feat: add optional MongoDB GridFS upload

README.md

@@ -48,3 +48,16 @@ If you prefer uploading to Google Drive:
 
 The app uses `google-api-python-client` to upload each WAV file into that folder. Errors will be shown in the results area if credentials or permissions are incorrect.
 
+## MongoDB Upload (Alternative)
+
+You can also upload recordings to MongoDB using GridFS.
+
+Secrets to configure in your Space:
+- `MONGO_URI`: your MongoDB connection string (supports `mongodb+srv://`)
+- `MONGO_DB`: database name (default: `spells`)
+- `GRIDFS_BUCKET`: GridFS bucket prefix (default: `fs`)
+
+Then in the UI, tick "Upload to MongoDB (GridFS)" before Submit.
+
+Each file is stored in GridFS with metadata: `spell`, `username`, `timestamp`, and original `filename`.
+

app.py

@@ -25,6 +25,14 @@ except Exception:
     build = None
     MediaFileUpload = None
 
+# MongoDB (GridFS) optional imports
+try:
+    from pymongo import MongoClient
+    import gridfs
+except Exception:
+    MongoClient = None
+    gridfs = None
+
 # Output directory for saved recordings
 OUT_DIR = "recordings"
 os.makedirs(OUT_DIR, exist_ok=True)
@@ -208,6 +216,83 @@ def upload_recordings_to_gdrive(paths: Sequence[str]) -> Tuple[int, Optional[str
     return uploaded, None
 
 
+def _parse_meta_from_filename(basename: str) -> Tuple[str, str, Optional[int]]:
+    """Parse (spell_slug, username, timestamp) from `<spell_slug>_<username>_<ts>.wav`.
+    Username and spell slug can contain underscores; timestamp is the last token.
+    """
+    name = basename
+    if name.endswith(".wav"):
+        name = name[:-4]
+    parts = name.split("_")
+    if len(parts) < 3:
+        return name, "", None
+    try:
+        ts = int(parts[-1])
+    except Exception:
+        ts = None
+    username = parts[-2]
+    spell_slug = "_".join(parts[:-2])
+    return spell_slug, username, ts
+
+
+def upload_recordings_to_mongo(paths: Sequence[str]) -> Tuple[int, Optional[str]]:
+    """Upload files to MongoDB using GridFS.
+
+    Env configuration:
+    - MONGO_URI: connection string (e.g., mongodb+srv://user:pass@cluster/)
+    - MONGO_DB: database name (default: spells)
+    - GRIDFS_BUCKET: GridFS bucket prefix (default: fs)
+    Returns (uploaded_count, error_message).
+    """
+    if not paths:
+        return 0, None
+    if not (MongoClient and gridfs):
+        return 0, "pymongo/gridfs not installed."
+    uri = os.getenv("MONGO_URI")
+    if not uri:
+        return 0, "Missing MONGO_URI."
+    db_name = os.getenv("MONGO_DB", "spells")
+    bucket = os.getenv("GRIDFS_BUCKET", "fs")
+
+    try:
+        client = MongoClient(uri, serverSelectionTimeoutMS=5000)
+        # quick connectivity check
+        client.admin.command("ping")
+        db = client[db_name]
+        fs = gridfs.GridFS(db, collection=bucket)
+    except Exception as e:
+        return 0, f"Mongo connect error: {e}"
+
+    uploaded = 0
+    try:
+        for p in paths:
+            if not os.path.isfile(p):
+                continue
+            base = os.path.basename(p)
+            spell_slug, username, ts = _parse_meta_from_filename(base)
+            with open(p, "rb") as f:
+                fs.put(
+                    f.read(),
+                    filename=base,
+                    contentType="audio/wav",
+                    metadata={
+                        "spell": spell_slug,
+                        "username": username,
+                        "timestamp": ts,
+                        "path": p,
+                    },
+                )
+            uploaded += 1
+    except Exception as e:
+        return uploaded, f"Mongo upload error: {e}"
+    finally:
+        try:
+            client.close()
+        except Exception:
+            pass
+    return uploaded, None
+
+
 def submit_recordings(
     username: str,
     lumos_path: Optional[str],
@@ -218,6 +303,7 @@ def submit_recordings(
     reparo_path: Optional[str],
     upload_flag: bool,
     gdrive_flag: bool,
+    mongo_flag: bool,
 ) -> str:
     user = sanitize_username(username)
 
@@ -270,6 +356,14 @@ def submit_recordings(
         else:
             lines.append(f"Drive upload: {gup} file(s) uploaded to folder.")
 
+    if mongo_flag:
+        mup, merr = upload_recordings_to_mongo(saved_paths)
+        lines.append("")
+        if merr:
+            lines.append(f"Mongo upload attempted: {mup} succeeded, error: {merr}")
+        else:
+            lines.append(f"Mongo upload: {mup} file(s) stored in GridFS.")
+
     return "\n".join(lines)
 
 
@@ -298,12 +392,13 @@ def build_ui() -> gr.Blocks:
         with gr.Row():
             upload_checkbox = gr.Checkbox(label="Upload to Hub (requires HF_TOKEN)", value=False)
             gdrive_checkbox = gr.Checkbox(label="Upload to Google Drive (service account)", value=False)
+            mongo_checkbox = gr.Checkbox(label="Upload to MongoDB (GridFS)", value=False)
         submit = gr.Button("Submit")
         result = gr.Markdown()
 
         submit.click(
             fn=submit_recordings,
-            inputs=[username, lumos, nox, alohomora, wingardium, accio, reparo, upload_checkbox, gdrive_checkbox],
+            inputs=[username, lumos, nox, alohomora, wingardium, accio, reparo, upload_checkbox, gdrive_checkbox, mongo_checkbox],
             outputs=[result],
         )
 
@@ -312,6 +407,7 @@ def build_ui() -> gr.Blocks:
     - Files are saved locally in `recordings/` with `<spell>_<username>_<timestamp>.wav`.
     - Check "Upload to Hub" to commit them to the repo (needs HF_TOKEN secret).
     - Or check "Upload to Google Drive" to upload via a service account.
+    - Or check "Upload to MongoDB (GridFS)" to store in your database.
     - 16 kHz mono WAV ensures consistent model training.
     - You can submit partial sets; only provided spells are saved.
         """)

requirements.txt

@@ -6,4 +6,5 @@ huggingface_hub
 google-api-python-client
 google-auth
 google-auth-httplib2
-google-auth-oauthlib
+google-auth-oauthlib
+pymongo