test]
fn ext_play_cid_differs_from_the_url_locator_for_the_same_url() {
// The whole contract lives in the codec slot: `0x1006` means "fetch and
// seed these bytes", `0x1009` means "open this, there are no bytes for
// us". Wire-identical otherwise — which is exactly why a decoder that
// matched on *shape* would seed a Bandcamp page as the file.
let url = "https://example.bandcamp.com/album/x";
let ext = compute_ext_play_cid(url).unwrap();
let card = compute_card_cid("x", url).unwrap();
assert_ne!(ext, card);
}
#[test]
fn ipfs_cid_single_leaf_matches_kubo() {
assert_eq!(
+1
View File
@@ -24,6 +24,7 @@ pub const DEFAULT_META_CORE_URL: &str = "http://metacore-app:9000";
pub mod cache;
pub mod common;
pub mod config;
pub mod domain;
pub mod enrich;
pub mod filename_meta;
pub mod hash;
+42
View File
@@ -184,6 +184,48 @@ impl FeederStore {
&self.meta_core_url
}
/// Raw keys containing `contains` (`GET /api/kv/search`). Returns
/// `(keys, truncated)`; `truncated` means the server capped the result and
/// there may be more.
///
/// This is a keyspace scan, so it belongs on rare control-plane paths only —
/// today, invalidating the gateway's search-coverage markers when a feeder's
/// upstream source set changes. Never call it per search.
pub async fn search_keys(&self, contains: &str, limit: usize) -> anyhow::Result<(Vec<String>, bool)> {
let url = format!("{}/api/kv/search", self.meta_core_url);
let resp = self
.http
.get(&url)
.query(&[("contains", contains), ("limit", &limit.to_string())])
.send()
.await?;
let status = resp.status();
if !status.is_success() {
anyhow::bail!("meta-core KV search `{contains}` returned {status}");
}
#[derive(Deserialize)]
struct KvSearchResponse {
#[serde(default)]
keys: Vec<String>,
#[serde(default)]
truncated: bool,
}
let body: KvSearchResponse = resp.json().await?;
Ok((body.keys, body.truncated))
}
/// Delete one raw KV key (`DELETE /api/kv/value?key=`). A missing key is not
/// an error.
pub async fn delete_kv(&self, key: &str) -> anyhow::Result<()> {
let url = format!("{}/api/kv/value", self.meta_core_url);
let resp = self.http.delete(&url).query(&[("key", key)]).send().await?;
let status = resp.status();
if status.is_success() || status == reqwest::StatusCode::NOT_FOUND {
return Ok(());
}
anyhow::bail!("meta-core KV DELETE {key} returned {status}");
}
/// Self-publish a **metadata-only** outcome to meta-core, returning the
/// **sparse** outcome (`record: None`) so the gateway's `(_, None)` branch
/// skips it — no double write. Byte outcomes and already-sparse outcomes are
+35
View File
@@ -74,6 +74,13 @@ pub struct HashOutcome {
/// digest is over the id set, not over any block, so the `.nzb` manifest
/// travels separately as an ordinary sha2-256 cid in the record's `manifest`
/// field. Produced by `hash::compute_nzb_posting_cid`.
/// - `YtVideo` — a **delegated-playback** reference, custom multicodec
/// `yt-video` `0x1008`. Like `CardLocator` it is always paired with
/// `bytes: None` and lands in the `(None, Some)` metadata-only branch, but it
/// means something stronger than "no bytes yet": the bytes **exist and are
/// permanently someone else's**. Nothing is fetched, seeded or
/// content-addressed, and meta-share's byte path REFUSES the cid with a `400`
/// rather than 404ing. Produced by `hash::compute_yt_video_cid`.
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
pub enum HashKind {
/// Reserved: no feeder currently emits midhash256 outcomes, but the family
@@ -85,6 +92,7 @@ pub enum HashKind {
NzbRelease,
CardLocator,
NzbPosting,
YtVideo,
}
/// Startup-only configuration error returned by [`FeederPlugin::configure`].
@@ -121,6 +129,33 @@ pub fn upstream_id_field(upstream_id: &str) -> String {
format!("{upstream_id}id")
}
/// Key-set prefix for the provenance `source/<label>` field (METADATA_KEYS §5).
/// Each member is its own hash field (`source/gateway:nyaa.si = "true"`) so two
/// producers that reach the same content by different routes union without a
/// last-writer-wins clobber. The `<label>` uses `:` internally (not `/`, which
/// is the key-set path separator) — e.g. `gateway:nyaa.si`, `gateway:tribler`.
pub const SOURCE_KEYSET_PREFIX: &str = "source/";
/// Stamp the SDK's default provenance member — `source/gateway:<upstream_id>` —
/// onto a record's fields, but **only if it carries no `source/*` member yet**.
/// This gives every gateway hit a provenance label for free (so meta-watch never
/// falls back to "unknown" for a feeder record), while a plugin that knows a
/// finer origin (torznab → the specific Prowlarr indexer) stamps its own flat
/// `source/*` facets first (`source/gateway`, `source/<protocol>`,
/// `source/<indexer-slug>`) and this default then no-ops.
pub fn stamp_default_source(
fields: &mut std::collections::BTreeMap<String, String>,
upstream_id: &str,
) {
if fields.keys().any(|k| k.starts_with(SOURCE_KEYSET_PREFIX)) {
return;
}
fields.insert(
format!("{SOURCE_KEYSET_PREFIX}gateway:{upstream_id}"),
"true".to_string(),
);
}
/// Static plugin contract. Each enabled `upstream_id` is implemented by
/// exactly one `Box<dyn FeederPlugin>`. Lives behind `&self` in steady state —
/// called concurrently, owns any interior mutability for caches.
+58
View File
@@ -27,6 +27,16 @@ pub fn record_matches(fields: &BTreeMap<String, String>, query: &GatewayQuery) -
}
continue;
}
// `genres` is a key-set too (`genres/<Name> = "true"`, METADATA_KEYS
// §`genres/{name}`) on everything the card tier writes. Without this arm
// a `genres:` filter looks up a flat `genres` field, finds nothing on a
// card, and drops it — so a genre-scoped row would return empty.
if key == "genres" {
if !genres_filter_matches(fields, allowed) {
return false;
}
continue;
}
let Some(field_value) = fields.get(key) else {
return false;
};
@@ -155,6 +165,54 @@ fn record_has_concrete_language(fields: &BTreeMap<String, String>) -> bool {
/// Key-set members are written as the literal `"true"`; tolerate a couple of
/// obvious truthy spellings and treat anything else as absent.
/// Fold a genre label to its comparison form: lowercase, alphanumerics only.
///
/// Genre display names carry spaces and ampersands (`"Action & Adventure"`,
/// `"Sci-Fi & Fantasy"`, `"Science Fiction"`), and **space is the query DSL's
/// token separator** — so a filter can never spell a genre the way the record
/// stores it. Callers pass a slug (`genres:action-adventure`); folding both
/// sides makes them one value without forcing the stored name to be
/// machine-shaped.
fn genre_fold(s: &str) -> String {
s.chars()
.filter(|c| c.is_alphanumeric())
.flat_map(|c| c.to_lowercase())
.collect()
}
/// True iff the record carries one of the requested genres, in **either**
/// encoding: a truthy `genres/<Name>` key-set member (the card tier's shape) or
/// the legacy flat comma-joined `genres` field. OR across the requested genres,
/// both sides folded — see [`genre_fold`].
///
/// **Fails closed, unlike [`languages_filter_matches`].** A language filter is a
/// viewer *preference* narrowing a set, so an unknown-language release is kept.
/// A genre filter *defines what a row is*: a record with no genre information is
/// not known to be Action, so putting it in an Action row states something
/// untrue. Upstream has already filtered by genre, so this is a re-validation.
fn genres_filter_matches(fields: &BTreeMap<String, String>, allowed: &[String]) -> bool {
let wanted: Vec<String> = allowed.iter().map(|g| genre_fold(g)).collect();
if wanted.iter().all(|g| g.is_empty()) {
return true; // a filter with no usable value constrains nothing
}
for (key, value) in fields {
let Some(name) = key.strip_prefix("genres/") else {
continue;
};
if is_truthy_member(value) && wanted.contains(&genre_fold(name)) {
return true;
}
}
if let Some(flat) = fields.get("genres") {
for member in flat.split(|c| c == ',' || c == '|') {
if wanted.contains(&genre_fold(member)) {
return true;
}
}
}
false
}
fn is_truthy_member(v: &str) -> bool {
matches!(v.trim().to_ascii_lowercase().as_str(), "true" | "1" | "yes")
}
+3
View File
@@ -57,6 +57,7 @@ pub enum HashKindDto {
NzbRelease,
CardLocator,
NzbPosting,
YtVideo,
}
impl From<HashKind> for HashKindDto {
@@ -68,6 +69,7 @@ impl From<HashKind> for HashKindDto {
HashKind::NzbRelease => HashKindDto::NzbRelease,
HashKind::CardLocator => HashKindDto::CardLocator,
HashKind::NzbPosting => HashKindDto::NzbPosting,
HashKind::YtVideo => HashKindDto::YtVideo,
}
}
}
@@ -81,6 +83,7 @@ impl From<HashKindDto> for HashKind {
HashKindDto::NzbRelease => HashKind::NzbRelease,
HashKindDto::CardLocator => HashKind::CardLocator,
HashKindDto::NzbPosting => HashKind::NzbPosting,
HashKindDto::YtVideo => HashKind::YtVideo,
}
}
}
+454 -16
View File
@@ -17,11 +17,14 @@
use std::collections::BTreeMap;
use std::sync::Arc;
use meta_feeder_sdk::hash::compute_card_cid;
use meta_feeder_sdk::hash::{compute_card_cid, compute_url_cid};
use meta_feeder_sdk::types::DiscoveryRecord;
use crate::consts::MAX_CARD_SEASONS;
use crate::tmdb_client::{TmdbClient, TmdbKind, TmdbSeasonSummary};
use crate::tmdb_client::{
iso639_1_to_3, norm_tokens, normalize_title_name, title_member_key, TmdbClient, TmdbImage,
TmdbKind, TmdbSeasonSummary,
};
/// The metadata source that published a card. One variant per bridge; the
/// string form is the CID's source namespace, so **changing it re-mints every
@@ -65,16 +68,30 @@ pub(crate) struct Card {
pub(crate) title: String,
/// Canonical + original + AKA titles. Rides along so the *content* feeders
/// can run their resemblance guard against the card instead of re-querying
/// TMDB for it.
/// TMDB for it (phase 2, once the indexer feeder's own anchor resolution is
/// retired).
///
/// Not read here yet — it is consumed by phase 2, which still resolves its
/// own anchor until the indexer feeder's `build_anchor_record` path is
/// retired. Resolving it now means the card is already complete when that
/// switch happens.
#[allow(dead_code)]
/// Read here for one thing: picking `searchTitle` ([`search_aka`]).
pub(crate) guard_titles: Arc<Vec<String>>,
/// Every name of the work as `(lang3, name)`, filed on the record as the
/// `titles/{lang3}/{name}` key-set (METADATA_KEYS.md). Built by
/// `tmdb_client::title_names`: original under its language, `title` under
/// `eng` when it differs, AKAs by market. Same names as `guard_titles`, with
/// the language each is written in.
pub(crate) names: Vec<(&'static str, String)>,
pub(crate) overview: Option<String>,
pub(crate) poster_path: Option<String>,
/// Editorial genre names (`"Animation"`, `"Drama"`, …) — METADATA_KEYS
/// §`genres`. The *work's* categorisation, distinct from `categories/*`,
/// which is the Newznab/Prowlarr distribution taxonomy a release is filed
/// under and which a card (having no release) never carries.
///
/// Free on both resolution paths: a details fetch returns named genre
/// objects, and a list hit's bare `genre_ids` are mapped through
/// `Resolver::genre_names`. May be empty — TMDB genre coverage is good but
/// not universal, and consumers must treat "no genres" as unknown rather
/// than as "none".
pub(crate) genres: Vec<String>,
pub(crate) year: Option<u16>,
// -- structure (TV only) ------------------------------------------------
@@ -87,6 +104,11 @@ pub(crate) struct Card {
/// from a mutated series card.
#[allow(dead_code)]
pub(crate) season_summaries: Arc<Vec<TmdbSeasonSummary>>,
/// Poster candidates (TMDB `images.posters`), filed as the
/// `posters/{lang3}/{cid}` key-set by [`poster_member_keys`]. Only the by-id
/// path has them; discovery/search cards leave this empty and file no set.
pub(crate) posters: Arc<Vec<TmdbImage>>,
}
impl Card {
@@ -123,6 +145,27 @@ impl Card {
n.clamp(1, MAX_CARD_SEASONS)
}
/// The card's synopsis, trimmed and non-empty.
pub(crate) fn overview_text(&self) -> Option<&str> {
self.overview.as_deref().map(str::trim).filter(|s| !s.is_empty())
}
/// The card's poster path, non-empty.
pub(crate) fn poster(&self) -> Option<&str> {
self.poster_path.as_deref().filter(|s| !s.is_empty())
}
/// Would this card actually render? Exactly [`Card::to_record`]'s emit
/// condition, hoisted so a caller can predict the drop *before* projecting.
