afv-library/skills/developing-datacloud-code-extension/quick-reference.md

Data Cloud Code Extension - Quick Reference

Command Cheat Sheet

Initialize Project

# Create script project
sf data-code-extension init <directory> --code-type script

# Create function project
sf data-code-extension init <directory> --code-type function

# Examples
sf data-code-extension init . --code-type script
sf data-code-extension init my-transform --code-type script

Scan for Permissions

# Basic scan
sf data-code-extension scan ./payload/entrypoint.py

# Preview without saving
sf data-code-extension scan ./payload/entrypoint.py --dry-run

# Custom config location
sf data-code-extension scan ./payload/entrypoint.py --config ./custom-config.json

# Skip requirements.txt
sf data-code-extension scan ./payload/entrypoint.py --no-requirements

Run Locally

# Basic run
sf data-code-extension run ./payload/entrypoint.py --target-org <org_alias>

# With custom config
sf data-code-extension run ./payload/entrypoint.py -o <org_alias> -c custom-config.json

# Examples
sf data-code-extension run ./payload/entrypoint.py --target-org afvibe
sf data-code-extension run ./payload/entrypoint.py -o afvibe

Deploy

# Minimal deployment (MUST include --path ./payload)
sf data-code-extension deploy \
  --target-org <org_alias> \
  --name <name> \
  --package-version <version> \
  --description "<description>" \
  --path ./payload

# Full options
sf data-code-extension deploy \
  --target-org <org_alias> \
  --name <name> \
  --package-version <version> \
  --description "<description>" \
  --cpu-size <CPU_L|CPU_XL|CPU_2XL|CPU_4XL> \
  --path ./payload

# Examples (CRITICAL: Always include --path ./payload)
sf data-code-extension deploy \
  --target-org afvibe \
  --name Employee_Upper \
  --package-version 1.0.0 \
  --description "Uppercase employee positions" \
  --path ./payload

sf data-code-extension deploy \
  -o afvibe \
  -n Employee_Upper \
  --package-version 1.0.0 \
  --description "Uppercase employee positions" \
  --cpu-size CPU_4XL \
  --path ./payload

Common Workflows

New Project from Scratch

# 1. Create directory
mkdir my-transform && cd my-transform

# 2. Initialize
sf data-code-extension init . --code-type script

# 3. Edit entrypoint.py
# (Add your transformation code)

# 4. Scan
sf data-code-extension scan ./payload/entrypoint.py

# 5. Test
sf data-code-extension run ./payload/entrypoint.py --target-org afvibe

# 6. Deploy (MUST include --path ./payload)
sf data-code-extension deploy \
  --target-org afvibe \
  --name MyTransform \
  --package-version 1.0.0 \
  --description "Uppercase employee positions" \
  --path ./payload

Update Existing Code Extension

# 1. Edit entrypoint.py

# 2. Re-scan
sf data-code-extension scan ./payload/entrypoint.py

# 3. Test
sf data-code-extension run ./payload/entrypoint.py -o afvibe

# 4. Deploy with new version (include --path ./payload)
sf data-code-extension deploy \
  -o afvibe \
  -n MyTransform \
  --package-version 1.1.0 \
  --description "Uppercase employee positions" \
  --path ./payload

Python Code Patterns

Read/Write DLO

from datacustomcode import Client

client = Client()

# Read
df = client.read_dlo('Employee__dll')

# Transform
df['new_field'] = df['old_field'].str.upper()

# Write (modes: 'overwrite', 'append')
client.write_to_dlo('Output__dll', df, 'overwrite')

Read/Write DMO

# Read
df = client.read_dmo('EmployeeDMO')

# Write (modes: 'upsert', 'insert')
client.write_to_dmo('EmployeeDMO', df, 'upsert')

Multiple DLO Operations

# Read multiple
employees = client.read_dlo('Employee__dll')
departments = client.read_dlo('Department__dll')

# Join
merged = employees.merge(departments, on='dept_id')

# Write multiple
client.write_to_dlo('Enriched__dll', merged, 'overwrite')
client.write_to_dmo('EmployeeDMO', merged, 'upsert')

