Fix radarr_cleanup.sh missing secondary edition files as false-positive orphans

Radarr 6+ supports multiple tracked files per movie (alternate editions/extras),
but the movie list's embedded movieFile.path only ever reflects the primary one.
Relying on it alone flagged legitimately-tracked secondary edition files as
orphans. Now batch-fetches /moviefile?movieId=X across all tracked movies to
catch every file, not just the primary.

Arrs_Stack/radarr_cleanup.sh

@@ -18,19 +18,26 @@
 # false-alarm abort. Mirrors the same fix built for lidarr_cleanup.sh 2026-07-16 after
 # a whole-library rescan there made trackFileCount read 22% of normal mid-scan.
 #
-# Cache-first movie list, no per-movie API calls at all (2026-07-17). The movie list comes
-# from the shared tracked-data cache via arr_get_tracked_data() — fresh (kept warm every
-# 30min by arr_cache_prefill.sh), live fetch as fallback. Radarr is structurally different
-# from Lidarr/Sonarr here: its movie list already embeds movieFile.path directly on every
-# hasFile=true entry (confirmed live, zero exceptions across the full library), so there's
-# no separate per-movie moviefile?movieId=X walk needed at all — what used to be up to 2896
-# individual API calls is now a jq filter over data already in hand. The filesystem is
-# walked once per run, not twice — classification records which paths are eligible for
-# deletion as it goes, and the delete pass (once the size-threshold check below passes)
-# just acts on that list instead of re-walking and re-classifying the whole tree. That single
-# walk also gets size+mtime straight from find -printf instead of a separate stat fork per
-# file — find already has to stat() every entry to know it's -type f, so this is free by
-# comparison. Measured ~130x faster per file (0.033ms vs 4.3ms).
+# Cache-first movie list (2026-07-17), batched moviefile fetch (2026-07-19). The movie list
+# comes from the shared tracked-data cache via arr_get_tracked_data() — fresh (kept warm
+# every 30min by arr_cache_prefill.sh), live fetch as fallback. Radarr's movie list embeds
+# movieFile.path directly on every hasFile=true entry, but that's only the *primary* file —
+# Radarr 6+ supports a second tracked file per movie (alternate editions/extras) that never
+# shows up there, so relying on it alone misclassified a movie's second edition as an orphan
+# (confirmed live 2026-07-19: The Crash, They Will Kill You, The Drama, Lee Cronin's The
+# Mummy, and Ready or Not: Here I Come all had a legitimately-tracked second file deleted-flagged
+# this way). /moviefile?movieId=X returns every file for a movie, including secondaries, and
+# accepts movieId as a repeated query param for a bulk fetch — but the whole library in one
+# request 414s (Request-URI Too Long, confirmed live), so _fetch_tracked_files() batches
+# movieId params BATCH_SIZE at a time instead: ~14 requests for a ~2800-movie library rather
+# than the up-to-2896 individual per-movie calls the 2026-07-17 optimization eliminated, and
+# rather than the one-shot list read that missed secondary files. The filesystem is walked
+# once per run, not twice — classification records which paths are eligible for deletion as
+# it goes, and the delete pass (once the size-threshold check below passes) just acts on that
+# list instead of re-walking and re-classifying the whole tree. That single walk also gets
+# size+mtime straight from find -printf instead of a separate stat fork per file — find
+# already has to stat() every entry to know it's -type f, so this is free by comparison.
+# Measured ~130x faster per file (0.033ms vs 4.3ms).
 #
 # ==============================================================================================
 # OPERATIONAL MODEL
@@ -320,26 +327,43 @@ info "Found $MOVIE_COUNT movies — fetching movie files..."
 TRACKED_FILE="$TMP_DIR/tracked_paths.txt"
 > "$TRACKED_FILE"
 
-# Fetches every movie's file path fresh into TRACKED_FILE/TRACKED_MAP/TRACKED_COUNT. Pulled
-# into a function so the rescan-aware retry below can re-fetch after waiting without
+# Fetches every movie's file path(s) fresh into TRACKED_FILE/TRACKED_MAP/TRACKED_COUNT.
+# Pulled into a function so the rescan-aware retry below can re-fetch after waiting without
 # duplicating this whole loop inline.
 #
-# Radarr's movie list already embeds movieFile.path directly on every entry with
-# hasFile=true (confirmed live 2026-07-17, zero exceptions across the full library) —
-# unlike Lidarr/Sonarr, where trackFile/episodefile genuinely require a separate per-item
-# call. A single live movie-list fetch replaces what used to be one API call per movie
-# (2896 of them) for the same data already sitting in that response. Still a fresh live
-# fetch on every call (not cache-first) — this function's whole purpose during the
+# Batched, not per-movie and not a single one-shot list read (2026-07-19) — see the header
+# comment above for why movie.movieFile.path alone misses secondary edition files. Still a
+# fresh live fetch on every call (not cache-first) — this function's whole purpose during the
 # rescan-aware retry below is to see Radarr's progress as the rescan updates hasFile/
 # movieFile, so it needs genuinely current data each time, not a stale snapshot.
 _fetch_tracked_files() {
     > "$TRACKED_FILE"
     local movies_now
     movies_now=$(arr_api "$RADARR_URL" "$RADARR_API_KEY" "v3" "movie" "Radarr" 2>/dev/null)
-    while IFS= read -r api_path; do
+
+    local _ids=() _id _qs="" _batch_count=0
+    local BATCH_SIZE=200   # 250 confirmed working live 2026-07-19; kept under that for margin
+    mapfile -t _ids < <(echo "$movies_now" | jq -r '.[] | select(.hasFile==true) | .id' 2>/dev/null)
+
+    {
+        for _id in "${_ids[@]}"; do
+            _qs+="movieId=${_id}&"
+            (( _batch_count++ ))
+            if [[ "$_batch_count" -ge "$BATCH_SIZE" ]]; then
+                arr_api "$RADARR_URL" "$RADARR_API_KEY" "v3" "moviefile?${_qs%&}" "Radarr" 2>/dev/null | \
+                    jq -r '.[].path' 2>/dev/null
+                _qs=""
+                _batch_count=0
+            fi
+        done
+        if [[ -n "$_qs" ]]; then
+            arr_api "$RADARR_URL" "$RADARR_API_KEY" "v3" "moviefile?${_qs%&}" "Radarr" 2>/dev/null | \
+                jq -r '.[].path' 2>/dev/null
+        fi
+    } | while IFS= read -r api_path; do
         [[ -z "$api_path" ]] && continue
         translate_path "$api_path" >> "$TRACKED_FILE"
-    done < <(echo "$movies_now" | jq -r '.[] | select(.hasFile==true) | .movieFile.path' 2>/dev/null)
+    done
 
     sort -u "$TRACKED_FILE" -o "$TRACKED_FILE"