Spaces:
Sleeping
Sleeping
import os | |
os.environ["OMP_NUM_THREADS"] = "1" | |
import re | |
import json | |
import datetime as dt | |
from urllib.parse import urlparse | |
import xml.etree.ElementTree as ET | |
import requests | |
import gradio as gr | |
from utils import extract_kyc_fields | |
# ------------------ HARD-CODED SALESFORCE CREDS (as requested) ------------------ | |
SF_USERNAME = "[email protected]" | |
SF_PASSWORD = "Lic@2025" | |
SF_SECURITY_TOKEN = "AmmfRcd6IiYaRtSGntBnzNMQU" | |
SF_DOMAIN = "login" # "login" for prod, "test" for sandbox | |
SF_API_VERSION = "60.0" # safe current Partner/REST version | |
# ------------------------------------------------------------------------------- | |
# ---------- helpers ---------- | |
def _parse_birthdate(dob_text: str): | |
""" | |
Normalize DOB to YYYY-MM-DD (Salesforce Date). | |
Supports dd/mm/yyyy, dd-mm-yyyy, dd.mm.yyyy, yyyy-mm-dd, or just YYYY (mapped to mid-year). | |
""" | |
if not dob_text or dob_text == "Not found": | |
return None | |
s = dob_text.strip() | |
m = re.fullmatch(r"(\d{4})-(\d{2})-(\d{2})", s) | |
if m: | |
y, mo, d = map(int, m.groups()) | |
try: | |
return dt.date(y, mo, d).isoformat() | |
except ValueError: | |
return None | |
m = re.fullmatch(r"(\d{2})[./-](\d{2})[./-](\d{4})", s) | |
if m: | |
d, mo, y = map(int, m.groups()) | |
try: | |
return dt.date(y, mo, d).isoformat() | |
except ValueError: | |
return None | |
m = re.fullmatch(r"(19|20)\d{2}", s) | |
if m: | |
y = int(s) | |
try: | |
return dt.date(y, 6, 15).isoformat() | |
except ValueError: | |
return None | |
return None | |
def _soap_login(username: str, password: str, token: str, domain: str = "login", api_version: str = SF_API_VERSION): | |
""" | |
Login via Salesforce Partner SOAP API (no connected app needed). | |
Returns (instance_url, access_token) on success; raises on error. | |
""" | |
endpoint = f"https://{domain}.salesforce.com/services/Soap/u/{api_version}" | |
combined_password = f"{password}{token}" | |
envelope = f"""<?xml version="1.0" encoding="utf-8" ?> | |
<env:Envelope xmlns:xsd="http://www.w3.org/2001/XMLSchema" | |
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" | |
xmlns:env="http://schemas.xmlsoap.org/soap/envelope/"> | |
<env:Body> | |
<n1:login xmlns:n1="urn:partner.soap.sforce.com"> | |
<n1:username>{username}</n1:username> | |
<n1:password>{combined_password}</n1:password> | |
</n1:login> | |
</env:Body> | |
</env:Envelope>""" | |
headers = { | |
"Content-Type": "text/xml; charset=UTF-8", | |
"SOAPAction": "login", | |
} | |
resp = requests.post(endpoint, data=envelope.encode("utf-8"), headers=headers, timeout=30) | |
if resp.status_code != 200: | |
raise RuntimeError(f"SOAP login HTTP {resp.status_code}: {resp.text[:400]}") | |
# Parse XML to find sessionId and serverUrl | |
try: | |
root = ET.fromstring(resp.text) | |
except ET.ParseError as e: | |
raise RuntimeError(f"SOAP login parse error: {e}") | |
# Find tags by localname to avoid namespace headaches | |
session_id = None | |
server_url = None | |
for elem in root.iter(): | |
tag = elem.tag.split('}', 1)[-1] # strip namespace | |
if tag == "sessionId": | |
session_id = elem.text | |
elif tag == "serverUrl": | |
server_url = elem.text | |
if not session_id or not server_url: | |
raise RuntimeError(f"SOAP login failed: sessionId/serverUrl not found. Raw: {resp.text[:600]}") | |
parsed = urlparse(server_url) | |
instance_url = f"{parsed.scheme}://{parsed.netloc}" | |
return instance_url, session_id | |
def _rest_create(instance_url: str, access_token: str, object_api: str, payload: dict, api_version: str = SF_API_VERSION): | |
""" | |
Create a record via REST API. Returns a dict with success flag and details. | |
""" | |
url = f"{instance_url}/services/data/v{api_version}/sobjects/{object_api}" | |
headers = { | |
"Authorization": f"Bearer {access_token}", | |
"Content-Type": "application/json", | |
} | |
resp = requests.post(url, headers=headers, data=json.dumps(payload), timeout=30) | |
try: | |
body = resp.json() | |
except Exception: | |
body = None | |
if 200 <= resp.status_code < 300: | |
rec_id = body.get("id") if isinstance(body, dict) else None | |
return {"success": True, "id": rec_id, "status_code": resp.status_code, "url": url, "response_json": body} | |
