Fix: replace api.pokemontcg.io with api.tcgdex.net (#20)

docs/assets-and-info-apis.md

@@ -20,28 +20,33 @@ Used by `MagicCardPreviewSource` to find a card printing from `name` + `setId`,
 
 Unified **Pokemon** Game menu entry. Per-card `region` (`West` / `Asia`) selects the backend below. Collection: `pokemon/collection.json`. West sets: `pokemon/sets-west.json`. Asia sets: `pokemon/sets-asia.json`. Language choices: West → English/German/French/Spanish/Italian/Russian; Asia → Japanese/S-Chinese/T-Chinese/Korean.
 
-### West (`Game::Pokemon`, pokemontcg.io)
+### West (`Game::Pokemon`, TCGdex EN)
 
-**Info API:** `https://api.pokemontcg.io/v2/sets`  
-Used by `PokemonSetSource` to fetch all sets. The parser maps `id`, `name`, and `releaseDate` directly into `Set`, then sorts ascending by release date.
+Upstream: [TCGdex REST API](https://tcgdex.dev/) locale `en`. No API key. Canonical West set ids are TCGdex EN ids (e.g. `base1`, `sv01`, `swsh12.5tg`). Legacy pokemontcg.io ids (`sv1`, `pgo`, `swsh12tg`, …) are rewritten via `canonicalizeWestSetId` on West collection load, preview/auto-detect lookups, and set-completion matching so existing collections keep working; the next save persists TCGdex ids.
+
+**Info API:** `https://api.tcgdex.net/v2/en/sets`  
+Used by `PokemonSetSource` to fetch the slim set list (`id`, `name`). Release dates are not on the list endpoint — each set’s `GET /v2/en/sets/{id}` supplies `releaseDate` as `YYYY-MM-DD`, rewritten to `YYYY/MM/DD`, then the list is sorted ascending by release date.
+
+**Asset API:** `https://api.tcgdex.net/v2/en/cards/{setId}-{localId}` (by id), `https://api.tcgdex.net/v2/en/cards?…` (filtered search), and set-detail `cards[]` for auto-detect. Image CDN bases live on `assets.tcgdex.net`; the preview source appends `/high.png` (wxImage decodes PNG, not webp).
 
-**Asset API:** `https://api.pokemontcg.io/v2/cards` (by id) and `https://api.pokemontcg.io/v2/cards?q=...` (search)  
 Used by `PokemonCardPreviewSource` in two ways:
 
-1. **Preview lookup (`fetchImageUrl`).** When both set id and collector number are present, prefers `GET /v2/cards/{setId}-{number}` (single-card `data` object) — same idea as Asia’s direct localId fetch — so Lucene `name:` ∩ `number:` misses cannot blank the preview after Auto-detect fills Set #. On HTTP failure or missing images, falls back to a name-less search `set.id:… number:…` (collector numbers are unique within a set). When Set # or set id is missing, keeps the older `name:"…"` search with optional `set.id` / `number`. The parser takes `images.large` first and falls back to `images.small`.
+1. **Preview lookup (`fetchImageUrl`).** When both set id and collector number are present, prefers `GET /v2/en/cards/{setId}-{localId}` (card object with `image` base). On HTTP failure or missing image, falls back to a filtered search `set.id=eq:…&localId=eq:…` (collector numbers are unique within a set). When Set # or set id is missing, uses `name=eq:…` with optional `set.id` / `localId`. Legacy set ids are canonicalized before URL build.
 
-2. **Auto-detect print (`detectFirstPrint` / `detectPrintVariants`, Pokémon edit dialog).** Uses the search endpoint with `name:"<name>"` and `set.id:<setId>` only — **no** `number:` clause — plus `select=name,number,rarity,set` and `pageSize=50` so the response stays small. If the set-scoped HTTP request fails, it retries with **`name:` only** and still filters rows in `PokemonCardPreviewSource::parsePrintVariants(...)` by the picker’s **`set.id`** (not the display set name). The dialog passes `card.set.id` into `CardPreviewService::detectPrintVariants(...)` on a worker thread so the modal stays responsive. Each matching `data[]` row whose **card name matches exactly** (case-insensitive) and whose embedded `set.id` equals the chosen set maps to `AutoDetectedPrint::setNo` as the API `number` field only (for example `25`, not `25/185`). `AutoDetectedPrint::rarity` is filled from the card’s `rarity` field but the Pokémon edit dialog does not auto-sync holo or other flags from it. Distinct `(setNo, rarity)` pairs are deduped. When both an exact card name and `set.id` are supplied, an upstream miss returns an error instead of blending unrelated sets from a broader payload. The edit dialog offers **Auto detect** (fills Set # from the first variant), **Next** (cycles distinct `setNo` values when multiple exist), silent prefetch on **Edit** open, and clears cached variants when **Name** or **Set** changes. The Set # field and persisted `PokemonCard::setNo` keep only the printed-number portion; values such as `4/104` are trimmed to `4` on load and save.
+2. **Auto-detect print (`detectFirstPrint` / `detectPrintVariants`, Pokémon edit dialog).** Prefers `GET /v2/en/sets/{setId}` and filters `cards[]` by exact case-insensitive card name. Maps `localId` → `AutoDetectedPrint::setNo` and `rarity` → `AutoDetectedPrint::rarity` (the edit dialog does not auto-sync holo flags from rarity). If set detail fails, falls back to a filtered cards search and still restricts rows to the chosen set id when present. Distinct `(setNo, rarity)` pairs are deduped. The edit dialog offers **Auto detect**, **Next**, silent prefetch on **Edit** open, and clears cached variants when **Name** or **Set** changes. The Set # field and persisted `PokemonCard::setNo` keep only the printed-number portion; values such as `4/104` are trimmed to `4` on load and save.
 
