Convert Salesforce JSON to CSV

Convert Salesforce SOQL query result to CSV

import json
import pandas as pd

with open("sf_contacts.json") as f:
    data = json.load(f)

records = data["records"]

def flatten_sf_record(record):
    """Flatten a Salesforce record, dropping attributes and promoting relationship fields."""
    out = {}
    for key, value in record.items():
        if key == "attributes":
            continue  # drop internal metadata
        if isinstance(value, dict) and "attributes" in value:
            # Relationship field — flatten it with the parent key as prefix
            for sub_key, sub_val in value.items():
                if sub_key == "attributes":
                    continue
                out[f"{key}_{sub_key}"] = sub_val
        else:
            out[key] = value
    return out

flat_records = [flatten_sf_record(r) for r in records]

df = pd.DataFrame(flat_records)

# Clean up date columns — Salesforce uses ISO 8601 with timezone offset
for col in df.select_dtypes(include="object").columns:
    if df[col].str.match(r"\d{4}-\d{2}-\d{2}T", na=False).any():
        df[col] = pd.to_datetime(df[col], errors="coerce", utc=True).dt.strftime("%Y-%m-%d %H:%M")

df.to_csv("contacts.csv", index=False, encoding="utf-8-sig")
print(f"Exported {len(df)} contacts")

Query and export using simple_salesforce (handles auth and pagination)

from simple_salesforce import Salesforce
import pandas as pd

sf = Salesforce(
    username="user@example.com",
    password="your_password",
    security_token="your_token",
)

# SOQL query — select the fields you need explicitly
result = sf.query_all("""
    SELECT Id, FirstName, LastName, Email, Title, Department,
           Phone, Account.Name, Account.Industry,
           CreatedDate, LastModifiedDate
    FROM Contact
    WHERE IsDeleted = false
    ORDER BY LastName
""")

print(f"Total records: {result['totalSize']}")

def flatten_sf_record(record):
    out = {}
    for key, value in record.items():
        if key == "attributes":
            continue
        if isinstance(value, dict) and "attributes" in value:
            for sub_key, sub_val in value.items():
                if sub_key != "attributes":
                    out[f"{key}_{sub_key}"] = sub_val
        else:
            out[key] = value
    return out

rows = [flatten_sf_record(r) for r in result["records"]]
df = pd.DataFrame(rows)
df.to_csv("sf_contacts.csv", index=False, encoding="utf-8-sig")
print(f"Exported {len(df)} rows")

Export Opportunities with multi-level relationships

from simple_salesforce import Salesforce
import pandas as pd

sf = Salesforce(username="...", password="...", security_token="...")

result = sf.query_all("""
    SELECT Id, Name, StageName, Amount, CloseDate,
           Probability, Type, LeadSource,
           Account.Name, Account.Industry, Account.BillingCity,
           Owner.Name, Owner.Email
    FROM Opportunity
    WHERE IsClosed = false
""")

def flatten_sf_record(record):
    out = {}
    for key, value in record.items():
        if key == "attributes":
            continue
        if isinstance(value, dict) and "attributes" in value:
            for sub_key, sub_val in value.items():
                if sub_key != "attributes":
                    out[f"{key}_{sub_key}"] = sub_val
        else:
            out[key] = value
    return out

df = pd.DataFrame([flatten_sf_record(r) for r in result["records"]])

# Amount is already a float in Salesforce (unlike Stripe's cents)
df["Amount"] = pd.to_numeric(df["Amount"], errors="coerce")
df["CloseDate"] = pd.to_datetime(df["CloseDate"], errors="coerce").dt.strftime("%Y-%m-%d")

df.to_csv("open_opportunities.csv", index=False, encoding="utf-8-sig")
print(f"Exported {len(df)} opportunities")