///
/// [`crate::discovery`] needs this: it walks catalog pages until it has N
/// cards, and counting raw TMDB hits would overcount — a hit with no poster
/// or no overview is silently declined at projection time, leaving a short
/// row. The two must not drift, hence one predicate rather than two.
pub(crate) fn is_displayable(&self) -> bool {
self.overview_text().is_some() && self.poster().is_some()
}
/// Project to the wire record.
///
/// Returns `None` when the card lacks a poster or a description — it would
@@ -139,16 +182,50 @@ impl Card {
tmdb: &TmdbClient,
query_filters: &BTreeMap<String, Vec<String>>,
) -> Option<DiscoveryRecord> {
let overview = self.overview.as_deref().map(str::trim).filter(|s| !s.is_empty())?;
let poster_path = self.poster_path.as_deref().filter(|s| !s.is_empty())?;
// Same two conditions as `Card::is_displayable`, which the discovery page
// walk uses to predict this drop — keep them in one place.
let overview = self.overview_text()?;
let poster_path = self.poster()?;
let mut fields: BTreeMap<String, String> = BTreeMap::new();
// Type axes. `card` on the fileType axis and the work kind on the
// contentKind axis stay orthogonal, so `fileType:card contentKind:series`
// is a well-formed query and routing needs no special case.
fields.insert("fileType".to_string(), "card".to_string());
fields.insert("contentKind".to_string(), self.content_kind().to_string());
let content_kind = self.content_kind();
fields.insert("contentKind".to_string(), content_kind.to_string());
// Third and fourth axes: which app the card is destined for, and what
// shape of work it is. `fileType=card` says "no bytes", `contentKind`
// says which work, `domain` says who wants it — meta-watch's wall is a
// `domain:screen` query — and `workForm` says film-or-serial, the split
// `domain` stopped carrying when film and tv merged (METADATA_KEYS.md
// §1, §14.17).
if let Some(domain) = meta_feeder_sdk::domain::domain_for_content_kind(content_kind) {
fields.insert("domain".to_string(), domain.to_string());
}
if let Some(work_form) =
meta_feeder_sdk::domain::work_form_for_content_kind(content_kind)
{
fields.insert("workForm".to_string(), work_form.to_string());
}
fields.insert("title".to_string(), self.title.clone());
// The keyword a consumer should search indexers with, only when `title`
// itself can't be searched (`3%` → `3 percent`). meta-watch's title page
// has no TMDB access of its own and reads this off the card; without it
// its free-text query sends `3%`, which one newznab answers with nothing
// and another floods with every release containing a `3`.
if let Some(aka) = search_aka(&self.title, self.guard_titles.iter()) {
// Normalised like a `titles/*/*` member name: `searchTitle` is always
// one of the record's names, spelled the same way.
fields.insert("searchTitle".to_string(), normalize_title_name(aka));
}
// Every name the work is known by, as the language-nested key-set. A
// key-set so two peers resolving different AKAs union on key-merge.
for (lang3, name) in &self.names {
if let Some(key) = title_member_key(lang3, name) {
fields.insert(key, "true".to_string());
}
}
// The id bag.
if let Some(id) = self.tmdbid {
@@ -171,6 +248,24 @@ impl Card {
// `overview`).
fields.insert("description/eng".to_string(), overview.to_string());
// Genres as a KEY-SET (`genres/<Name> = "true"`), not the legacy
// comma-joined `genres` value the tmdb/jellyfin plugins still write.
//
// METADATA_KEYS §14.12 is explicit that `csv-set` is the shape being
// migrated *away* from and that "new writers must not introduce more
// csv-set fields — reach for key-set, even if the surrounding family is
// still on the legacy shape". This is a new writer, and the reason bites
// here specifically: a card is the one record type multiple peers derive
// independently at the same CID (§9.4), so two peers resolving different
// genre sets must union by key-merge rather than diverge into two
// comma-joined strings that need string-diffing to reconcile.
for g in &self.genres {
let g = g.trim();
if !g.is_empty() {
fields.insert(format!("genres/{g}"), "true".to_string());
}
}
// The bare-CID key-set member, exactly as every other feeder stamps on
// its search records (`indexer-feeder`'s `torznab/xml.rs` does the same
// with the release's infohash).
@@ -204,9 +299,30 @@ impl Card {
"poster_url".to_string(),
tmdb.poster_cdn_url(poster_path),
);
// The alternative posters, as the language-nested key-set (METADATA_KEYS
// §6 `posters/{lang3}/{cid}`). Every member is a `url` locator, so listing
// them costs no fetch. The primary member wraps the exact `poster_url`
// string above: the gateway matches on it and renames the member to the
// content cid it seeds into `poster` — that is how "`poster` is a member"
// holds for a feeder card.
for key in poster_member_keys(tmdb, poster_path, &self.posters) {
fields.insert(key, "true".to_string());
}
for (key, allowed) in query_filters {
if key == "languages" || fields.contains_key(key) || allowed.is_empty() {
// `genres` joins `languages` as an echo exclusion, for the same
// reason: it is stored as a key-set, so the flat field this loop
// would write is not the shape a reader looks for. Worse, the filter
// value is a *slug* (`action-adventure` — the query DSL can't spell
// "Action & Adventure", space being its token separator), so echoing
// it would persist a slug masquerading as a genre name alongside the
// real `genres/<Name>` members. Both re-validating tiers read the
// key-set, so nothing needs the echo.
if key == "languages"
|| key == "genres"
|| fields.contains_key(key)
|| allowed.is_empty()
{
continue;
}
fields.insert(key.clone(), allowed.join(","));
@@ -220,6 +336,117 @@ impl Card {
}
}
/// Most members a record's `posters/*` key-set may carry, the primary included
/// (METADATA_KEYS.md §6, writer rule 1).
pub(crate) const MAX_POSTER_MEMBERS: usize = 10;
/// The `posters/{lang3}/{cid}` keys for a card: the primary poster first, then
/// the best-voted alternatives, capped at [`MAX_POSTER_MEMBERS`]. Empty when
/// TMDB returned no image list — without one the primary's language is unknown,
/// and a set is better absent than mislabelled.
///
/// Ranked by `vote_average`, then `vote_count`, then `file_path`, the last so
/// two peers holding the same payload emit the same keys. The language is that
/// of the text printed on the poster: `zxx` for a textless one (TMDB `null`),
/// `und` for a code outside [`iso639_1_to_3`] or a primary TMDB did not list.
///
/// CROSS-BINARY CONTRACT: same ranking, cap, CDN size and language rule as the
/// tmdb plugin's `posterMembers` (`metamesh-plugin-tmdb/src/posters.ts`).
pub(crate) fn poster_member_keys(
tmdb: &TmdbClient,
primary_path: &str,
candidates: &[TmdbImage],
) -> Vec<String> {
if candidates.is_empty() {
return Vec::new();
}
let lang3 = |img: &TmdbImage| -> &'static str {
match img.iso_639_1.as_deref().map(str::trim) {
None | Some("") => "zxx",
Some(code) => iso639_1_to_3(code).unwrap_or("und"),
}
};
let mut keys: Vec<String> = Vec::new();
let primary_lang = candidates
.iter()
.find(|img| img.file_path == primary_path)
.map_or("und", lang3);
if let Some(cid) = compute_url_cid(&tmdb.poster_cdn_url(primary_path)) {
keys.push(format!("posters/{primary_lang}/{cid}"));
}
let mut ranked: Vec<&TmdbImage> = candidates
.iter()
.filter(|img| !img.file_path.trim().is_empty() && img.file_path != primary_path)
.collect();
ranked.sort_by(|a, b| {
b.vote_average
.total_cmp(&a.vote_average)
.then(b.vote_count.cmp(&a.vote_count))
.then(a.file_path.cmp(&b.file_path))
});
for img in ranked {
if keys.len() >= MAX_POSTER_MEMBERS {
break;
}
if let Some(cid) = compute_url_cid(&tmdb.poster_cdn_url(&img.file_path)) {
let key = format!("posters/{}/{cid}", lang3(img));
if !keys.contains(&key) {
keys.push(key);
}
}
}
keys
}
/// Punctuation keyword searches fold to a space. Only used to *exclude* these
/// from [`is_search_hostile_symbol`] — a folded character searches fine.
///
/// CROSS-BINARY CONTRACT: identical to the indexer feeder's
/// `torznab/mod.rs::KEYWORD_PUNCTUATION`, and to meta-watch's
/// `clean_title` + `sanitize_free_text` sets combined.
const KEYWORD_PUNCTUATION: [char; 11] = ['.', '_', ':', ';', ',', '-', '/', '(', ')', '=', '!'];
/// An ASCII symbol an indexer's search box can neither carry nor safely lose
/// (`%` in `3%`): punctuation outside [`KEYWORD_PUNCTUATION`], apostrophe
/// excepted. See the indexer feeder's `is_search_hostile_symbol` for the
/// measurements.
///
/// CROSS-BINARY CONTRACT: same rule as the indexer feeder's
/// `torznab/mod.rs::is_search_hostile_symbol` and meta-watch's
/// `catalog::query::has_search_hostile_symbol`.
fn is_search_hostile_symbol(c: char) -> bool {
c.is_ascii_punctuation() && c != '\'' && !KEYWORD_PUNCTUATION.contains(&c)
}
/// The AKA to search with in place of `title` — the card's `searchTitle`.
/// `None` unless `title` carries a [search-hostile symbol](is_search_hostile_symbol)
/// **and** an AKA is symbol-free and strictly more specific (all of the title's
/// word tokens, plus at least one): `3 percent` for `3%`, never the bare `3`.
/// First qualifying AKA wins; `TmdbAltTitles::all` orders English markets first.
///
/// CROSS-BINARY CONTRACT: mirrors the indexer feeder's
/// `torznab/mod.rs::search_aka`, which picks the anchored jobs' `q=` the same
/// way. Change both.
pub(crate) fn search_aka<'a>(
title: &str,
akas: impl IntoIterator<Item = &'a String>,
) -> Option<&'a str> {
if !title.chars().any(is_search_hostile_symbol) {
return None;
}
let want = norm_tokens(title);
if want.is_empty() {
return None;
}
akas.into_iter()
.filter(|aka| !aka.chars().any(is_search_hostile_symbol))
.find(|aka| {
let have = norm_tokens(aka);
have.len() > want.len() && want.iter().all(|w| have.contains(w))
})
.map(String::as_str)
}
/// Split a card `record_id` (`"tmdb:tv:95479"`) back into `(source, source_id)`
/// — the two halves of the CID preimage. `None` when the id carries no source
/// prefix.
@@ -241,11 +468,18 @@ mod tests {
imdb_id: Some("tt22248376".to_string()),
title: "Frieren: Beyond Journey's End".to_string(),
guard_titles: Arc::new(vec!["Sousou no Frieren".to_string()]),
names: vec![
("jpn", "葬送のフリーレン".to_string()),
("eng", "Frieren: Beyond Journey's End".to_string()),
("jpn", "Sousou no Frieren".to_string()),
],
overview: Some("The story follows the elf mage Frieren.".to_string()),
poster_path: Some("/dqZENchTd7lp5zit1Q7Bkjzcxpi.jpg".to_string()),
genres: vec!["Animation".to_string(), "Sci-Fi & Fantasy".to_string()],
year: Some(2023),
seasons: 1,
season_summaries: Arc::new(Vec::new()),
posters: Arc::new(Vec::new()),
}
}
@@ -314,6 +548,102 @@ mod tests {
assert!(no_overview.to_record(&tmdb, &BTreeMap::new()).is_none());
}
/// `3%` (TMDB tv 68467): the card spells the symbol out for searchers, and
/// keeps the real title for display.
#[test]
fn symbol_title_card_carries_a_search_title() {
let tmdb = TmdbClient::new("token".to_string());
let mut card = tv_card();
card.title = "3%".to_string();
card.guard_titles = Arc::new(
["3%", "3 %", "3", "Three Percent", "3 percent", "3 Pourcent"]
.iter()
.map(|s| s.to_string())
.collect(),
);
let rec = card.to_record(&tmdb, &BTreeMap::new()).expect("record");
assert_eq!(rec.fields["title"], "3%");
assert_eq!(rec.fields.get("searchTitle").map(String::as_str), Some("3 percent"));
// A searchable title gets no searchTitle — Frieren's apostrophe and
// colon are not hostile.
let rec = tv_card().to_record(&tmdb, &BTreeMap::new()).expect("record");
assert!(!rec.fields.contains_key("searchTitle"));
}
#[test]
fn record_files_every_name_as_a_language_nested_key_set() {
let tmdb = TmdbClient::new("token".to_string());
let rec = tv_card().to_record(&tmdb, &BTreeMap::new()).expect("record");
for key in [
"titles/jpn/葬送のフリーレン",
"titles/eng/Frieren: Beyond Journey's End",
"titles/jpn/Sousou no Frieren",
] {
assert_eq!(rec.fields.get(key).map(String::as_str), Some("true"), "{key}");
}
// Leaves only: no scalar at the language level.
assert!(!rec.fields.keys().any(|k| k.starts_with("titles/") && k.matches('/').count() == 1));
}
/// The METADATA_KEYS.md example, end to end from TMDB's data.
