Update app.py

app.py

@@ -88,11 +88,13 @@ def read_gen_logs():
 
 def list_shards(folder):
     if not folder or not os.path.isdir(folder): return "❌ Provide a valid folder path."
+    # ⬇ Only JSONL shards; ignore manifest files
     jsonl = sorted(glob.glob(os.path.join(folder,"*.jsonl")))
     gz    = sorted(glob.glob(os.path.join(folder,"*.jsonl.gz")))
-    total = len(jsonl)+len(gz)
-    if total==0: return "No shards found."
-    preview=(jsonl+gz)[:10]
+    files = [p for p in (jsonl+gz) if "manifest" not in os.path.basename(p).lower()]
+    total = len(files)
+    if total==0: return "No shards found (*.jsonl / *.jsonl.gz)."
+    preview=files[:10]
     lines=[f"Found {total} shard(s). Showing first {len(preview)}:"]+[f"- {os.path.basename(p)}" for p in preview]
     return "\n".join(lines)
 
@@ -117,12 +119,12 @@ def _train_worker(dataset_path: str, shards_folder: str):
     with open(LOG_FILE,"a") as log:
         if shards_folder:
             log.write(f"📂 Folder mode: {shards_folder}\n")
+            # ⬇ Only JSONL shards; ignore manifest files
             paths = sorted(glob.glob(os.path.join(shards_folder,"*.jsonl"))) + \
-                    sorted(glob.glob(os.path.join(shards_folder,"*.json")))  + \
-                    sorted(glob.glob(os.path.join(shards_folder,"*.jsonl.gz"))) + \
-                    sorted(glob.glob(os.path.join(shards_folder,"*.json.gz")))
+                    sorted(glob.glob(os.path.join(shards_folder,"*.jsonl.gz")))
+            paths = [p for p in paths if "manifest" not in os.path.basename(p).lower()]
             if not paths:
-                log.write("❌ No shards found. Aborting.\n"); ok=False
+                log.write("❌ No shards found (*.jsonl / *.jsonl.gz). Aborting.\n"); ok=False
             else:
                 tmp="tmp_train.jsonl"
                 for i,pth in enumerate(paths,1):
@@ -232,7 +234,7 @@ with gr.Blocks(title="JSON AI Trainer (with Dataset Generator)") as app:
         list_btn.click(fn=list_shards, inputs=list_folder, outputs=list_out)
 
     with gr.Tab("🧠 Train"):
-        gr.Markdown("Upload a single JSON/JSONL file *or* train on a folder of shards (.json, .jsonl, .jsonl.gz, .json.gz).")
+        gr.Markdown("Upload a single JSON/JSONL file *or* train on a folder of shards (.jsonl, .jsonl.gz). Manifests are ignored.")
         with gr.Row():
             file_input = gr.File(label="Upload single dataset file", file_types=[".json",".jsonl"])
             upload_btn = gr.Button("📤 Upload (single file)")