ai-apply / FindEmailWorkFlowV2.py
sk31415's picture
AiApply
5e54e45
"""
Contact deduplication workflow using per-user email history.
"""
import json
def is_email_okay_to_send(email, domains):
"""
Check if an email is safe to send (not already contacted at this domain).
Args:
email: Email address to check
domains: Set of already-contacted domains
Returns:
bool: True if okay to send, False otherwise
"""
if "@" not in email:
return False
domain = email.split("@")[1]
return domain not in domains
def filter_contacts(contacts, user_emails_sent, user_domains_contacted):
"""
Filter contacts based on user's email history.
Args:
contacts: List of dicts with company_name, contact_name, email_address
user_emails_sent: Set of emails already sent by this user
user_domains_contacted: Set of domains already contacted by this user
Returns:
list: Filtered list of contacts that are safe to contact
"""
cleaned_contacts = []
for contact in contacts:
email = contact.get("email_address", "")
if is_email_okay_to_send(email, user_domains_contacted):
cleaned_contacts.append(contact)
return cleaned_contacts
def main(contacts, user_emails_sent=None, user_domains_contacted=None):
"""
Main workflow function - filters contacts based on user's history.
Args:
contacts: List of contact dicts from contact finder
user_emails_sent: Set of emails already sent by this user (optional)
user_domains_contacted: Set of domains already contacted by this user (optional)
Returns:
list: Deduplicated contacts ready for email generation
"""
if user_emails_sent is None:
user_emails_sent = set()
if user_domains_contacted is None:
user_domains_contacted = set()
return filter_contacts(contacts, user_emails_sent, user_domains_contacted)
if __name__ == "__main__":
# Test with sample data
sample_contacts = [
{
"company_name": "Test Company 1",
"contact_name": "John Doe",
"email_address": "john@testcompany1.com",
},
{
"company_name": "Test Company 2",
"contact_name": None,
"email_address": "info@testcompany2.com",
},
]
result = main(sample_contacts)
print(f"Cleaned contacts: {json.dumps(result, indent=2)}")