#[test]
fn title_names_match_the_three_percent_example() {
use crate::tmdb_client::{title_names, TmdbAltTitle};
let aka = |m: &str, t: &str| TmdbAltTitle { title: t.to_string(), iso_3166_1: m.to_string() };
let names = title_names(
"3%",
Some("3%"),
Some("pt"),
&[aka("US", "3 Percent"), aka("FR", "3 Pourcent"), aka("BR", "3%"), aka("CA", "3 Por Cento ")],
);
assert_eq!(
names,
vec![
("por", "3%".to_string()),
("eng", "3 Percent".to_string()),
("fra", "3 Pourcent".to_string()),
("und", "3 Por Cento".to_string()),
],
"title == original is filed once; BR `3%` repeats por/3%; CA is multilingual → und"
);
// A translated title is filed under eng beside the original.
let names = title_names("Naruto", Some("ナルト"), Some("ja"), &[]);
assert_eq!(names, vec![("jpn", "ナルト".to_string()), ("eng", "Naruto".to_string())]);
// Unknown original language → und; TMDB's placeholder is not a name.
assert_eq!(title_names("(untitled)", Some("X"), Some("xx"), &[]), vec![("und", "X".to_string())]);
}
#[test]
fn title_member_key_normalises_like_two_peers_must() {
assert_eq!(title_member_key("jpn", "Fate/Zero").as_deref(), Some("titles/jpn/Fate\u{2215}Zero"));
assert_eq!(title_member_key("eng", " 3 Percent ").as_deref(), Some("titles/eng/3 Percent"));
assert_eq!(title_member_key("eng", "Pokémon").as_deref(), Some("titles/eng/Pokémon"));
assert_eq!(title_member_key("eng", " "), None);
}
#[test]
fn search_aka_never_picks_something_weaker() {
let akas = |v: &[&str]| v.iter().map(|s| s.to_string()).collect::<Vec<_>>();
assert_eq!(search_aka("M*A*S*H", akas(&["MASH"]).iter()), None);
assert_eq!(search_aka("3%", akas(&["3", "3 %"]).iter()), None);
assert_eq!(search_aka("Law & Order", akas(&["Law and Order"]).iter()), Some("Law and Order"));
assert_eq!(search_aka("Naruto", akas(&["Naruto Shippuden"]).iter()), None);
}
#[test]
fn tmdb_alt_titles_put_english_markets_first() {
let alt: crate::tmdb_client::TmdbAltTitles = serde_json::from_str(
r#"{"results":[{"iso_3166_1":"FR","title":"3 Pourcent"},{"iso_3166_1":"US","title":"3 percent"},{"title":"no market"}]}"#,
)
.expect("decode");
assert_eq!(alt.all(), vec!["3 percent", "3 Pourcent", "no market"]);
assert!(alt.has_markets());
let legacy: crate::tmdb_client::TmdbAltTitles =
serde_json::from_str(r#"{"results":[{"title":"3 Pourcent"},{"title":"3 percent"}]}"#).expect("decode");
assert!(!legacy.has_markets(), "a pre-market cache entry must refetch");
assert!(crate::tmdb_client::TmdbAltTitles::default().has_markets(), "no AKAs is complete");
}
#[test]
fn movie_card_uses_the_movie_content_kind() {
let mut card = tv_card();
@@ -326,17 +656,125 @@ mod tests {
}
/// Query filters are echoed so the record survives `record_matches` on the
/// way out, but `languages` is deliberately skipped (it fails open).
/// way out — except the two key-set fields, which a reader resolves from
/// their `<prefix>/<member>` members rather than from a flat value.
#[test]
fn query_filters_are_echoed_except_languages() {
fn query_filters_are_echoed_except_the_key_set_fields() {
let tmdb = TmdbClient::new("token".to_string());
let mut filters = BTreeMap::new();
filters.insert("fileType".to_string(), vec!["card".to_string()]);
filters.insert("genres".to_string(), vec!["fantasy".to_string()]);
filters.insert("popular".to_string(), vec!["true".to_string()]);
filters.insert("genres".to_string(), vec!["action-adventure".to_string()]);
filters.insert("languages".to_string(), vec!["jpn".to_string()]);
let rec = tv_card().to_record(&tmdb, &filters).expect("record");
assert_eq!(rec.fields["genres"], "fantasy");
assert_eq!(rec.fields["popular"], "true", "ordinary filters echo");
assert_eq!(rec.fields["fileType"], "card", "own value wins the echo");
assert!(!rec.fields.contains_key("languages"));
// The genre filter carries a *slug* the DSL can spell; echoing it would
// persist "action-adventure" as if it were a genre name, beside the real
// `genres/Action & Adventure` members.
assert!(
!rec.fields.contains_key("genres"),
"the genre slug must not be echoed as a flat field"
);
assert_eq!(rec.fields["genres/Animation"], "true");
assert_eq!(rec.fields["genres/Sci-Fi & Fantasy"], "true");
}
fn img(path: &str, lang: Option<&str>, avg: f64, count: u32) -> TmdbImage {
TmdbImage {
file_path: path.to_string(),
iso_639_1: lang.map(str::to_string),
vote_average: avg,
vote_count: count,
}
}
/// METADATA_KEYS §6 `posters/{lang3}/{cid}`: the primary is a member, filed
/// under its own language, as the locator of the exact `poster_url` the
/// gateway seeds — that equality is what lets the gateway rename it.
#[test]
fn record_files_the_primary_poster_as_a_locator_member() {
let tmdb = TmdbClient::new("token".to_string());
let mut card = tv_card();
let primary = card.poster_path.clone().expect("fixture poster");
card.posters = Arc::new(vec![
img("/alt-fr.jpg", Some("fr"), 5.5, 10),
img(&primary, Some("ja"), 5.0, 3),
img("/textless.jpg", None, 6.0, 1),
]);
let rec = card.to_record(&tmdb, &BTreeMap::new()).expect("record");
let primary_cid = compute_url_cid(&rec.fields["poster_url"]).expect("locator");
assert_eq!(
rec.fields.get(&format!("posters/jpn/{primary_cid}")).map(String::as_str),
Some("true")
);
let fr = compute_url_cid(&tmdb.poster_cdn_url("/alt-fr.jpg")).expect("locator");
let textless = compute_url_cid(&tmdb.poster_cdn_url("/textless.jpg")).expect("locator");
assert!(rec.fields.contains_key(&format!("posters/fra/{fr}")));
assert!(rec.fields.contains_key(&format!("posters/zxx/{textless}")));
let members = rec.fields.keys().filter(|k| k.starts_with("posters/")).count();
assert_eq!(members, 3, "the primary is listed once, not twice");
// Leaves only, and never a 639-2/B code.
assert!(!rec.fields.keys().any(|k| k.starts_with("posters/") && k.matches('/').count() != 2));
assert!(!rec.fields.keys().any(|k| k.starts_with("posters/fre/")));
}
#[test]
fn poster_members_rank_by_votes_and_cap_at_ten() {
let tmdb = TmdbClient::new("token".to_string());
let candidates: Vec<TmdbImage> = (0..15)
.map(|i| img(&format!("/p{i:02}.jpg"), Some("en"), f64::from(i), 1))
.collect();
let keys = poster_member_keys(&tmdb, "/primary.jpg", &candidates);
assert_eq!(keys.len(), MAX_POSTER_MEMBERS);
// Primary first; TMDB didn't list it among its images, so `und`.
let primary = compute_url_cid(&tmdb.poster_cdn_url("/primary.jpg")).expect("locator");
assert_eq!(keys[0], format!("posters/und/{primary}"));
// Then the best vote_average: p14, p13, … p06 — p05 is cut.
let best = compute_url_cid(&tmdb.poster_cdn_url("/p14.jpg")).expect("locator");
assert_eq!(keys[1], format!("posters/eng/{best}"));
let cut = compute_url_cid(&tmdb.poster_cdn_url("/p05.jpg")).expect("locator");
assert!(!keys.contains(&format!("posters/eng/{cut}")));
// vote_count breaks a vote_average tie.
let tie = poster_member_keys(
&tmdb,
"/primary.jpg",
&[img("/few.jpg", None, 5.0, 1), img("/many.jpg", None, 5.0, 9)],
);
let many = compute_url_cid(&tmdb.poster_cdn_url("/many.jpg")).expect("locator");
assert_eq!(tie[1], format!("posters/zxx/{many}"));
// An unmapped language is `und`, never a 2-letter key.
let odd = poster_member_keys(&tmdb, "/primary.jpg", &[img("/x.jpg", Some("xx"), 1.0, 1)]);
assert!(odd[1].starts_with("posters/und/"), "{odd:?}");
}
/// No image list (a discovery hit, a pre-append cache entry): no set at all,
/// rather than a primary filed under a guessed `und`.
#[test]
fn card_without_an_image_list_files_no_poster_set() {
let tmdb = TmdbClient::new("token".to_string());
let rec = tv_card().to_record(&tmdb, &BTreeMap::new()).expect("record");
assert!(rec.fields.contains_key("poster_url"), "poster itself is unaffected");
assert!(!rec.fields.keys().any(|k| k.starts_with("posters/")));
}
#[test]
fn tmdb_images_append_decodes() {
let d: crate::tmdb_client::TmdbTvDetails = serde_json::from_str(
r#"{"name":"X","images":{"posters":[{"file_path":"/a.jpg","iso_639_1":null,"vote_average":5.3,"vote_count":4},{"file_path":"/b.jpg","iso_639_1":"fr","vote_average":0}]}}"#,
)
.expect("decode");
let posters = d.images.expect("images").posters;
assert_eq!(posters.len(), 2);
assert_eq!(posters[0].iso_639_1, None);
assert_eq!(posters[1].iso_639_1.as_deref(), Some("fr"));
assert_eq!(posters[1].vote_count, 0);
// A pre-append cache entry still decodes, with no image list.
let old: crate::tmdb_client::TmdbTvDetails =
serde_json::from_str(r#"{"name":"X"}"#).expect("decode");
assert!(old.images.is_none());
}
}
+73
View File
@@ -30,6 +30,70 @@ pub(crate) const TMDB_RATE_LIMIT_DEFAULT_RETRY_SECS: u64 = 5;
/// Maximum time to wait for a TMDB budget permit before giving up (seconds).
pub(crate) const TMDB_WAIT_DEADLINE_SECS: u64 = 3000;
/// Budget-permit deadline for a **catalog page** ([`crate::discovery`]),
/// deliberately far shorter than [`TMDB_WAIT_DEADLINE_SECS`].
///
/// That 3000 s ceiling is sized for a query someone is *waiting on* — a user
/// typed a title and wants that specific answer, so out-waiting a 429 pause
/// beats returning nothing. A browse row is the opposite: nobody asked for
/// *this* row in particular, it refreshes on a timer, and its client keeps the
/// previously-filled row on an empty pass (meta-watch's sticky per-row merge).
/// Inheriting 3000 s would let a single 429 pin a row's search stream open for
/// most of an hour to eventually produce what a 5 s bail produces now.
pub(crate) const TMDB_DISCOVERY_WAIT_DEADLINE_SECS: u64 = 5;
/// Hard ceiling on catalog pages walked for one discovery row, whatever the
/// requested card count. Bounds a misconfigured `card_discovery_n` (or a client
/// asking for 500) to a predictable per-row TMDB cost.
pub(crate) const DISCOVERY_MAX_PAGES: u32 = 5;
/// TMDB keyword ids excluded from an `anime:true` discovery row, as
/// `without_keywords`: `hentai`, `softcore`, `erotica`, `pornography`,
/// `adult animation`.
///
/// **Why this is needed at all.** `include_adult=false` does *not* cover it:
/// TMDB's `adult` boolean is reserved for actual pornography, and every one of
/// the titles this excludes comes back `adult: false`. Measured on the live
/// endpoint, the flag removed 84 of 4465 anime series and none of the offenders.
///
/// **Why they surface in the first place.** `sort_by=popularity.desc` sorts on
/// TMDB's *traffic* metric (page views, API hits, list adds), not on audience
/// size or rating. Adult titles attract disproportionate scraping and search
/// traffic, so they score far above their audience: measured, one carried
/// popularity 313 on **26 votes** and another 242 on **7 votes**, against
/// Bleach's 92 on **2197**. Narrowing the pool to the anime keyword removes the
/// mainstream ballast that keeps them off the front of `/tv/popular`, which is
/// why only the anime rows show it.
///
/// **Why these five and not `erotic`/`ecchi`.** Those two are *content
/// descriptors* a mainstream show can legitimately carry — Mushoku Tensei
/// (tmdb 94664, 1626 votes) carries both — so excluding them costs real titles.
/// The five here tag adult *productions*, and no false positive was observed.
/// Keyword ids are stable; names resolved via `search/keyword`.
pub(crate) const DISCOVERY_EXCLUDED_KEYWORDS: &str = "198385,155477,325693,445,161919";
/// Minimum TMDB vote count for a discovery row entry.
///
/// The residue filter behind [`DISCOVERY_EXCLUDED_KEYWORDS`]: a handful of
/// erotic-tagged titles carry none of the five excluded keywords, and they are
/// uniformly low-signal (7–37 votes). A vote requires a real logged-in account,
/// which is exactly the signal the traffic-based popularity score lacks, so a
/// floor separates them from genuinely obscure-but-real titles far better than
/// popularity can.