-The preview path normalizes collector numbers before request build. For example, `4/102` is reduced to `4` because the remote `number:` query and card-id path expect only the printed-number component (unquoted `number:4` / `number:TG14`; do not wrap alphanumeric numbers in Lucene quotes when combining with other clauses — that has been observed to 500 on the live API).
+The preview path normalizes collector numbers before request build. For example, `4/102` is reduced to `4` because the remote `localId` path expects only the printed-number component.
 
 ### Set-completion catalog (West)
 
 **Sets → Update Pokemon** uses `PokemonSetSource::fetchAllWithCatalog()` so the West path writes:
 
-1. The set list (`pokemon/sets-west.json`) from `/v2/sets` (same as before)
-2. A pack checklist at `<dataStorage>/pokemon/set-catalog-west.json` from a paginated `/v2/cards?select=name,number,set&pageSize=250` dump
+1. The set list (`pokemon/sets-west.json`) from `/v2/en/sets` + per-set detail dates
+2. A pack checklist at `<dataStorage>/pokemon/set-catalog-west.json` from each set’s detail `cards[]` (`localId` → `setNo`, `name` → name)
 
-Each catalog pack stores `id` (pokemontcg.io set id), `name` (display), and `cards[]` of `{ setNo, name }` keyed by the API `number` field (normalized by stripping anything after `/`). Duplicate collector numbers within a pack collapse to one checklist row. The Pokemon **Set Completion** tab reads this file offline; ownership for a West pack requires `PokemonRegion::West`, matching `card.set.id`, and a normalized collector number match. Amount / holo / 1st Edition are ignored for completion counts.
+Each catalog pack stores `id` (TCGdex EN set id), `name` (display), and `cards[]` of `{ setNo, name }` keyed by `localId` (normalized by stripping anything after `/`). Duplicate collector numbers within a pack collapse to one checklist row. The Pokemon **Set Completion** tab reads this file offline; ownership for a West pack requires `PokemonRegion::West`, a canonicalized `card.set.id` match, and a normalized collector number match. Amount / holo / 1st Edition are ignored for completion counts.
+
+After a successful Update, `PokemonGameView` also runs `syncPokemonCollectionSets` against the refreshed set lists: West cards get legacy set-id migration plus `set.name` / `releaseDate` refresh when the id is present; Asia cards refresh name/date the same way. Changed cards are persisted via `CollectionService::saveAll`.
 
 If `set-catalog-west.json` is missing (and the active region filter is West or All with no Asia catalog either), the Set Completion tab prompts the user to run Update Pokemon.
 
@@ -156,7 +161,7 @@ Empty search array / `{"error":"..."}` → `NotFound`; bad JSON / HTTP → `Tran
 
 ## Japanese Pokémon TCG APIs (TCGdex `ja`) — Asia region backend
 
-Asia Pokémon is routed internally as `Game::JapanesePokemon` (`dirName` `pokemon`, same data directory as West). It is **not** a separate Game menu entry: the unified **Pokemon** UI stores both West and Asia cards in `pokemon/collection.json` with a per-card `region` (`West` / `Asia`). Set caches are split by filename under that directory (`pokemon/sets-west.json` vs `pokemon/sets-asia.json`). `JsonSetRepository` migrate-on-load promotes legacy `pokemon/sets.json` → `sets-west.json` and `pokemonjp/sets.json` → `sets-asia.json` when the new files are missing. **Sets > Update Pokemon** refreshes both lists. Upstream: [TCGdex REST API](https://tcgdex.dev/). No API key. Japanese set IDs (e.g. `PMCG1`, `SV1a`) are never merged into Western pokemontcg.io ids.
+Asia Pokémon is routed internally as `Game::JapanesePokemon` (`dirName` `pokemon`, same data directory as West). It is **not** a separate Game menu entry: the unified **Pokemon** UI stores both West and Asia cards in `pokemon/collection.json` with a per-card `region` (`West` / `Asia`). Set caches are split by filename under that directory (`pokemon/sets-west.json` vs `pokemon/sets-asia.json`). `JsonSetRepository` migrate-on-load promotes legacy `pokemon/sets.json` → `sets-west.json` and `pokemonjp/sets.json` → `sets-asia.json` when the new files are missing. **Sets > Update Pokemon** refreshes both lists. Upstream: [TCGdex REST API](https://tcgdex.dev/). No API key. Japanese set IDs (e.g. `PMCG1`, `SV1a`) are never merged into Western TCGdex EN ids.
 
