fix: sha1-based large file restore, fix broken data dir urls

Replace grep-based restore with SHA1 matching via database.json.
The old grep heuristic failed for assets with renamed basenames
(dsi_nand_batocera42.bin) or special characters (MAME dots vs
spaces), and only restored to the first .gitignore match when
multiple paths shared a basename.

Fix 3 broken data directory sources:
- opentyrian: buildbot URL 404, use release asset
- syobonaction: invalid git_subtree URL, use GitHub archive
- stonesoup: same fix, adds 532 game data files

.github/workflows/build.yml

@@ -58,16 +58,32 @@ jobs:
         run: |
           mkdir -p .cache/large
           gh release download large-files -D .cache/large/ 2>/dev/null || true
-          for f in .cache/large/*; do
-            [ -f "$f" ] || continue
-            name=$(basename "$f")
-            target=$(grep "$name" .gitignore | head -1)
-            if [ -n "$target" ] && [ ! -f "$target" ]; then
-              mkdir -p "$(dirname "$target")"
-              cp "$f" "$target"
-              echo "Restored: $target"
-            fi
-          done
+          python3 -c "
+          import hashlib, json, os, shutil
+          db = json.load(open('database.json'))
+          with open('.gitignore') as f:
+              ignored = {l.strip() for l in f if l.strip().startswith('bios/')}
+          cache = '.cache/large'
+          if not os.path.isdir(cache):
+              exit(0)
+          idx = {}
+          for fn in os.listdir(cache):
+              fp = os.path.join(cache, fn)
+              if os.path.isfile(fp):
+                  h = hashlib.sha1(open(fp, 'rb').read()).hexdigest()
+                  idx[h] = fp
+          restored = 0
+          for sha1, entry in db['files'].items():
+              path = entry['path']
+              if path in ignored and not os.path.exists(path):
+                  src = idx.get(sha1)
+                  if src:
+                      os.makedirs(os.path.dirname(path), exist_ok=True)
+                      shutil.copy2(src, path)
+                      print(f'Restored: {path}')
+                      restored += 1
+          print(f'Total: {restored} files restored')
+          "
         env:
           GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 

.github/workflows/deploy-site.yml

@@ -40,16 +40,32 @@ jobs:
         run: |
           mkdir -p .cache/large
           gh release download large-files -D .cache/large/ 2>/dev/null || true
-          for f in .cache/large/*; do
-            [ -f "$f" ] || continue
-            name=$(basename "$f")
-            target=$(grep "$name" .gitignore | head -1)
-            if [ -n "$target" ] && [ ! -f "$target" ]; then
-              mkdir -p "$(dirname "$target")"
-              cp "$f" "$target"
-              echo "Restored: $target"
-            fi
-          done
+          python3 -c "
+          import hashlib, json, os, shutil
+          db = json.load(open('database.json'))
+          with open('.gitignore') as f:
+              ignored = {l.strip() for l in f if l.strip().startswith('bios/')}
+          cache = '.cache/large'
+          if not os.path.isdir(cache):
+              exit(0)
+          idx = {}
+          for fn in os.listdir(cache):
+              fp = os.path.join(cache, fn)
+              if os.path.isfile(fp):
+                  h = hashlib.sha1(open(fp, 'rb').read()).hexdigest()
+                  idx[h] = fp
+          restored = 0
+          for sha1, entry in db['files'].items():
+              path = entry['path']
+              if path in ignored and not os.path.exists(path):
+                  src = idx.get(sha1)
+                  if src:
+                      os.makedirs(os.path.dirname(path), exist_ok=True)
+                      shutil.copy2(src, path)
+                      print(f'Restored: {path}')
+                      restored += 1
+          print(f'Total: {restored} files restored')
+          "
         env:
           GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 

platforms/_data_dirs.yml

@@ -150,22 +150,23 @@ data_directories:
     description: "SDLPAL Chinese Paladin game data (.mkf archives)"
 
   # ref: OpenTyrian2000 — system/opentyrian/tyrian/
-  # Tyrian 2.1 freeware data (also on buildbot as OpenTyrian.zip)
+  # Tyrian 2.1 freeware data (buildbot URLs removed, sourced from release asset)
   opentyrian:
-    source_url: "https://buildbot.libretro.com/assets/system/OpenTyrian%20%28Game%20Data%29.zip"
+    source_url: "https://github.com/Abdess/retrobios/releases/download/large-files/opentyrian-data.zip"
     source_type: zip
     for_platforms: [retroarch, lakka, retropie]
     local_cache: data/opentyrian
     description: "OpenTyrian Tyrian 2.1 freeware game data"
 
   # ref: syobonaction — system/syobonaction/
-  # Freeware game data from OpenSyobonAction
+  # Freeware game data from OpenSyobonAction (BGM, res, SE directories)
   syobonaction:
-    source_url: "https://github.com/akemin-dayo/OpenSyobonAction"
-    source_type: git_subtree
-    source_path: "res"
+    source_url: "https://github.com/akemin-dayo/OpenSyobonAction/archive/refs/heads/{version}.tar.gz"
+    source_type: tarball
+    source_path: "OpenSyobonAction-master"
     version: master
     local_cache: data/syobonaction
+    exclude: [DxLib.cpp, DxLib.h, icon.ico, joyconfig.h, loadg.cpp, main.cpp, main.h, Makefile, README_ja.md, README.md]
     description: "Syobon Action (Cat Mario) game data (sprites, BGM, SE)"
 
   # =========================================================================
@@ -190,9 +191,10 @@ data_directories:
   # Not on buildbot — sourced from libretro repo
   # 532 files (tiles, fonts, databases, lua scripts, level descriptions)
   stonesoup:
-    source_url: "https://github.com/libretro/crawl-ref"
-    source_type: git_subtree
-    source_path: "crawl-ref/source/dat"
+    source_url: "https://github.com/libretro/crawl-ref/archive/refs/heads/{version}.tar.gz"
+    source_type: tarball
+    source_path: "crawl-ref-master/crawl-ref/source/dat"
+    version: master
     local_cache: data/stonesoup
     description: "DCSS game data (tiles, fonts, databases, lua, level descriptions)"