///
/// 50 is deliberately low. It keeps Ranma½ (169) and Doraemon (237) — it is a
/// noise gate, not a popularity contest. Raising it much further starts costing
/// legitimate niche titles (Akane-banashi 25, Gravitation 17), which is why the
/// keyword exclusion carries the main load and this only sweeps up behind it.
///
/// Applies to `/discover` rows, which is now every mode except non-anime
/// **trending** — the one signal with no `/discover` equivalent, and therefore
/// the one row that cannot carry this gate. The popular/top-rated modes were
/// moved off `/{kind}/popular` and `/{kind}/top_rated` precisely so they could:
/// three hentai titles were measured at positions 4, 5 and 17 of the
/// *non-anime* `tv/popular` row.
pub(crate) const DISCOVERY_MIN_VOTES: u32 = 50;
/// Default number of cards returned for a free-text query (top-N by TMDB
/// popularity).
///
@@ -47,6 +111,15 @@ pub(crate) const DEFAULT_CARD_TOP_N: usize = 10;
/// never needs a cache wipe — reads slice the cached list to the configured N.
pub(crate) const CACHED_PRINCIPAL_DEPTH: usize = 30;
/// Default number of cards a **keyword-less catalog** query returns (a home
/// row: `popular:true contentKind:series fileType:card`).
///
/// 20 is TMDB's own catalog page size, so the default row costs exactly one
/// TMDB call. Raising it past 20 is legal — [`crate::discovery`] walks further
/// pages — but each page is another call against the shared budget, paid on
/// every warm pass of every row.
pub(crate) const DEFAULT_CARD_DISCOVERY_N: usize = 20;
/// Defensive ceiling on a card's reported season count (guards a corrupt
/// `number_of_seasons`).
pub(crate) const MAX_CARD_SEASONS: u32 = 50;
+722
View File
@@ -0,0 +1,722 @@
//! Keyword-less catalog discovery: TMDB's popular / trending / top-rated lists
//! → [`Card`]s.
//!
//! The card tier's *browse* surface, as opposed to its search surface. A client
//! with a home page (meta-watch's wall) cannot ask "what should I show?" through
//! a text query — there is no text. This branch answers a query whose intent is
//! carried entirely by structured filters, by calling TMDB's catalog endpoints.
//!
//! ## Query dialect
//!
//! Deliberately identical to the indexer feeder's older `discovery.rs`, which
//! answers the same dialect with metadata-only *seeds*. Same words, so a client
//! switches tiers by swapping `fileType:video` for `fileType:card` and nothing
//! else:
//!
//! - **mode** (one of) — `popular:true` and `top_rated:true` are served by
//! `/discover/{movie,tv}` sorted by popularity / rating (rather than the
//! `/{movie,tv}/popular` list endpoints, which cannot carry the adult gate —
//! see `DISCOVERY_EXCLUDED_KEYWORDS`); `trending:true` → the real
//! `/trending/{movie,tv}/week`, the one mode with no `/discover` equivalent
//! and therefore the one row that stays ungated.
//! - **kind** (required) — `contentKind:movie` → films,
//! `contentKind:series` (or `episode`/`tvshow`/`tv`) → shows. Also what routes
//! the query here at all, alongside `fileType:card`.
//! - **anime** (optional) — `anime:true` narrows `/discover` to TMDB's *anime*
//! keyword, sorted to match the mode. Anime trending has no keyword-filtered
//! endpoint either, so it approximates with popularity — and stays gated.
//!
//! ## One call, complete cards
//!
//! A TMDB catalog list already carries `title`, `overview`, `poster_path` and
//! `year` per entry — everything [`Card::to_record`] needs to emit a renderable
//! card. So a row costs **one** TMDB call, not one plus twenty by-id lookups.
//!
//! The price is a thinner id bag: `tvdbid`/`imdbid` come from `external_ids`,
//! which a list response has no room for, so a discovery card carries only its
//! `tmdbid`. That is the id consumers actually key on (it is what the detail
//! page's phase-2 `tmdbid:` query is built from), and the full bag is resolved
//! anyway the moment someone clicks the card — `card_by_id` is a different code
//! path and still fetches it. Paying twenty extra calls per row up front to
//! pre-fill ids nobody reads until a click would defeat the branch's purpose.
//!
//! ## Why nothing is cached here
//!
//! A catalog list is **mutable by nature** — "popular this week" is the one
//! thing that must not be frozen. The SDK's redb tables are permanent-hit caches
//! sized for immutable data (`external_ids` is explicitly "cached forever"), so
//! persisting a popularity list there would pin the wall to whatever it looked
//! like on first boot. The layer above already de-duplicates the repeat cost:
//! the gateway persists each emitted card to meta-core and its search-coverage
//! gate then skips a repeated `(upstream, query)` for an hour, serving meta-core
//! instead. Every call here is budget-gated, so a burst cannot run away.
use std::collections::HashMap;
use std::sync::Arc;
use meta_feeder_sdk::plugin::GatewayQuery;
use tracing::debug;
use crate::card::{Card, CardSource};
use crate::consts::{DISCOVERY_EXCLUDED_KEYWORDS, DISCOVERY_MAX_PAGES, DISCOVERY_MIN_VOTES};
use crate::resolve::{dedup_titles, Resolver};
use crate::tmdb_client::{title_names, TmdbHit, TmdbKind};
/// TMDB keyword id for "anime". `/discover` is the only catalog endpoint that
/// takes a keyword filter, which is why `anime:true` switches endpoints rather
/// than adding a parameter.
const TMDB_ANIME_KEYWORD: &str = "210024";
/// Results TMDB returns per catalog page — used to translate a desired card
/// count into a page count.
const TMDB_PAGE_SIZE: usize = 20;
/// The catalog-list flavour a discovery query asks for.
#[derive(Clone, Copy, PartialEq, Eq, Debug)]
pub(crate) enum DiscoveryMode {
Popular,
Trending,
TopRated,
}
impl DiscoveryMode {
/// The query-filter key that selects this mode.
///
/// It is also echoed onto every emitted record — not here, but by
/// [`Card::to_record`]'s trailing filter echo, which stamps back every
/// filter the query carried that the card did not already set. That echo is
/// load-bearing: both the gateway dispatcher and meta-search's consumer-side
/// `record_matches` re-apply the query's filters to each record, and a
/// *missing* field fails the match, so an un-echoed `popular` would see
/// every card dropped at one of the two tiers.
fn marker(self) -> &'static str {
match self {
DiscoveryMode::Popular => "popular",
DiscoveryMode::Trending => "trending",
DiscoveryMode::TopRated => "top_rated",
}
}
/// `sort_by` value for the `/discover` (anime) path. `/discover` has no
/// "trending", so trending approximates to popularity there — the non-anime
/// path still uses the real `/trending` endpoint.
fn discover_sort(self) -> &'static str {
match self {
DiscoveryMode::TopRated => "vote_average.desc",
DiscoveryMode::Popular | DiscoveryMode::Trending => "popularity.desc",
}
}
}
/// True when this is a keyword-less catalog query this branch should answer.
///
/// Three conditions, and each excludes a query that belongs to a *different*
/// branch of `handle_query`: no free text (that is the principal search), no
/// `tmdbid:` filter (that is the direct by-id lookup a deep link uses), and a
/// truthy mode marker. Note `contentKind` alone deliberately does **not** select
/// this branch — a text query carries it too.
pub(crate) fn is_discovery_query(query: &GatewayQuery) -> bool {
query.free_text.trim().is_empty()
&& !query.filters.contains_key("tmdbid")
&& discovery_mode(query).is_some()
}
/// First truthy mode marker on the query, in precedence order.
fn discovery_mode(query: &GatewayQuery) -> Option<DiscoveryMode> {
for mode in [
DiscoveryMode::Trending,
DiscoveryMode::Popular,
DiscoveryMode::TopRated,
] {
if filter_is_true(query, mode.marker()) {
return Some(mode);
}
}
None
}
/// True iff `query.filters[key]` carries a truthy value.
fn filter_is_true(query: &GatewayQuery, key: &str) -> bool {
query
.filters
.get(key)
.is_some_and(|v| v.iter().any(|s| s.eq_ignore_ascii_case("true")))
}
/// Map the query's `contentKind` filter to a TMDB media kind.
///
/// `series` is the canonical spelling for a card (it describes a whole work),
/// but `episode`/`tvshow`/`tv` are accepted so the dialect stays a superset of
/// the indexer feeder's — a client that has not yet switched its row queries
/// still resolves the right endpoint here, even though its records will then be
/// dropped downstream for the mismatch.
fn discovery_kind(query: &GatewayQuery) -> Option<TmdbKind> {
let values = query.filters.get("contentKind")?;
for v in values {
match v.trim().to_ascii_lowercase().as_str() {
"movie" => return Some(TmdbKind::Movie),
"series" | "episode" | "tvshow" | "tv" => return Some(TmdbKind::Tv),
_ => {}
}
}
None
}
/// The genre slugs a `genres:` filter requested, folded for comparison.
///
/// Multi-valued = OR, matching the filter semantics on both re-validating tiers.
fn requested_genres(query: &GatewayQuery) -> Vec<String> {
query
.filters
.get("genres")
.map(|vs| {
vs.iter()
.map(|v| genre_fold(v))
.filter(|v| !v.is_empty())
.collect()
})
.unwrap_or_default()
}
/// Fold a genre label for comparison: lowercase, alphanumerics only.
///
/// Must stay identical to `genres_filter_matches`'s fold in the SDK and in
/// meta-search — those two re-apply the filter to every record this emits, so a
/// divergence here means the row's own results get dropped downstream.
fn genre_fold(s: &str) -> String {
s.chars()
.filter(|c| c.is_alphanumeric())
.flat_map(|c| c.to_lowercase())
.collect()
}
/// Resolve the requested genre slugs to TMDB genre ids via the id → name table.
/// Unknown slugs are dropped; if none resolve the caller omits `with_genres`
/// rather than sending an empty filter that TMDB would reject.
fn with_genres_ids(wanted: &[String], names: &HashMap<u32, String>) -> Vec<u32> {
let mut ids: Vec<u32> = names
.iter()
.filter(|(_, name)| wanted.contains(&genre_fold(name)))
.map(|(id, _)| *id)
.collect();
ids.sort_unstable();
ids
}
/// Build the `path_and_query` for `TmdbClient::discovery_list` from the mode,
/// kind, anime flag, requested genre ids, and 1-based `page`.
fn build_path_and_query(
mode: DiscoveryMode,
kind: TmdbKind,
anime: bool,
genre_ids: &[u32],
page: u32,
) -> String {
let seg = match kind {
TmdbKind::Movie => "movie",
TmdbKind::Tv => "tv",
};
// Trending is the one mode with no `/discover` equivalent: `/trending` is a
// distinct, TMDB-computed "this week" signal, not a sort order. So it keeps
// the list endpoint — and, being a list endpoint, **cannot carry the adult
// gate below**. The anime variant still routes through `/discover` (there is
// no keyword-filtered trending at all), approximating it with popularity.
// …and a genre-scoped row can't use it either: `/trending` takes no
// `with_genres`, and returning an unfiltered trending page for an "Action"
// row would fill it with whatever is trending. `/discover` sorted by
// popularity is the closest filterable approximation.
if matches!(mode, DiscoveryMode::Trending) && !anime && genre_ids.is_empty() {
return format!("trending/{seg}/week?page={page}");
}
// Everything else goes through `/discover`, including the non-anime rows
// that used to use `/{seg}/popular` and `/{seg}/top_rated`.
//
// Those fixed list endpoints accept neither `without_keywords` nor
// `vote_count.gte`, and measurement showed they genuinely need them: three
// hentai titles reached positions 4, 5 and 17 of the *non-anime* `tv/popular`
// row for the same traffic-metric reason described on
// [`DISCOVERY_EXCLUDED_KEYWORDS`]. `/discover` with `sort_by=popularity.desc`
// is the filterable equivalent, so one code path now carries the gate for
// every mode that can hold it.
//
// Top-rated floors higher than the shared gate: sorting by rating puts a
// single 10.0 vote on top, which a 50-vote gate doesn't stop.
let votes = match mode {
DiscoveryMode::TopRated => 200,
_ => DISCOVERY_MIN_VOTES,
};
let keyword = if anime {
format!("with_keywords={TMDB_ANIME_KEYWORD}&")
} else {
String::new()
};
// `with_genres` is comma-joined = AND on TMDB's side, pipe-joined = OR. A
// multi-valued filter means OR everywhere else in the DSL, so join with `|`
// to keep the wire semantics and the upstream semantics saying the same
// thing. URL-encoded, since a bare `|` is not valid in a query string.
let genres = if genre_ids.is_empty() {
String::new()
} else {
let joined = genre_ids
.iter()
.map(|id| id.to_string())
.collect::<Vec<_>>()
.join("%7C");
format!("with_genres={joined}&")
};
format!(
"discover/{seg}?{keyword}{genres}sort_by={}&include_adult=false\
&without_keywords={DISCOVERY_EXCLUDED_KEYWORDS}&vote_count.gte={votes}&page={page}",
mode.discover_sort(),
)
}
/// Project one catalog hit into a [`Card`], with no further network calls.