 ### Info API: TCGdex `GET /v2/ja/sets` (+ per-set detail)
 
@@ -294,7 +299,7 @@ Three mechanisms reduce preview latency for **all** games (Magic, Pokemon West/A
 
 - **In-memory preview LRU** (`CardPreviewService`). Successful `fetchPreviewBytes` results are cached keyed by `(game, name, setId, setNo)`; successful `fetchImageBytesByUrl` results are cached keyed by URL (used for the per-game card-back fallback). Re-selecting a previously viewed row is decode-only — no HTTP at all. The cache is bounded by `CardPreviewService::kCacheCapacity` (currently 128 entries) and uses a list+map LRU under a mutex (the preview pipeline is invoked from a worker thread in `BaseSelectedCardPanel`). **Source errors are split** by `PreviewLookupError::Kind`: `NotFound` (the upstream answered cleanly that the record has no image) is *negative-cached* in this tier so subsequent selections short-circuit without HTTP, while `Transient` (HTTP/network/parse failures) is **never** cached so a brief outage cannot permanently disable a card's preview.
 - **Persistent disk byte cache** (`LocalPreviewByteCache`, port `IPreviewByteCache`). Wraps the in-memory tier with an on-disk store under `<exeDir>/.cache/preview-cache/` — pinned **next to the executable**, in the same scope as `config.json`, **not** under the user-configurable `Configuration.dataStorage` path. The cache stays put when the user reconfigures or relocates their collection data, and it is not part of the user's data directory backups; it is install-scoped, not collection-scoped. Both positive previews and `NotFound` verdicts survive an app restart. Each entry is a mutually-exclusive `<hash>.bin` (positive payload) or `<hash>.neg` (negative marker) plus a `<hash>.idx` sidecar containing the original key — load-time mismatch on the sidecar treats the entry as a miss, so a hash collision degrades to a one-time HTTP refetch instead of serving the wrong card's bytes (or the wrong card's "no image" verdict). Hashing is FNV-1a 64-bit (no crypto dependency). The cache is bounded by total `.bin` payload bytes (default `kDefaultMaxBytes = 64 MiB`) and evicts oldest entries by mtime when a new write would exceed the cap; reading an entry touches its mtime so frequently-viewed cards survive eviction. Negative `.neg` markers are tiny and not counted against the cap — their count is naturally bounded by the user's actively-viewed records. Filesystem mutations route through `IFileSystem`; size and mtime queries (which the port does not expose) use `std::filesystem` directly inside the adapter. The persistent tier is **fire-and-forget on the way down** — every adapter operation swallows I/O errors so a flaky or full disk never breaks the preview path.
-- **Persistent HTTP session** (`CprHttpClient`). The adapter owns one long-lived `cpr::Session` (libcurl easy handle) for the lifetime of the app. Per-request configuration is limited to `SetUrl(...)`; headers, timeout, and redirect policy are configured once in the constructor. Default **`Accept: */*`** keeps JSON responses and raw image bodies working on the same session (avoid tying every GET to `application/json`). libcurl's connection pool keeps the TLS connection to each host warm, so repeat calls to `api.scryfall.com`, `api.pokemontcg.io`, `db.ygoprodeck.com`, `yugipedia.com`, `ms.yugipedia.com`, `digimoncard.io`, and `images.digimoncard.io` skip the TLS handshake. A `std::mutex` serializes callers — libcurl easy handles are not thread-safe, and the preview pipeline is single-flight per panel anyway.
+- **Persistent HTTP session** (`CprHttpClient`). The adapter owns one long-lived `cpr::Session` (libcurl easy handle) for the lifetime of the app. Per-request configuration is limited to `SetUrl(...)`; headers, timeout, and redirect policy are configured once in the constructor. Default **`Accept: */*`** keeps JSON responses and raw image bodies working on the same session (avoid tying every GET to `application/json`). libcurl's connection pool keeps the TLS connection to each host warm, so repeat calls to `api.scryfall.com`, `api.tcgdex.net`, `assets.tcgdex.net`, `db.ygoprodeck.com`, `yugipedia.com`, `ms.yugipedia.com`, `digimoncard.io`, and `images.digimoncard.io` skip the TLS handshake. A `std::mutex` serializes callers — libcurl easy handles are not thread-safe, and the preview pipeline is single-flight per panel anyway.
 
 `CardPreviewService` consults the tiers in order **memory → disk → source/HTTP**. On a disk hit (positive *or* negative) the entry is promoted into the in-memory LRU so the next click on the same row never re-touches the disk cache. On HTTP success the bytes are written through to both tiers in one shot. On a `NotFound` source error the **negative** marker is written through to both tiers; on `Transient` source errors nothing is written, so the next selection retries cleanly.