Update app.py

app.py

@@ -3,6 +3,7 @@
 # Advice + Inspirational quotes + Emotion-based color + SQLite DB
 # GoEmotions model + loads GoEmotions dataset ("simplified" config)
 # ================================
+import os
 import re
 import random
 import sqlite3
@@ -13,15 +14,27 @@ import torch
 from transformers import AutoTokenizer, AutoModelForSequenceClassification, TextClassificationPipeline
 from datasets import load_dataset
 
-# --- Paths (persist across Space restarts when Persistent storage is ON) ---
-DB_PATH = "/data/moodmirror.db"
+# --- Storage paths (robust across local dev vs. HF Spaces) ---
+def _pick_data_dir():
+    # Prefer /data if it exists AND is writable (Spaces with persistent storage).
+    if os.path.isdir("/data") and os.access("/data", os.W_OK):
+        return "/data"
+    # Otherwise, fall back to the repo working directory.
+    return os.getcwd()
+
+DATA_DIR = os.getenv("MM_DATA_DIR", _pick_data_dir())
+os.makedirs(DATA_DIR, exist_ok=True)
+DB_PATH = os.path.join(DATA_DIR, "moodmirror.db")
+print(f"[MM] Using data dir: {DATA_DIR}")
+print(f"[MM] SQLite path: {DB_PATH}")
 
 # --- Load GoEmotions dataset ("simplified") ---
 # This pulls from: google-research-datasets/go_emotions
-# The "simplified" config uses standard splits and provides label indices + names.
+# The "simplified" config uses train/validation/test splits and label indices.
 try:
     ds = load_dataset("google-research-datasets/go_emotions", "simplified")
     LABEL_NAMES = ds["train"].features["labels"].feature.names  # e.g. ['admiration', ..., 'neutral']
+    print("[MM] GoEmotions dataset loaded.")
 except Exception as e:
     ds = None
     LABEL_NAMES = None
@@ -156,32 +169,47 @@ THRESHOLD = 0.35  # tune to be more/less sensitive
 
 # --- SQLite setup ---
 def get_conn():
-    return sqlite3.connect(DB_PATH, check_same_thread=False)
+    # timeout helps if multiple requests hit the DB at once
+    return sqlite3.connect(DB_PATH, check_same_thread=False, timeout=10)
 
 def init_db():
-    conn = get_conn()
-    c = conn.cursor()
-    c.execute("""
-    CREATE TABLE IF NOT EXISTS sessions(
-        id INTEGER PRIMARY KEY AUTOINCREMENT,
-        ts TEXT,
-        country TEXT,
-        user_text TEXT,
-        main_emotion TEXT
-    )
-    """)
-    conn.commit()
-    conn.close()
+    conn = None
+    try:
+        conn = get_conn()
+        c = conn.cursor()
+        c.execute("""
+        CREATE TABLE IF NOT EXISTS sessions(
+            id INTEGER PRIMARY KEY AUTOINCREMENT,
+            ts TEXT,
+            country TEXT,
+            user_text TEXT,
+            main_emotion TEXT
+        )
+        """)
+        conn.commit()
+    finally:
+        try:
+            if conn is not None:
+                conn.close()
+        except Exception:
+            pass
 
 def log_session(country, msg, emotion):
-    conn = get_conn()
-    c = conn.cursor()
-    c.execute(
-        "INSERT INTO sessions(ts, country, user_text, main_emotion) VALUES(?,?,?,?)",
-        (datetime.utcnow().isoformat(timespec="seconds"), country, msg[:500], emotion),
-    )
-    conn.commit()
-    conn.close()
+    conn = None
+    try:
+        conn = get_conn()
+        c = conn.cursor()
+        c.execute(
+            "INSERT INTO sessions(ts, country, user_text, main_emotion) VALUES(?,?,?,?)",
+            (datetime.utcnow().isoformat(timespec="seconds"), country, msg[:500], emotion),
+        )
+        conn.commit()
+    finally:
+        try:
+            if conn is not None:
+                conn.close()
+        except Exception:
+            pass
 
 # --- Emotion detection (multi-label via model) ---
 def detect_emotions(text: str):
@@ -191,7 +219,7 @@ def detect_emotions(text: str):
       - main_app: top mapped category for UI/tips/colors
     """
     try:
-        preds = pipe(text)[0]  # list of {'label': 'joy', 'score': 0.82}, for all labels
+        preds = pipe(text)[0]  # list of {'label': 'joy', 'score': 0.82} for all labels
         chosen = [p for p in preds if p["score"] >= THRESHOLD]
         chosen.sort(key=lambda x: x["score"], reverse=True)