///
/// `source_id` uses the same `"tv:{id}"` / `"movie:{id}"` shape
/// `Resolver::card_by_id` produces, so a discovery card's `record_id` — and
/// therefore its `0x1007` locator CID — is byte-identical to the one the by-id
/// path derives for the same work. That is what lets a card the wall surfaced
/// and a card the search box surfaced converge on one record instead of two.
fn card_from_hit(hit: &TmdbHit, kind: TmdbKind, genres: &HashMap<u32, String>) -> Card {
let seg = match kind {
TmdbKind::Movie => "movie",
TmdbKind::Tv => "tv",
};
let mut titles = vec![hit.title.clone()];
if let Some(o) = &hit.original_title {
titles.push(o.clone());
}
titles.extend(hit.alt_titles.iter().cloned());
Card {
source: CardSource::Tmdb,
kind,
source_id: format!("{seg}:{}", hit.tmdbid),
tmdbid: Some(hit.tmdbid),
// Not available on a list response — see the module doc. Resolved on
// the click-time `card_by_id` path instead.
tvdb_id: None,
imdb_id: None,
title: hit.title.clone(),
guard_titles: Arc::new(dedup_titles(titles)),
// A list hit carries no AKAs, so this is title + original only; the
// click-time `card_by_id` card adds the AKAs under the same CID.
names: title_names(
&hit.title,
hit.original_title.as_deref(),
hit.original_language.as_deref(),
&hit.akas,
),
overview: hit.overview.clone().filter(|s| !s.trim().is_empty()),
poster_path: hit.poster_path.clone().filter(|s| !s.trim().is_empty()),
// A list hit carries bare genre ids; the table names them.
genres: hit.genre_ids.iter().filter_map(|id| genres.get(id).cloned()).collect(),
year: hit.year,
// A list response carries no season structure. `seasonCount` is simply
// omitted from the record (`Card::to_record` skips it at `seasons == 0`).
seasons: 0,
season_summaries: Arc::new(Vec::new()),
// A list response carries no `images` append, so a discovery card files
// no `posters/*` set; the click-time `card_by_id` card adds it under the
// same CID.
posters: Arc::new(Vec::new()),
}
}
/// Resolve a keyword-less catalog query to up to `max_results` cards, walking
/// TMDB pages until the cap is reached or a page comes back empty.
///
/// Best-effort throughout: no mode or no `contentKind` yields `[]`, and a
/// rate-limit or miss breaks the walk and returns what was already gathered.
/// The branch never errors — an empty row is the failure mode, matching
/// `Resolver::cards_for_text`.
pub(crate) async fn cards_for_discovery(
resolver: &Resolver,
query: &GatewayQuery,
max_results: usize,
) -> Vec<Card> {
let (Some(mode), Some(kind)) = (discovery_mode(query), discovery_kind(query)) else {
return Vec::new();
};
let anime = filter_is_true(query, "anime");
// Fetched once per process, not per row (see `Resolver::genre_names`).
let gmap = resolver.genre_names(kind).await;
// A `genres:` filter narrows the row upstream via `with_genres`, so the row
// is a full page OF that genre rather than a page filtered down to it.
let wanted = requested_genres(query);
let genre_ids = with_genres_ids(&wanted, &gmap);
if !wanted.is_empty() && genre_ids.is_empty() {
// Asked for a genre this kind doesn't have (e.g. `genres:kids` on the
// movie side, where TMDB files it as "Family"). Returning an unfiltered
// page would silently fill the row with the wrong thing.
debug!(
target: "meta-share::gateway", upstream = "tmdb",
?wanted, "discovery: no TMDB genre id matched; row left empty"
);
return Vec::new();
}
// One page of headroom over the arithmetic minimum: `out` counts *displayable*
// cards, and a catalog page reliably contains a few entries with no overview
// or no poster (newly-added or non-English works), which are declined at
// projection. Without the spare page a 20-card row comes back at 17.
let pages = (max_results.div_ceil(TMDB_PAGE_SIZE).max(1) as u32 + 1).min(DISCOVERY_MAX_PAGES);
let mut out: Vec<Card> = Vec::new();
for page in 1..=pages {
if out.len() >= max_results {
break;
}
let path_and_query = build_path_and_query(mode, kind, anime, &genre_ids, page);
let Some(hits) = resolver.discovery_page(kind, &path_and_query).await else {
break; // budget exhausted, 429, or miss — keep what we have
};
if hits.is_empty() {
break; // walked past the last catalog page
}
out.extend(
hits.iter()
.map(|h| card_from_hit(h, kind, &gmap))
.filter(Card::is_displayable),
);
}
out.truncate(max_results);
debug!(
target: "meta-share::gateway",
upstream = "tmdb",
mode = mode.marker(),
anime,
count = out.len(),
"resolved discovery cards"
);
out
}
#[cfg(test)]
mod tests {
use super::*;
use std::collections::BTreeMap;
fn query(filters: &[(&str, &str)]) -> GatewayQuery {
let mut map: BTreeMap<String, Vec<String>> = BTreeMap::new();
for (k, v) in filters {
map.entry(k.to_string()).or_default().push(v.to_string());
}
GatewayQuery {
raw_text: String::new(),
free_text: String::new(),
filters: map,
ranges: Vec::new(),
negations: Vec::new(),
}
}
#[test]
fn selects_only_keyword_less_mode_queries() {
assert!(is_discovery_query(&query(&[
("popular", "true"),
("contentKind", "series"),
("fileType", "card"),
])));
// A mode marker but with free text → the principal search owns it.
let mut with_text = query(&[("popular", "true"), ("contentKind", "series")]);
with_text.free_text = "frieren".to_string();
assert!(!is_discovery_query(&with_text));
// An explicit id → the by-id lookup owns it.
assert!(!is_discovery_query(&query(&[
("popular", "true"),
("tmdbid", "95479"),
])));
// contentKind alone must NOT select this branch.
assert!(!is_discovery_query(&query(&[
("contentKind", "series"),
("fileType", "card"),
])));
// A falsey marker is not a marker.
assert!(!is_discovery_query(&query(&[("popular", "false")])));
}
/// `contentKind:series` is the canonical card spelling; the indexer feeder's
/// older `episode` spelling still resolves the same endpoint.
#[test]
fn content_kind_maps_series_and_the_legacy_spellings() {
assert_eq!(
discovery_kind(&query(&[("contentKind", "series")])),
Some(TmdbKind::Tv)
);
assert_eq!(
discovery_kind(&query(&[("contentKind", "episode")])),
Some(TmdbKind::Tv)
);
assert_eq!(
discovery_kind(&query(&[("contentKind", "movie")])),
Some(TmdbKind::Movie)
);
assert_eq!(discovery_kind(&query(&[("contentKind", "audiobook")])), None);
assert_eq!(discovery_kind(&query(&[])), None);
}
#[test]
fn builds_the_catalog_endpoints() {
// Popular / top-rated route through /discover (so the gate applies);
// only the sort and the vote floor differ.
let pop = build_path_and_query(DiscoveryMode::Popular, TmdbKind::Movie, false, &[], 1);
assert!(pop.starts_with("discover/movie?"), "{pop}");
assert!(pop.contains("sort_by=popularity.desc"), "{pop}");
assert!(pop.contains("page=1"), "{pop}");
let paged = build_path_and_query(DiscoveryMode::Popular, TmdbKind::Tv, false, &[], 2);
assert!(paged.starts_with("discover/tv?"), "{paged}");
assert!(paged.contains("page=2"), "{paged}");
let top = build_path_and_query(DiscoveryMode::TopRated, TmdbKind::Movie, false, &[], 1);
assert!(top.contains("sort_by=vote_average.desc"), "{top}");
assert!(top.contains("vote_count.gte=200"), "{top}");
// Trending keeps the real endpoint — see the dedicated test.
assert_eq!(
build_path_and_query(DiscoveryMode::Trending, TmdbKind::Tv, false, &[], 1),
"trending/tv/week?page=1"
);
}
#[test]
fn anime_switches_to_the_keyword_filtered_discover_endpoint() {
let p = build_path_and_query(DiscoveryMode::Popular, TmdbKind::Tv, true, &[], 1);
assert!(p.starts_with("discover/tv?"), "{p}");
assert!(p.contains("with_keywords=210024"), "{p}");
assert!(p.contains("sort_by=popularity.desc"), "{p}");
let top = build_path_and_query(DiscoveryMode::TopRated, TmdbKind::Movie, true, &[], 1);
assert!(top.contains("sort_by=vote_average.desc"), "{top}");
// Sorting by rating puts a 1-vote 10.0 on top, so top-rated floors higher.
assert!(top.contains("vote_count.gte=200"), "{top}");
}
/// The gate is not anime-specific. Three hentai titles were measured at
/// positions 4, 5 and 17 of the **non-anime** `tv/popular` row, which is why
/// the plain popular/top-rated modes were moved off the fixed list endpoints
/// (which cannot carry `without_keywords`) and onto `/discover`.
#[test]
fn non_anime_popular_and_top_rated_also_gate() {
for mode in [DiscoveryMode::Popular, DiscoveryMode::TopRated] {
let p = build_path_and_query(mode, TmdbKind::Tv, false, &[], 1);
assert!(p.starts_with("discover/tv?"), "{p}");
assert!(!p.contains("with_keywords="), "no anime narrowing here: {p}");
assert!(p.contains("without_keywords="), "{p}");
assert!(p.contains("vote_count.gte="), "{p}");
}
}
/// Trending is the documented exception: `/trending` is a TMDB-computed
/// signal with no `/discover` sort equivalent, so the non-anime trending row
/// keeps the list endpoint and cannot carry the gate.
#[test]
fn non_anime_trending_keeps_the_real_trending_endpoint() {
let p = build_path_and_query(DiscoveryMode::Trending, TmdbKind::Tv, false, &[], 1);
assert_eq!(p, "trending/tv/week?page=1");
// …but the anime variant has no keyword-filtered trending at all, so it
// still routes through /discover and stays gated.
let a = build_path_and_query(DiscoveryMode::Trending, TmdbKind::Tv, true, &[], 1);
assert!(a.starts_with("discover/tv?"), "{a}");
assert!(a.contains("without_keywords="), "{a}");
}
/// A genre-scoped row narrows **upstream** via `with_genres`, so the row is a
/// full page of that genre rather than a page filtered down to a handful.
#[test]
fn a_genre_row_narrows_upstream() {
let p = build_path_and_query(DiscoveryMode::Popular, TmdbKind::Tv, false, &[10759], 1);
assert!(p.contains("with_genres=10759"), "{p}");
// Multi-valued = OR, so pipe-joined (comma would mean AND to TMDB).
let multi = build_path_and_query(DiscoveryMode::Popular, TmdbKind::Tv, false, &[16, 35], 1);
assert!(multi.contains("with_genres=16%7C35"), "{multi}");
// No genre filter ⇒ the parameter is absent, not empty.
let none = build_path_and_query(DiscoveryMode::Popular, TmdbKind::Tv, false, &[], 1);
assert!(!none.contains("with_genres"), "{none}");
}
/// A genre row can't use `/trending` (it takes no `with_genres`), so it falls
/// back to the filterable `/discover` rather than returning an unfiltered
/// trending page under a genre heading.
#[test]
fn a_genre_scoped_trending_row_falls_back_to_discover() {
let p = build_path_and_query(DiscoveryMode::Trending, TmdbKind::Tv, false, &[10759], 1);
assert!(p.starts_with("discover/tv?"), "{p}");
assert!(p.contains("with_genres=10759"), "{p}");
}
/// Slugs resolve to TMDB ids through the id → name table, folded on both
/// sides so the DSL never has to spell a space or an ampersand.
#[test]
fn genre_slugs_resolve_to_tmdb_ids() {
let names = HashMap::from([
(16u32, "Animation".to_string()),
(10759, "Action & Adventure".to_string()),
(10765, "Sci-Fi & Fantasy".to_string()),
]);
let want = |s: &str| with_genres_ids(&[genre_fold(s)], &names);
assert_eq!(want("action-adventure"), vec![10759]);
assert_eq!(want("Action & Adventure"), vec![10759]);
assert_eq!(want("scififantasy"), vec![10765]);
assert_eq!(want("animation"), vec![16]);
assert!(want("horror").is_empty(), "unknown slug resolves to nothing");
}
/// The fold must stay identical to the one both re-validating tiers use, or
/// a row's own results get dropped downstream.
#[test]
fn genre_fold_matches_the_consumer_side_rule() {
assert_eq!(genre_fold("Action & Adventure"), "actionadventure");
assert_eq!(genre_fold("action-adventure"), "actionadventure");
assert_eq!(genre_fold("Sci-Fi & Fantasy"), "scififantasy");
assert_eq!(genre_fold("Science Fiction"), "sciencefiction");
assert_eq!(genre_fold(" "), "");
}
/// The adult-content gate on anime rows, and *why* it is two filters.
///
/// `include_adult=false` is not enough on its own — measured live, every
/// offending title returns `adult: false` (TMDB reserves that flag for
/// pornography), so the keyword exclusion carries the real load and the vote
/// floor sweeps up the erotic-tagged residue behind it. Dropping either one
/// puts hentai back on the home page.
