56 lines
1.9 KiB
Python
56 lines
1.9 KiB
Python
import json
|
|
import csv
|
|
import re
|
|
|
|
# Parse the SERP output file (it's JSON content inside a text file, usually)
|
|
# The previous view_file showed it's valid JSON.
|
|
|
|
input_file = r'C:\Users\a931627\.gemini\antigravity\brain\6060ab5d-4406-4d40-803f-c8d1df8bb430\.system_generated\steps\141\output.txt'
|
|
output_csv = 'cologne_duesseldorf_data/cologne_leads.csv'
|
|
|
|
def parse_serp():
|
|
with open(input_file, 'r', encoding='utf-8') as f:
|
|
data = json.load(f)
|
|
|
|
leads = []
|
|
items = data.get('items', [])
|
|
|
|
email_regex = re.compile(r'[\w\.-]+@[\w\.-]+\.\w+')
|
|
|
|
for item in items:
|
|
if item.get('type') == 'organic':
|
|
desc = item.get('description', '')
|
|
title = item.get('title', '')
|
|
snippet = item.get('pre_snippet', '')
|
|
full_text = f"{title} {desc} {snippet}"
|
|
|
|
emails = email_regex.findall(full_text)
|
|
for email in emails:
|
|
email = email.rstrip('.')
|
|
|
|
# Use title or domain as Innung name
|
|
innung_name = item.get('website_name') or item.get('domain') or title
|
|
|
|
if any(l['Email'] == email for l in leads):
|
|
continue
|
|
|
|
leads.append({
|
|
'Firm/Innung': innung_name,
|
|
'Contact': "N/A",
|
|
'Email': email,
|
|
'Phone': "N/A",
|
|
'Region': 'Köln'
|
|
})
|
|
|
|
with open(output_csv, 'w', newline='', encoding='utf-8') as f:
|
|
writer = csv.DictWriter(f, fieldnames=['Firm/Innung', 'Contact', 'Email', 'Phone', 'Region'])
|
|
writer.writeheader()
|
|
writer.writerows(leads)
|
|
|
|
print(f"Extracted {len(leads)} leads from Cologne SERP.")
|
|
for l in leads:
|
|
print(f"{l['Firm/Innung']}: {l['Email']}")
|
|
|
|
if __name__ == "__main__":
|
|
parse_serp()
|