Data Transformations

import pandas as pd

# Filter
active = df[df['status'] == 'Active']

# Computed column
df['full_name'] = df['first'] + ' ' + df['last']

# Aggregate
summary = df.groupby('dept')['salary'].mean()

# Conditional
df['grade'] = df['position'].apply(
    lambda x: 'Senior' if 'VP' in x else 'Junior'
)

Option Reference

--code-type

• script - Batch transformation (default)
• function - Real-time function

--cpu-size

• CPU_L - Small datasets (< 1M records)
• CPU_XL - Medium datasets (1M-5M)
• CPU_2XL - Large datasets (5M-10M) [default]
• CPU_4XL - Very large (> 10M records)

Write Modes

• overwrite - Replace all data
• append - Add to existing data
• upsert - Update or insert (DMO only)
• insert - Insert only (DMO only)

Troubleshooting Quick Fixes

# Plugin not found
sf plugins install @salesforce/plugin-data-codeextension

# Python SDK missing
pip install salesforce-data-customcode

# Verify Python version (must be 3.11.x)
python --version

# Org not connected
sf org login web --alias <org_alias>

# Config missing
sf data-code-extension scan ./payload/entrypoint.py

# Docker not running (for deploy)
# Start Docker Desktop

File Structure

my-project/
├── payload/
│   ├── entrypoint.py      # Main code
│   ├── config.json        # Auto-generated permissions
│   └── requirements.txt   # Auto-generated dependencies
└── README.md

config.json Format

{
  "version": "1.0",
  "permissions": {
    "read": ["Employee__dll", "Department__dll"],
    "write": ["Enriched__dll"]
  },
  "resources": {
    "cpu_size": "CPU_2XL"
  }
}

Common Errors

Error	Quick Fix
Plugin not found	sf plugins install @salesforce/plugin-data-codeextension
Python SDK missing	pip install salesforce-data-customcode
Wrong Python version	Use pyenv to install 3.11.0
Org not connected	sf org login web --alias <alias>
Config missing	Run scan command
DLO not found	Check DLO name, use DLO Schema skill
Docker error	Start Docker Desktop

Deployment Checklist

• Code written in entrypoint.py
• Scanned for permissions
• Tested locally
• Version number decided
• Description added
• CPU size chosen
• Docker running
• Org authenticated

Next Steps After Deploy

1. Go to Data Cloud in Salesforce UI
2. Navigate to Code Extensions
3. Find your deployment
4. Click "Run Now" to test
5. Schedule for recurring execution
6. Monitor execution logs

Quick Examples

Example 1: Simple Transform

from datacustomcode import Client
client = Client()

df = client.read_dlo('Employee__dll')
df['upper_pos'] = df['position'].str.upper()
client.write_to_dlo('Employee_Upper__dll', df, 'overwrite')

Example 2: Filter and Write

from datacustomcode import Client
client = Client()

df = client.read_dlo('Employee__dll')
managers = df[df['position'].str.contains('Manager')]
client.write_to_dlo('Managers__dll', managers, 'overwrite')

Example 3: Join Two DLOs

from datacustomcode import Client
client = Client()

employees = client.read_dlo('Employee__dll')
departments = client.read_dlo('Department__dll')

merged = employees.merge(departments, left_on='dept_id', right_on='id')
client.write_to_dlo('Employee_With_Dept__dll', merged, 'overwrite')

Resources

• Plugin: https://github.com/salesforcecli/plugin-data-code-extension
• Python SDK: https://github.com/forcedotcom/datacloud-customcode-python-sdk
• Data Cloud Docs: https://help.salesforce.com/s/articleView?id=sf.c360_a_intro.htm