#[test]
fn anime_rows_gate_adult_content_three_ways() {
for mode in [
DiscoveryMode::Popular,
DiscoveryMode::Trending,
DiscoveryMode::TopRated,
] {
for kind in [TmdbKind::Tv, TmdbKind::Movie] {
let p = build_path_and_query(mode, kind, true, &[], 1);
assert!(p.contains("include_adult=false"), "{p}");
assert!(
p.contains(&format!("without_keywords={DISCOVERY_EXCLUDED_KEYWORDS}")),
"{p}"
);
assert!(p.contains("vote_count.gte="), "{p}");
}
}
}
fn genre_table() -> HashMap<u32, String> {
HashMap::from([(16, "Animation".to_string()), (18, "Drama".to_string())])
}
fn hit() -> TmdbHit {
TmdbHit {
akas: Vec::new(),
tmdbid: 95479,
title: "Frieren: Beyond Journey's End".to_string(),
original_title: Some("葬送のフリーレン".to_string()),
original_language: Some("ja".to_string()),
overview: Some("A mage reflects on her long life.".to_string()),
year: Some(2023),
poster_path: Some("/poster.jpg".to_string()),
genre_ids: vec![16],
alt_titles: Vec::new(),
posters: Vec::new(),
}
}
/// A list hit must carry everything a renderable card needs, so the branch
/// costs one TMDB call per page and no by-id follow-ups.
#[test]
fn a_list_hit_becomes_a_complete_card() {
let c = card_from_hit(&hit(), TmdbKind::Tv, &genre_table());
assert_eq!(c.tmdbid, Some(95479));
assert!(c.overview.is_some(), "overview is what to_record requires");
assert!(c.poster_path.is_some(), "poster is what to_record requires");
assert_eq!(c.year, Some(2023));
assert_eq!(c.content_kind(), "series");
// Genres come from the id table — the category axis a card used to lack.
assert_eq!(c.genres, vec!["Animation".to_string()]);
// The id bag beyond tmdbid needs external_ids — deliberately absent.
assert_eq!(c.tvdb_id, None);
assert_eq!(c.imdb_id, None);
}
/// The locator CID is derived from `record_id`, so a discovery card and a
/// by-id card for the same work must address identically or the two tiers
/// would publish two records for one work.
#[test]
fn record_id_matches_the_by_id_path() {
assert_eq!(card_from_hit(&hit(), TmdbKind::Tv, &genre_table()).record_id(), "tmdb:tv:95479");
assert_eq!(
card_from_hit(&hit(), TmdbKind::Movie, &genre_table()).record_id(),
"tmdb:movie:95479"
);
}
/// The whole point of the branch: `contentKind` on the emitted record is the
/// *card's* kind, never an echo of the query filter. A row asking
/// `contentKind:episode` must still produce `series` here — which is exactly
/// why a client's row query has to ask for `series` (the mismatch is then
/// caught by `record_matches` downstream rather than silently mislabelling).
/// **The invariant this whole branch rests on.** The mode/anime markers are
/// stamped onto the record by `Card::to_record`'s trailing filter echo, not
/// by anything here — and both the gateway dispatcher and meta-search
/// re-apply the query's filters to every record, where a *missing* field
/// fails the match. So if that echo is ever "tidied up", every discovery
/// card is silently dropped in transit and the wall goes blank with no
/// error. This test fails that day.
#[test]
fn a_discovery_card_survives_the_consumer_re_filter() {
let q = query(&[
("popular", "true"),
("contentKind", "series"),
("fileType", "card"),
("anime", "true"),
]);
let card = card_from_hit(&hit(), TmdbKind::Tv, &genre_table());
let tmdb = crate::tmdb_client::TmdbClient::new("token".to_string());
let rec = card.to_record(&tmdb, &q.filters).expect("displayable card emits");
assert_eq!(rec.fields.get("popular").map(String::as_str), Some("true"));
assert_eq!(rec.fields.get("anime").map(String::as_str), Some("true"));
assert_eq!(rec.fields.get("fileType").map(String::as_str), Some("card"));
assert_eq!(
rec.fields.get("contentKind").map(String::as_str),
Some("series"),
"the card's own kind must win over the query filter"
);
assert!(rec.fields.contains_key("description/eng"));
assert!(rec.fields.contains_key("poster_url"));
// Genres ride as a key-set, never the legacy comma-joined value
// (METADATA_KEYS §14.12 — new writers must not add csv-set fields).
assert_eq!(rec.fields.get("genres/Animation").map(String::as_str), Some("true"));
assert!(!rec.fields.contains_key("genres"), "must not write the legacy csv-set");
// The thin id bag: resolved lazily on the click-time by-id path.
assert!(!rec.fields.contains_key("tvdbid"));
assert!(!rec.fields.contains_key("imdbid"));
assert!(!rec.fields.contains_key("seasonCount"));
assert!(
meta_feeder_sdk::query_eval::record_matches(&rec.fields, &q),
"a discovery card must survive the same filter re-application the \
dispatcher and meta-search perform"
);
}
/// A hit TMDB has not written up yet is declined at projection — which is
/// why the page walk counts displayable cards, not raw hits.
#[test]
fn an_incomplete_hit_is_not_displayable() {
let mut no_overview = hit();
no_overview.overview = None;
assert!(!card_from_hit(&no_overview, TmdbKind::Tv, &genre_table()).is_displayable());
let mut no_poster = hit();
no_poster.poster_path = None;
assert!(!card_from_hit(&no_poster, TmdbKind::Tv, &genre_table()).is_displayable());
assert!(card_from_hit(&hit(), TmdbKind::Tv, &genre_table()).is_displayable());
}
#[test]
fn content_kind_comes_from_the_card_not_the_filter() {
let tv = card_from_hit(&hit(), TmdbKind::Tv, &genre_table());
assert_eq!(tv.content_kind(), "series");
let movie = card_from_hit(&hit(), TmdbKind::Movie, &genre_table());
assert_eq!(movie.content_kind(), "movie");
}
}
+1
View File
@@ -12,6 +12,7 @@
pub mod card;
pub mod consts;
pub mod discovery;
pub mod resolve;
pub mod tmdb;
pub mod tmdb_budget;
+142 -7
View File
@@ -9,6 +9,7 @@
//! The redb table names are unchanged from the indexer feeder's cache, so a
//! warm cache carries straight over.
use std::collections::HashMap;
use std::sync::Arc;
use std::time::Duration;
@@ -16,10 +17,13 @@ use meta_feeder_sdk::cache::MidhashCache;
use tracing::debug;
use crate::card::{Card, CardSource};
use crate::consts::{CACHED_PRINCIPAL_DEPTH, TMDB_WAIT_DEADLINE_SECS};
use crate::consts::{
CACHED_PRINCIPAL_DEPTH, TMDB_DISCOVERY_WAIT_DEADLINE_SECS, TMDB_WAIT_DEADLINE_SECS,
};
use crate::tmdb_budget::{Lease, TmdbBudget};
use crate::tmdb_client::{
principal_top_n, TmdbCall, TmdbClient, TmdbExternalIds, TmdbHit, TmdbKind, TmdbTvDetails,
principal_top_n, title_names, TmdbAltTitles, TmdbCall, TmdbClient, TmdbExternalIds, TmdbHit,
TmdbKind, TmdbTvDetails,
};
/// TMDB resolution with a persistent cache in front and a token budget behind.
@@ -28,6 +32,14 @@ pub(crate) struct Resolver {
pub(crate) client: Arc<TmdbClient>,
pub(crate) cache: MidhashCache,
pub(crate) budget: Arc<TmdbBudget>,
/// `genre id → name` per kind, held for the process lifetime.
///
/// In memory rather than in the redb cache on purpose: the table is 16 TV /
/// 19 movie entries and effectively static, so a process-local map costs one
/// call per kind per boot and needs no invalidation story — whereas the redb
/// tables are permanent-hit and shared byte-for-byte with the indexer
/// feeder's, which this is not part of.
genre_names: Arc<std::sync::RwLock<HashMap<&'static str, Arc<HashMap<u32, String>>>>>,
}
impl Resolver {
@@ -36,20 +48,68 @@ impl Resolver {
client,
cache,
budget,
genre_names: Arc::new(std::sync::RwLock::new(HashMap::new())),
}
}
/// The `genre id → name` table for `kind`, fetched once and then reused.
///
/// Needed only by the **list** paths (discovery/search), where TMDB gives
/// bare `genre_ids`; a details fetch returns named genre objects directly.
/// Best-effort: a failed or empty fetch yields an empty map (cards simply
/// carry no genres) and is **not** cached, so the next row retries.
pub(crate) async fn genre_names(&self, kind: TmdbKind) -> Arc<HashMap<u32, String>> {
let key = match kind {
TmdbKind::Tv => "tv",
TmdbKind::Movie => "movie",
};
if let Some(m) = self.genre_names.read().ok().and_then(|g| g.get(key).cloned()) {
return m;
}
let fetched: HashMap<u32, String> = self
.budgeted_with_deadline(
TMDB_DISCOVERY_WAIT_DEADLINE_SECS,
self.client.genre_list(kind),
|gs| {
gs.into_iter()
.filter(|g| !g.name.trim().is_empty())
.map(|g| (g.id, g.name))
.collect()
},
)
.await
.unwrap_or_default();
let arc = Arc::new(fetched);
if !arc.is_empty() {
if let Ok(mut g) = self.genre_names.write() {
g.insert(key, arc.clone());
}
}
arc
}
/// Acquire a budget token, run `call`, map a hit through `on_hit`. A 429
/// pauses the shared bucket globally rather than being retried here.
async fn budgeted<T, R>(
&self,
call: impl std::future::Future<Output = TmdbCall<T>>,
on_hit: impl FnOnce(T) -> R,
) -> Option<R> {
self.budgeted_with_deadline(TMDB_WAIT_DEADLINE_SECS, call, on_hit)
.await
}
/// [`Self::budgeted`] with a caller-chosen permit deadline, for callers whose
/// cost of waiting differs from a user-facing lookup's — see
/// [`TMDB_DISCOVERY_WAIT_DEADLINE_SECS`].
async fn budgeted_with_deadline<T, R>(
&self,
deadline_secs: u64,
call: impl std::future::Future<Output = TmdbCall<T>>,
on_hit: impl FnOnce(T) -> R,
) -> Option<R> {
if matches!(
self.budget
.acquire(Duration::from_secs(TMDB_WAIT_DEADLINE_SECS))
.await,
self.budget.acquire(Duration::from_secs(deadline_secs)).await,
Lease::DeadlineExceeded
) {
return None;
@@ -90,7 +150,11 @@ impl Resolver {
if let Ok(Some(json)) = self.cache.get_tmdb_moviedetails(&key) {
if json != "null" {
if let Ok(hit) = serde_json::from_str::<TmdbHit>(&json) {
return Some(hit);
// Self-heal: AKAs cached without their markets can't be
// filed under `titles/{lang3}/{name}` — refetch once.
if hit.alt_titles.is_empty() || !hit.akas.is_empty() {
return Some(hit);
}
}
}
}
@@ -124,6 +188,29 @@ impl Resolver {
.await
}
/// One page of a TMDB **catalog list** (popular / trending / top-rated /
/// `discover`) — the keyword-less browse primitive behind
/// [`crate::discovery`]. `None` on a budget timeout, a 429, or a miss, which
/// the caller treats as "stop walking pages and keep what you have".
///
/// Deliberately **uncached**, unlike every other method here: a catalog list
/// is the one TMDB response that is *supposed* to change under you, and the
/// redb tables in front of the others are permanent-hit caches with no TTL.
/// See the module doc of [`crate::discovery`] for why the repeat cost is
/// already absorbed a layer up.
pub(crate) async fn discovery_page(
&self,
kind: TmdbKind,
path_and_query: &str,
) -> Option<Vec<TmdbHit>> {
self.budgeted_with_deadline(
TMDB_DISCOVERY_WAIT_DEADLINE_SECS,
self.client.discovery_list(kind, path_and_query),
|hits| hits,
)
.await
}
/// Cached principal `search/multi`: free text → up to `n` confident
/// `(tmdbid, kind)` candidates, most popular first.
