Spaces:
Sleeping
Sleeping
File size: 3,680 Bytes
f6e88dd b644e58 f6e88dd b644e58 f6e88dd b644e58 f6e88dd b644e58 f6e88dd b644e58 f6e88dd b644e58 f6e88dd | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 | import os
# Force Streamlit to use /tmp for config, cache, metrics
os.environ["HOME"] = "/tmp"
os.environ["XDG_CONFIG_HOME"] = "/tmp"
os.environ["XDG_CACHE_HOME"] = "/tmp"
os.environ["STREAMLIT_BROWSER_GATHERUSAGESTATS"] = "false"
# Create /tmp/.streamlit manually so Streamlit doesn't try to write to /
os.makedirs("/tmp/.streamlit", exist_ok=True)
# --- now import the rest ---
import json
from datetime import datetime
import streamlit as st
from supabase import create_client
from openai import OpenAI
# --- env vars ---
SUPABASE_URL = os.environ["SUPABASE_URL"]
SUPABASE_SERVICE_ROLE_KEY = os.environ["SUPABASE_SERVICE_ROLE_KEY"]
OPENAI_API_KEY = os.environ["OPENAI_API_KEY"]
sb = create_client(SUPABASE_URL, SUPABASE_SERVICE_ROLE_KEY)
client = OpenAI(api_key=OPENAI_API_KEY)
# --- UI ---
st.set_page_config(page_title="Email → LLM → Supabase Demo", layout="centered")
st.title("📧 Email → 🤖 LLM → 📦 Supabase")
st.caption("Paste an email, classify with LLM, store in Supabase, approve or decline the action.")
raw = st.text_area("Email body", height=200, placeholder="Paste email text here…")
col1, col2 = st.columns(2)
with col1:
subject = st.text_input("Subject (optional)")
with col2:
sender = st.text_input("From (optional)")
# --- helpers ---
def classify_email(email_text: str) -> dict:
prompt = f"""
Classify the email as one of: offer_update, rental_request, general.
Return STRICT JSON: {{"label":"...","confidence":0.0,"suggestion":"..."}}
Email:
{email_text}
"""
resp = client.chat.completions.create(
model="gpt-4o-mini",
temperature=0.2,
messages=[{"role": "user", "content": prompt}],
)
text = resp.choices[0].message.content or "{}"
try:
start, end = text.find("{"), text.rfind("}") + 1
return json.loads(text[start:end])
except Exception:
return {"label": "general", "confidence": 0.6, "suggestion": "Acknowledge receipt."}
def insert_email(raw_text, subject, from_addr):
res = sb.table("emails").insert(
{"raw_text": raw_text, "subject": subject, "from_addr": from_addr}
).execute()
return res.data[0]
def insert_classification(email_id, cls):
sb.table("classifications").insert({
"email_id": email_id,
"label": cls["label"],
"confidence": cls["confidence"],
"suggestion": cls["suggestion"],
}).execute()
def approve_action(email_id, suggestion):
sb.table("action_log").insert({
"email_id": email_id,
"action": suggestion,
"approved": True,
"approved_at": datetime.utcnow().isoformat(),
}).execute()
# --- flow ---
if st.button("🔎 Classify and Save", type="primary", disabled=not raw.strip()):
try:
email_row = insert_email(raw, subject, sender)
cls = classify_email(raw)
insert_classification(email_row["id"], cls)
st.session_state["current"] = {"email": email_row, "cls": cls}
st.success("✅ Email classified and saved.")
except Exception as e:
st.error(f"Error: {e}")
current = st.session_state.get("current")
if current:
st.subheader("Result")
st.write(f"**Label:** {current['cls']['label']}")
st.write(f"**Confidence:** {current['cls']['confidence']:.2f}")
st.write("**Suggested action:**")
st.code(current["cls"]["suggestion"])
if st.button("👍 Approve"):
try:
approve_action(current["email"]["id"], current["cls"]["suggestion"])
st.success("Approved and logged.")
except Exception as e:
st.error(f"Error approving: {e}")
if st.button("👎 Decline"):
st.info("Declined. No action logged.") |