fix: cross-reference checks inside ZIP archives

_build_supplemental_index scans both data/ directories and
contents of bios/ ZIP files. Eliminates 197 false positives
where files existed inside archive ZIPs (neogeo.zip, pgm.zip,
stvbios.zip, etc.) but were counted as missing. True missing
drops from 645 to 448.

README.md

@@ -2,7 +2,7 @@
 
 Complete BIOS and firmware packs for Batocera, BizHawk, EmuDeck, Lakka, Recalbox, RetroArch, RetroBat, RetroDECK, RetroPie, and RomM.
 
-**7,532** verified files across **352** systems, ready to extract into your emulator's BIOS directory.
+**7,545** verified files across **352** systems, ready to extract into your emulator's BIOS directory.
 
 ## Download BIOS packs
 
@@ -29,8 +29,8 @@ Each file is checked against the emulator's source code to match what the code a
 - **10 platforms** supported with platform-specific verification
 - **328 emulators** profiled from source (RetroArch cores + standalone)
 - **352 systems** covered (NES, SNES, PlayStation, Saturn, Dreamcast, ...)
-- **7,532 files** verified with MD5, SHA1, CRC32 checksums
-- **8112 MB** total collection size
+- **7,545 files** verified with MD5, SHA1, CRC32 checksums
+- **8115 MB** total collection size
 
 ## Supported systems
 
@@ -113,4 +113,4 @@ See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
 
 This repository provides BIOS files for personal backup and archival purposes.
 
-*Auto-generated on 2026-03-28T16:30:00Z*
+*Auto-generated on 2026-03-28T16:59:13Z*

database.json

@@ -1,5 +1,5 @@
 {
-  "generated_at": "2026-03-28T16:53:02Z",
+  "generated_at": "2026-03-28T16:58:50Z",
   "total_files": 7545,
   "total_size": 8508938896,
   "files": {

scripts/cross_reference.py

@@ -50,16 +50,29 @@ def load_platform_files(platforms_dir: str) -> tuple[dict[str, set[str]], dict[s
     return declared, platform_data_dirs
 
 
-def _build_data_dir_index(data_root: str = "data") -> set[str]:
-    """Build a set of filenames available in data/ directories."""
+def _build_supplemental_index(data_root: str = "data",
+                               bios_root: str = "bios") -> set[str]:
+    """Build a set of filenames in data/ directories and inside bios/ ZIPs."""
     names: set[str] = set()
     root_path = Path(data_root)
-    if not root_path.is_dir():
-        return names
-    for fpath in root_path.rglob("*"):
-        if fpath.is_file() and not fpath.name.startswith("."):
-            names.add(fpath.name)
-            names.add(fpath.name.lower())
+    if root_path.is_dir():
+        for fpath in root_path.rglob("*"):
+            if fpath.is_file() and not fpath.name.startswith("."):
+                names.add(fpath.name)
+                names.add(fpath.name.lower())
+    bios_path = Path(bios_root)
+    if bios_path.is_dir():
+        import zipfile
+        for zpath in bios_path.rglob("*.zip"):
+            try:
+                with zipfile.ZipFile(zpath) as zf:
+                    for member in zf.namelist():
+                        if not member.endswith("/"):
+                            basename = member.rsplit("/", 1)[-1] if "/" in member else member
+                            names.add(basename)
+                            names.add(basename.lower())
+            except (zipfile.BadZipFile, OSError):
+                pass
     return names
 
 
@@ -226,7 +239,7 @@ def main():
 
     declared, plat_data_dirs = load_platform_files(args.platforms_dir)
     db = load_database(args.db)
-    data_names = _build_data_dir_index()
+    data_names = _build_supplemental_index()
     report = cross_reference(profiles, declared, db, plat_data_dirs, data_names)
 
     if args.json:

scripts/verify.py

@@ -594,8 +594,8 @@ def verify_platform(
         status_counts[s] = status_counts.get(s, 0) + 1
 
     # Cross-reference undeclared files
-    from cross_reference import _build_data_dir_index
-    data_names = _build_data_dir_index()
+    from cross_reference import _build_supplemental_index
+    data_names = _build_supplemental_index()
     undeclared = find_undeclared_files(config, emulators_dir, db, emu_profiles,
                                        target_cores=target_cores, data_names=data_names)
     exclusions = find_exclusion_notes(config, emulators_dir, emu_profiles, target_cores=target_cores)