///
@@ -183,7 +270,18 @@ impl Resolver {
.as_deref()
.and_then(|d| d.get(0..4))
.and_then(|y| y.parse::<u16>().ok());
let names = title_names(
&details.name,
details.original_name.as_deref(),
details.original_language.as_deref(),
&details
.alternative_titles
.as_ref()
.map(TmdbAltTitles::entries)
.unwrap_or_default(),
);
Some(Card {
names,
source: CardSource::Tmdb,
kind,
source_id: format!("tv:{tmdbid}"),
@@ -194,19 +292,43 @@ impl Resolver {
guard_titles: Arc::new(dedup_titles(details.guard_titles())),
overview: details.overview.clone().filter(|s| !s.is_empty()),
poster_path: details.poster_path.clone().filter(|s| !s.is_empty()),
genres: genre_names_of(&details.genres),
year,
seasons: Card::clamp_seasons(details.number_of_seasons),
season_summaries: Arc::new(details.seasons.clone()),
posters: Arc::new(
details
.images
.as_ref()
.map(|i| i.posters.clone())
.unwrap_or_default(),
),
})
}
TmdbKind::Movie => {
let hit = self.movie_hit(tmdbid).await?;
// The movie path round-trips through `TmdbHit` (that is what the
// details cache stores), which keeps genres as bare ids — so
// unlike the TV branch above it needs the id → name table.
let gmap = self.genre_names(TmdbKind::Movie).await;
let genres: Vec<String> = hit
.genre_ids
.iter()
.filter_map(|id| gmap.get(id).cloned())
.collect();
let mut titles = vec![hit.title.clone()];
if let Some(o) = &hit.original_title {
titles.push(o.clone());
}
titles.extend(hit.alt_titles.iter().cloned());
let names = title_names(
&hit.title,
hit.original_title.as_deref(),
hit.original_language.as_deref(),
&hit.akas,
);
Some(Card {
names,
source: CardSource::Tmdb,
kind,
source_id: format!("movie:{tmdbid}"),
@@ -217,9 +339,11 @@ impl Resolver {
guard_titles: Arc::new(dedup_titles(titles)),
overview: hit.overview.filter(|s| !s.is_empty()),
poster_path: hit.poster_path.filter(|s| !s.is_empty()),
genres,
year: hit.year,
seasons: 0,
season_summaries: Arc::new(Vec::new()),
posters: Arc::new(hit.posters),
})
}
}
@@ -244,8 +368,19 @@ impl Resolver {
}
}
/// Genre display names off a details payload's `{id, name}` array, trimmed and
/// de-duplicated, order preserved.
pub(crate) fn genre_names_of(genres: &[crate::tmdb_client::TmdbGenre]) -> Vec<String> {
let mut seen = std::collections::HashSet::new();
genres
.iter()
.map(|g| g.name.trim().to_string())
.filter(|n| !n.is_empty() && seen.insert(n.to_lowercase()))
.collect()
}
/// Case-insensitive de-dup preserving first-seen order.
fn dedup_titles(titles: Vec<String>) -> Vec<String> {
pub(crate) fn dedup_titles(titles: Vec<String>) -> Vec<String> {
let mut seen = std::collections::HashSet::new();
titles
.into_iter()
+49 -4
View File
@@ -22,7 +22,7 @@ use meta_feeder_sdk::types::{DiscoveryRecord, GatewayError, Hash, PluginHealth};
use tracing::warn;
use crate::card::{split_record_id, Card, CardSource};
use crate::consts::DEFAULT_CARD_TOP_N;
use crate::consts::{DEFAULT_CARD_DISCOVERY_N, DEFAULT_CARD_TOP_N};
use crate::resolve::Resolver;
use crate::tmdb_budget::{TmdbBudget, DEFAULT_TMDB_BURST, DEFAULT_TMDB_REFILL_PER_SEC};
use crate::tmdb_client::TmdbClient;
@@ -39,6 +39,8 @@ pub struct TmdbConfig {
pub tmdb_burst: Option<f64>,
#[serde(default)]
pub card_top_n: Option<usize>,
#[serde(default)]
pub card_discovery_n: Option<usize>,
}
impl TmdbConfig {
@@ -51,6 +53,9 @@ impl TmdbConfig {
.and_then(|v| v.parse().ok()),
tmdb_burst: std::env::var("TMDB_BURST").ok().and_then(|v| v.parse().ok()),
card_top_n: std::env::var("CARD_TOP_N").ok().and_then(|v| v.parse().ok()),
card_discovery_n: std::env::var("CARD_DISCOVERY_N")
.ok()
.and_then(|v| v.parse().ok()),
}
}
}
@@ -62,6 +67,8 @@ pub struct TmdbCardPlugin {
/// `depends_on` is satisfied; queries just return nothing.
resolver: Option<Resolver>,
top_n: usize,
/// Cards per keyword-less catalog row (see [`DEFAULT_CARD_DISCOVERY_N`]).
discovery_n: usize,
/// Test hook: `(api_base, image_base)` overriding TMDB's real endpoints, so
/// the contract test can point the client at a wiremock server.
api_bases: Option<(String, String)>,
@@ -79,6 +86,7 @@ impl TmdbCardPlugin {
config: TmdbConfig::from_env(),
resolver: None,
top_n: DEFAULT_CARD_TOP_N,
discovery_n: DEFAULT_CARD_DISCOVERY_N,
api_bases: None,
}
}
@@ -111,6 +119,9 @@ impl TmdbCardPlugin {
if file_cfg.card_top_n.is_some() {
self.config.card_top_n = file_cfg.card_top_n;
}
if file_cfg.card_discovery_n.is_some() {
self.config.card_discovery_n = file_cfg.card_discovery_n;
}
}
}
}
@@ -142,6 +153,11 @@ impl FeederPlugin for TmdbCardPlugin {
self.load_config(cache_dir);
let cache: MidhashCache = meta_feeder_sdk::common::open_midhash_cache(cache_dir, "tmdb")?;
self.top_n = self.config.card_top_n.filter(|n| *n > 0).unwrap_or(DEFAULT_CARD_TOP_N);
self.discovery_n = self
.config
.card_discovery_n
.filter(|n| *n > 0)
.unwrap_or(DEFAULT_CARD_DISCOVERY_N);
let token = self.config.tmdb_token.trim().to_string();
if token.is_empty() {
@@ -189,15 +205,33 @@ impl FeederPlugin for TmdbCardPlugin {
self.served_file_types(),
self.served_content_kinds(),
) {
// One misconfiguration lands here often enough to deserve a name.
// A browse row written against the *indexer* feeder's dialect says
// `contentKind:episode`, which is not in `served_content_kinds` — so
// the row is rejected here, before TMDB is touched, and the operator
// sees an empty wall with no error anywhere. A card describes a WORK,
// and a work is a `series`, never an `episode`.
if crate::discovery::is_discovery_query(query) {
warn!(
target: "meta-share::gateway",
upstream = "tmdb",
content_kind = ?query.filters.get("contentKind"),
"discovery row rejected by the served-kinds gate: a card is a \
WORK — use contentKind:series for TV rows (not episode/tv/tvshow)"
);
}
return Ok(Vec::new());
}
let Some(resolver) = self.resolver() else {
return Ok(Vec::new()); // soft-skipped (no token)
};
// An explicit `tmdbid:` filter resolves exactly one card — the direct
// "give me this work" lookup a deep link uses. Otherwise the free text
// goes through the principal search.
// Three shapes, in precedence order:
// 1. an explicit `tmdbid:` filter — the direct "give me this work"
// lookup a deep link (or the detail page) uses;
// 2. a keyword-less catalog query (`popular:true contentKind:series`)
// — a client's home row, answered from TMDB's catalog endpoints;
// 3. free text — the principal search.
let cards: Vec<Card> = if let Some(id) = query
.filters
.get("tmdbid")
@@ -206,6 +240,9 @@ impl FeederPlugin for TmdbCardPlugin {
{
let kind = kind_hint(query, resolver, id).await;
resolver.card_by_id(id, kind).await.into_iter().collect()
} else if crate::discovery::is_discovery_query(query) {
let n = self.discovery_n.min(max_results.max(1));
crate::discovery::cards_for_discovery(resolver, query, n).await
} else {
let free_text = query.free_text.trim();
if free_text.is_empty() {
@@ -284,6 +321,13 @@ impl FeederPlugin for TmdbCardPlugin {
Blank keeps the built-in default (10). Unlike the old anchor \
top-N this costs no indexer requests — only TMDB.",
),
F::number("card_discovery_n", "Cards per discovery row").with_help(
"How many cards a keyword-less browse row (popular / trending / \
top-rated) returns. Blank keeps the built-in default (20 — \
exactly one TMDB catalog page, so a row costs ONE request). \
Separate from 'Cards per search': a row is a browse surface, \
not an answer to a question.",
),
],
}
}
@@ -294,6 +338,7 @@ impl FeederPlugin for TmdbCardPlugin {
"tmdb_rate_per_sec": self.config.tmdb_rate_per_sec,
"tmdb_burst": self.config.tmdb_burst,
"card_top_n": self.config.card_top_n,
"card_discovery_n": self.config.card_discovery_n,
})
}
}
+259 -14
View File
@@ -190,7 +190,7 @@ impl TmdbClient {
/// any failure returns `None` so validation degrades to "accept".
pub(crate) async fn tv_details(&self, tmdbid: u64) -> TmdbCall<TmdbTvDetails> {
let url = format!(
"{}/tv/{}?append_to_response=alternative_titles",
"{}/tv/{}?append_to_response=alternative_titles,images",
self.api_base.trim_end_matches('/'),
tmdbid,
);
@@ -253,7 +253,7 @@ impl TmdbClient {
/// [`tv_details`].
pub(crate) async fn movie_details(&self, tmdbid: u64) -> TmdbCall<TmdbMovieDetails> {
let url = format!(
"{}/movie/{}?append_to_response=alternative_titles",
"{}/movie/{}?append_to_response=alternative_titles,images",
self.api_base.trim_end_matches('/'),
tmdbid
);
@@ -309,6 +309,27 @@ impl TmdbClient {
}
}
/// `GET /3/genre/{tv,movie}/list` — the id → name table for the `genre_ids`
/// a discovery/search list hit carries.
///
/// TMDB returns genres as bare ids on list responses and as `{id, name}`
/// objects only on a details fetch, so the *browse* path has no way to name
/// a genre without this table. It is tiny (16 TV / 19 movie entries) and
/// effectively static, which is why the caller holds it for the process
/// lifetime instead of paying it per row.
pub(crate) async fn genre_list(&self, kind: TmdbKind) -> TmdbCall<Vec<TmdbGenre>> {
let seg = match kind {
TmdbKind::Movie => "movie",
TmdbKind::Tv => "tv",
};
let url = format!("{}/genre/{seg}/list", self.api_base.trim_end_matches('/'));
match self.get_json::<TmdbGenreList>(&url, "genre_list").await {
TmdbCall::Hit(r) => TmdbCall::Hit(r.genres),
TmdbCall::Miss => TmdbCall::Miss,
TmdbCall::RateLimited(d) => TmdbCall::RateLimited(d),
}
}
/// Shared `GET <url>` → JSON helper with the standard bearer auth, search
/// timeout, 429→`RateLimited`, non-2xx/transient→`Miss` contract. Factored
/// out of [`tv_details`]/[`external_ids`]/[`movie_details`] (identical
@@ -393,6 +414,21 @@ pub(crate) enum TmdbKind {
Tv,
}
/// One `{id, name}` genre entry, as returned by `genre/{kind}/list` and by the
/// details endpoints' `genres` array.
#[derive(Clone, Debug, serde::Deserialize, serde::Serialize)]
pub(crate) struct TmdbGenre {
pub(crate) id: u32,
#[serde(default)]
pub(crate) name: String,
}
#[derive(serde::Deserialize)]
struct TmdbGenreList {
#[serde(default)]
genres: Vec<TmdbGenre>,
}
#[derive(serde::Deserialize)]
pub(crate) struct TmdbSearchResponse {
#[serde(default)]
@@ -421,8 +457,8 @@ pub(crate) struct TmdbSearchItem {
#[serde(default)]
pub(crate) genre_ids: Vec<u32>,
/// ISO 639-1 (2-letter) language the title was originally produced in.
/// Mapped to `lang3` and used to file `original_title` under
/// `titles/{lang3}` (METADATA_KEYS.md §3).
/// Mapped to `lang3` and used to file `original_title` as a
/// `titles/{lang3}/{name}` member (METADATA_KEYS.md §3).
#[serde(default)]
pub(crate) original_language: Option<String>,
}
@@ -457,6 +493,8 @@ impl TmdbSearchItem {
genre_ids: self.genre_ids,
// search-list items carry no AKAs (a details/append fetch does)
alt_titles: Vec::new(),
akas: Vec::new(),
posters: Vec::new(),
}
}
}
@@ -464,8 +502,8 @@ impl TmdbSearchItem {
/// Map an ISO 639-1 (2-letter) code to its ISO 639-3 (`lang3`) equivalent,
/// matching the store's convention (`eng`, `jpn`, `fra`; the 639-2/T variant
/// where B/T differ, e.g. `deu` not `ger`). Covers the languages TMDB
/// commonly returns; unknown codes return `None` so the caller skips the
/// `titles/{lang3}` write rather than persisting a non-`lang3` key.
/// commonly returns; unknown codes return `None` so the caller files the name under `und`
/// rather than persisting a non-`lang3` key.
pub(crate) fn iso639_1_to_3(code: &str) -> Option<&'static str> {
Some(match code.to_ascii_lowercase().as_str() {
"en" => "eng",
@@ -505,6 +543,113 @@ pub(crate) fn iso639_1_to_3(code: &str) -> Option<&'static str> {
})
}
/// The language TMDB's `name`/`title` is written in. Every request goes out
/// without a `language` param, so TMDB answers in its default, en-US.
pub(crate) const METADATA_LANG3: &str = "eng";
/// A title as a `titles/{lang3}/{name}` member name (METADATA_KEYS.md): trimmed,
/// internal whitespace collapsed to one space, case/diacritics/`%` verbatim.
/// The `/` → `∕` mapping is the key's job ([`title_member_key`]), not the name's.
pub(crate) fn normalize_title_name(name: &str) -> String {
name.split_whitespace().collect::<Vec<_>>().join(" ")
}
/// The `titles/{lang3}/{name}` key-set member for `name`, or `None` when the
/// name normalises to empty. `/` is the key-set separator, so a `/` inside a
/// name (`Fate/Zero`) is written as U+2215 `∕` (METADATA_KEYS.md).