else: | |
# Salesforce error lists come back as JSON arrays usually | |
return { | |
"success": False, | |
"status_code": resp.status_code, | |
"url": url, | |
"response_json": body, | |
"response_text": resp.text, | |
} | |
def sf_push_kyc_record(ocr_results: dict): | |
""" | |
Combine Aadhaar + PAN into one KYC_Record__c and create via REST. | |
Fields expected in SF (API names): | |
Aadhaar_Number__c, Aadhaar_Name__c, Aadhaar_DOB__c (Date) | |
PAN_Number__c, Pan_Name__c, Pan_DOB__c (Date) | |
""" | |
a = ocr_results.get("aadhaar") or {} | |
p = ocr_results.get("pan") or {} | |
aadhaar_number = a.get("aadhaar_number") if (a.get("card_type") == "AADHAAR") else None | |
aadhaar_name = a.get("name") if (a.get("card_type") == "AADHAAR") else None | |
aadhaar_dob = _parse_birthdate(a.get("dob")) if (a.get("card_type") == "AADHAAR") else None | |
pan_number = p.get("pan_number") if (p.get("card_type") == "PAN") else None | |
pan_name = p.get("name") if (p.get("card_type") == "PAN") else None | |
pan_dob = _parse_birthdate(p.get("dob")) if (p.get("card_type") == "PAN") else None | |
payload = { | |
"Aadhaar_Number__c": aadhaar_number, | |
"Aadhaar_Name__c": aadhaar_name, | |
"Aadhaar_DOB__c": aadhaar_dob, | |
"PAN_Number__c": pan_number, | |
"Pan_Name__c": pan_name, | |
"Pan_DOB__c": pan_dob, | |
} | |
payload = {k: v for k, v in payload.items() if v is not None} | |
# 1) SOAP login | |
instance_url, access_token = _soap_login(SF_USERNAME, SF_PASSWORD, SF_SECURITY_TOKEN, SF_DOMAIN, SF_API_VERSION) | |
# 2) REST create | |
return _rest_create(instance_url, access_token, "KYC_Record__c", payload, SF_API_VERSION) | |
# ---------- gradio callback ---------- | |
def process_documents(aadhaar_file, pan_file, push_to_sf): | |
""" | |
- Runs OCR on Aadhaar and PAN separately. | |
- Optionally pushes a single KYC_Record__c to Salesforce via SOAP+REST (no simple-salesforce). | |
""" | |
results = {"aadhaar": None, "pan": None} | |
if not aadhaar_file and not pan_file: | |
return {"error": "Please upload at least one file (Aadhaar and/or PAN)."} | |
# OCR Aadhaar | |
if aadhaar_file: | |
try: | |
res = extract_kyc_fields(aadhaar_file.name) | |
res["source_file"] = os.path.basename(aadhaar_file.name) | |
results["aadhaar"] = res | |
except Exception as e: | |
results["aadhaar"] = {"error": f"Aadhaar OCR failed: {str(e)}", "card_type": "UNKNOWN"} | |
# OCR PAN | |
if pan_file: | |
try: | |
res = extract_kyc_fields(pan_file.name) | |
res["source_file"] = os.path.basename(pan_file.name) | |
results["pan"] = res | |
except Exception as e: | |
results["pan"] = {"error": f"PAN OCR failed: {str(e)}", "card_type": "UNKNOWN"} | |
output = {"ocr": results} | |
if push_to_sf: | |
try: | |
created = sf_push_kyc_record(results) | |
output["salesforce"] = {"pushed": created.get("success", False), **created} | |
except Exception as e: | |
output["salesforce"] = {"pushed": False, "error": {"type": e.__class__.__name__, "message": str(e)}} | |
return output | |
# ---------- UI ---------- | |
with gr.Blocks(title="Smart KYC OCR → Salesforce (KYC_Record__c)") as demo: | |
gr.Markdown( | |
""" | |
# 🧾 Smart KYC OCR → Salesforce | |
Upload **Aadhaar** and **PAN** in separate boxes, then (optional) push one **KYC_Record__c**. | |
""" | |
) | |
with gr.Row(): | |
with gr.Column(scale=1): | |
aadhaar_uploader = gr.File( | |
label="📤 Aadhaar Upload", | |
file_types=[".jpg", ".jpeg", ".png"] | |
) | |
with gr.Column(scale=1): | |
pan_uploader = gr.File( | |
label="📤 PAN Upload", | |
file_types=[".jpg", ".jpeg", ".png"] | |
) | |
push_to_sf = gr.Checkbox(label="Push to Salesforce (create KYC_Record__c)", value=False) | |
submit_btn = gr.Button("🔍 Extract KYC Info", variant="primary") | |
output_json = gr.JSON(label="📋 Output (OCR + Salesforce)") | |
submit_btn.click( | |
fn=process_documents, | |
inputs=[aadhaar_uploader, pan_uploader, push_to_sf], | |
outputs=output_json, | |
) | |
gr.Markdown("---") | |
gr.Markdown( | |
""" | |
You’ll see `status_code`, `url`, and Salesforce’s `response_json` on errors (e.g., wrong field API name or permissions). | |
""" | |
) | |
# Keep `demo` at module level for Spaces too | |
if __name__ == "__main__": | |
demo.launch() | |