///
/// CROSS-BINARY CONTRACT: the indexer feeder's `tmdb.rs::title_member_key`
/// writes the same shape; two peers must produce the same key for one name.
pub(crate) fn title_member_key(lang3: &str, name: &str) -> Option<String> {
let name = normalize_title_name(name);
if name.is_empty() {
return None;
}
Some(format!("titles/{lang3}/{}", name.replace('/', "\u{2215}")))
}
/// The language a TMDB AKA market (`iso_3166_1`) clearly implies, or `und` for
/// a multilingual or unknown market (METADATA_KEYS.md `titles/{lang3}/{name}`:
/// map what is unambiguous, never guess). `CA`, `CH`, `BE`, `IN`, `HK`, `SG`,
/// empty, … are `und` on purpose.
pub(crate) fn market_lang3(market: &str) -> &'static str {
match market.trim().to_ascii_uppercase().as_str() {
"US" | "GB" | "AU" | "NZ" | "IE" => "eng",
"FR" => "fra",
"DE" | "AT" => "deu",
"ES" | "MX" | "AR" | "CO" | "CL" | "PE" | "VE" | "UY" => "spa",
"BR" | "PT" => "por",
"IT" => "ita",
"JP" => "jpn",
"KR" => "kor",
"CN" | "TW" => "zho",
"RU" => "rus",
"UA" => "ukr",
"PL" => "pol",
"NL" => "nld",
"SE" => "swe",
"NO" => "nor",
"DK" => "dan",
"FI" => "fin",
"TR" => "tur",
"GR" => "ell",
"HU" => "hun",
"CZ" => "ces",
"SK" => "slk",
"RO" => "ron",
"BG" => "bul",
"RS" => "srp",
"HR" => "hrv",
"IL" => "heb",
"IR" => "fas",
"TH" => "tha",
"VN" => "vie",
"ID" => "ind",
_ => "und",
}
}
/// Every name of a work as `(lang3, name)` — what the card files under
/// `titles/{lang3}/{name}`:
///
/// - `original` under its `original_language` (`und` when unmapped/absent);
/// - `title` under [`METADATA_LANG3`], **unless it equals the original** — TMDB
/// falls back to the original name when it has no en-US translation, so `3%`
/// is filed once, as `titles/por/3%`, not also as an English name;
/// - every AKA under the language its market implies ([`market_lang3`]).
///
/// Names are normalised ([`normalize_title_name`]); blanks, TMDB's
/// `(untitled)` placeholder and exact `(lang3, name)` repeats are dropped.
pub(crate) fn title_names(
title: &str,
original: Option<&str>,
original_language: Option<&str>,
akas: &[TmdbAltTitle],
) -> Vec<(&'static str, String)> {
let mut out: Vec<(&'static str, String)> = Vec::new();
let mut push = |lang3: &'static str, name: &str| {
let name = normalize_title_name(name);
if !name.is_empty() && !out.iter().any(|(l, n)| *l == lang3 && *n == name) {
out.push((lang3, name));
}
};
let original = original.map(normalize_title_name).filter(|o| !o.is_empty());
if let Some(o) = &original {
let lang3 = original_language
.and_then(|l| iso639_1_to_3(l.trim()))
.unwrap_or("und");
push(lang3, o);
}
let title = normalize_title_name(title);
if title != "(untitled)" && original.as_deref() != Some(title.as_str()) {
push(METADATA_LANG3, &title);
}
for aka in akas {
push(market_lang3(&aka.iso_3166_1), &aka.title);
}
out
}
// `Clone` is required so the single-flight `Shared` future (whose `Output`
// must be `Clone`) can hand the same hit to every coalesced caller; serde so
// hits persist in the redb TMDB-search cache.
@@ -530,17 +675,53 @@ pub(crate) struct TmdbHit {
/// Empty for hits built from a search list (which carries no AKAs).
#[serde(default)]
pub(crate) alt_titles: Vec<String>,
/// The same AKAs **with their market**, which the card needs to file each
/// under `titles/{lang3}/{name}` ([`title_names`]). Empty on a movie-details
/// cache entry written before it existed — `Resolver::movie_hit` refetches
/// those once.
#[serde(default)]
pub(crate) akas: Vec<TmdbAltTitle>,
/// Poster candidates from the details `images` append, filed by the card as
/// `posters/{lang3}/{cid}` (METADATA_KEYS.md §6). Empty for list hits and for
/// a movie-details cache entry written before it existed — deliberately NOT a
/// self-heal trigger in `Resolver::movie_hit`: no backfill, no TMDB budget.
#[serde(default)]
pub(crate) posters: Vec<TmdbImage>,
}
impl TmdbHit {
/// TMDB `original_language` (ISO 639-1) mapped to the store's `lang3`
/// (ISO 639-3), or `None` when the language is outside the common set or
/// absent. Used to file `original_title` under `titles/{lang3}` (§3).
/// absent. Used to file `original_title` under `titles/{lang3}/{name}` (§3).
pub(crate) fn original_lang3(&self) -> Option<&'static str> {
iso639_1_to_3(self.original_language.as_deref()?.trim())
}
}
/// One poster from a details payload's `images` append — a candidate for the
/// card's `posters/{lang3}/{cid}` key-set ([`crate::card::poster_member_keys`]).
#[derive(Debug, Clone, Default, PartialEq, serde::Serialize, serde::Deserialize)]
pub(crate) struct TmdbImage {
#[serde(default)]
pub(crate) file_path: String,
/// Language of the text printed on the poster (ISO 639-1); `None` = textless.
#[serde(default)]
pub(crate) iso_639_1: Option<String>,
#[serde(default)]
pub(crate) vote_average: f64,
#[serde(default)]
pub(crate) vote_count: u32,
}
/// The `images` object `?append_to_response=images` adds to `GET /3/{tv,movie}/{id}`.
/// Only `posters` is decoded. No `language` param is sent, so TMDB returns the
/// posters of every language.
#[derive(Debug, Clone, Default, serde::Serialize, serde::Deserialize)]
pub(crate) struct TmdbImages {
#[serde(default)]
pub(crate) posters: Vec<TmdbImage>,
}
/// Authoritative TV structure from `GET /3/tv/{id}`, used to bounds-check
/// a title-parsed season/episode. Only the two structural fields are
/// decoded; everything else in the (large) details payload is ignored.
@@ -550,6 +731,10 @@ impl TmdbHit {
pub(crate) struct TmdbTvDetails {
#[serde(default)]
pub(crate) number_of_seasons: u32,
/// Genre objects (`{id, name}`) — present on a details fetch, unlike the
/// bare `genre_ids` a list response carries. Free here: same payload.
#[serde(default)]
pub(crate) genres: Vec<TmdbGenre>,
/// One entry per season TMDB knows about, including season 0
/// ("Specials"). `episode_count` is the released-episode total.
#[serde(default)]
@@ -577,6 +762,10 @@ pub(crate) struct TmdbTvDetails {
/// fetch self-heals with one refetch (same pattern as the display fields).
#[serde(default)]
pub(crate) alternative_titles: Option<TmdbAltTitles>,
/// Poster candidates from `?append_to_response=images`. `None` on a cache
/// entry written before the append — NOT a self-heal trigger (no backfill).
#[serde(default)]
pub(crate) images: Option<TmdbImages>,
}
impl TmdbTvDetails {
@@ -590,8 +779,15 @@ impl TmdbTvDetails {
/// True when the AKA field was populated by the current (append-enabled)
/// decoder. A `false` on a cache hit means a pre-AKA entry → the cached
/// fetch refetches once to upgrade it (mirrors [`has_display`]).
///
/// Also `false` for an AKA list decoded before `iso_3166_1` was: the
/// search-term pick (`card::search_aka`) prefers English-market AKAs, and a
/// market-less list would silently keep TMDB's order (`3 Pourcent` over
/// `3 percent`) forever. One refetch per such entry upgrades it.
pub(crate) fn has_akas(&self) -> bool {
self.alternative_titles.is_some()
self.alternative_titles
.as_ref()
.is_some_and(TmdbAltTitles::has_markets)
}
/// The show's canonical + original + AKA titles, for the id-based-job
@@ -632,6 +828,16 @@ impl TmdbTvDetails {
.as_ref()
.map(TmdbAltTitles::all)
.unwrap_or_default(),
akas: self
.alternative_titles
.as_ref()
.map(TmdbAltTitles::entries)
.unwrap_or_default(),
posters: self
.images
.as_ref()
.map(|i| i.posters.clone())
.unwrap_or_default(),
})
}
}
@@ -659,17 +865,42 @@ pub(crate) struct TmdbAltTitles {
pub(crate) struct TmdbAltTitle {
#[serde(default)]
pub(crate) title: String,
/// The market the AKA is used in (`US`, `FR`, …). Empty on a cache entry
/// written before it was decoded — ordering then falls back to TMDB's.
#[serde(default)]
pub(crate) iso_3166_1: String,
}
impl TmdbAltTitles {
/// Was this list decoded with markets? TMDB stamps `iso_3166_1` on every
/// AKA, so a non-empty list where none carries one is a pre-market cache
/// entry. An empty list is complete ("fetched, no AKAs").
pub(crate) fn has_markets(&self) -> bool {
let mut titles = self.results.iter().chain(self.titles.iter()).peekable();
titles.peek().is_none() || titles.any(|t| !t.iso_3166_1.is_empty())
}
/// The non-empty AKA titles across both the TV (`results`) and movie
/// (`titles`) shapes.
/// (`titles`) shapes, **English-market (`US`, `GB`) titles first**, TMDB's
/// order otherwise. The relevance guard only asks "any of these?", so the
/// order is free for it; it matters to the search-term AKA pick
/// (`card::search_aka`), where `3 percent` (US) must beat `3 Pourcent` (FR).
pub(crate) fn all(&self) -> Vec<String> {
self.results
.iter()
.chain(self.titles.iter())
.map(|t| t.title.trim().to_string())
.filter(|t| !t.is_empty())
self.entries().into_iter().map(|t| t.title).collect()
}
/// [`Self::all`] with each AKA's market kept — same order, same filtering
/// (titles trimmed, blanks dropped).
pub(crate) fn entries(&self) -> Vec<TmdbAltTitle> {
let mut titles: Vec<&TmdbAltTitle> = self.results.iter().chain(self.titles.iter()).collect();
titles.sort_by_key(|t| !matches!(t.iso_3166_1.as_str(), "US" | "GB"));
titles
.into_iter()
.map(|t| TmdbAltTitle {
title: t.title.trim().to_string(),
iso_3166_1: t.iso_3166_1.trim().to_string(),
})
.filter(|t| !t.title.is_empty())
.collect()
}
}
@@ -691,6 +922,9 @@ pub(crate) struct TmdbExternalIds {
#[derive(Debug, Clone, serde::Deserialize)]
pub(crate) struct TmdbMovieDetails {
pub(crate) id: u64,
/// See [`TmdbTvDetails::genres`].
#[serde(default)]
pub(crate) genres: Vec<TmdbGenre>,
#[serde(default)]
pub(crate) title: Option<String>,
#[serde(default)]
@@ -705,6 +939,9 @@ pub(crate) struct TmdbMovieDetails {
pub(crate) poster_path: Option<String>,
#[serde(default)]
pub(crate) alternative_titles: Option<TmdbAltTitles>,
/// See [`TmdbTvDetails::images`].
#[serde(default)]
pub(crate) images: Option<TmdbImages>,
}
impl TmdbMovieDetails {
@@ -719,6 +956,12 @@ impl TmdbMovieDetails {
.as_ref()
.map(TmdbAltTitles::all)
.unwrap_or_default();
let akas = self
.alternative_titles
.as_ref()
.map(TmdbAltTitles::entries)
.unwrap_or_default();
let posters = self.images.map(|i| i.posters).unwrap_or_default();
TmdbHit {
tmdbid: self.id,
title: self.title.unwrap_or_else(|| "(untitled)".to_string()),
@@ -729,6 +972,8 @@ impl TmdbMovieDetails {
poster_path: self.poster_path,
genre_ids: Vec::new(),
alt_titles,
akas,
posters,
}
}
}
+10 -1
View File
@@ -64,7 +64,7 @@ fn tv_details_json() -> serde_json::Value {
"poster_path": "/poster.jpg",
"number_of_seasons": 1,
"seasons": [{"season_number": 1, "episode_count": 28}],
"alternative_titles": { "results": [{"title": "Sousou no Frieren"}] }
"alternative_titles": { "results": [{"iso_3166_1": "JP", "title": "Sousou no Frieren"}] }
})
}
@@ -148,6 +148,15 @@ async fn query_returns_a_card_with_both_axes_and_the_id_bag() {
assert_eq!(f["fileType"], "card");
assert_eq!(f["contentKind"], "series");
assert_eq!(f["title"], "Frieren: Beyond Journey's End");
// Every name as the language-nested key-set: original (ja), the en-US
// title, and the JP-market AKA.
for key in [
"titles/jpn/\u{845}\u{9001}\u{306e}\u{30d5}\u{30ea}\u{30fc}\u{30ec}\u{30f3}",
"titles/eng/Frieren: Beyond Journey's End",
"titles/jpn/Sousou no Frieren",
] {
assert_eq!(f.get(key).map(String::as_str), Some("true"), "{key}");
}
// The id bag — phase 2 selects from this rather than assuming `tmdbid`.
assert_eq!(f["tmdbid"], "95479");
assert_eq!(f["tvdbid"], "367189");