name: frappe-syntax-whitelisted description: > Use when creating Frappe Whitelisted Methods (Python API endpoints) for v14/v15/v16. Covers @frappe.whitelist() decorator, frappe.call/frm.call invocations, permission checks, error handling, response formats, and client-server communication. Keywords: whitelisted, API endpoint, frappe.call, frm.call, REST API, @frappe.whitelist, allow_guest, API endpoint example, frappe.whitelist syntax, how to expose function. license: MIT compatibility: "Claude Code, Claude.ai Projects, Claude API. Frappe v14-v16." metadata: author: OpenAEC-Foundation version: "2.0"

Frappe Syntax: Whitelisted Methods

Whitelisted methods expose Python functions as HTTP API endpoints via /api/method/.

Quick Reference

import frappe
from frappe import _

# Authenticated endpoint (default)
@frappe.whitelist()
def get_customer_summary(customer):
    frappe.has_permission("Customer", "read", throw=True)
    return frappe.get_doc("Customer", customer).as_dict()

# Public endpoint — ALWAYS validate input thoroughly
@frappe.whitelist(allow_guest=True, methods=["POST"])
def submit_contact(name, email, message):
    if not name or not email:
        frappe.throw(_("Name and email required"), frappe.ValidationError)
    return {"success": True}

# Controller method — called via frm.call('method_name')
class SalesOrder(Document):
    @frappe.whitelist()
    def calculate_taxes(self, include_shipping=False):
        return {"tax": self.grand_total * 0.21}

Endpoint URL: /api/method/myapp.module.function_name

Decorator Signature [v14+]

@frappe.whitelist(
    allow_guest=False,   # True = accessible without login
    xss_safe=False,      # True = do NOT escape HTML in response
    methods=None,        # ["GET"], ["POST"], or ["GET","POST"] — default: all
    force_types=None     # True = require type annotations [v15+]
)

Full details: decorator-options.md

Decision Tree

What kind of endpoint?
|
+-- Standalone API (utility, integration, dashboard)?
|   --> @frappe.whitelist() on a module-level function
|   --> Call via: frappe.call('myapp.api.function')
|   --> URL: /api/method/myapp.api.function
|
+-- Document-specific action?
|   --> @frappe.whitelist() on a Document class method
|   --> Call via: frm.call('method_name')
|   --> URL: /api/method/run_doc_method (internal)
|
+-- Server Script (no-code)?
    --> Use Server Script DocType instead (no decorator needed)

Who may call the API?
|
+-- Anyone (including guests)?
|   --> allow_guest=True + thorough input validation + rate limiting
|
+-- Logged-in users only?
    +-- Specific role? --> frappe.only_for("RoleName")
    +-- DocType-level? --> frappe.has_permission(doctype, ptype, throw=True)
    +-- Document-level? --> frappe.has_permission(doctype, ptype, doc, throw=True)

Which HTTP methods?
|
+-- Read only? --> methods=["GET"]
+-- Write only? --> methods=["POST"]
+-- Both? --> methods=["GET","POST"] or default

Permission Patterns

ALWAYS check permissions inside every whitelisted method. The @frappe.whitelist() decorator only verifies the user is logged in — it does NOT check DocType or document-level permissions.

# DocType-level permission (throw=True raises PermissionError automatically)
@frappe.whitelist()
def get_orders():
    frappe.has_permission("Sales Order", "read", throw=True)
    return frappe.get_all("Sales Order", limit=20)

# Document-level permission
@frappe.whitelist()
def get_order(name):
    frappe.has_permission("Sales Order", "read", name, throw=True)
    return frappe.get_doc("Sales Order", name).as_dict()

# Role-based restriction
@frappe.whitelist()
def admin_action():
    frappe.only_for("System Manager")  # throws if user lacks role
    return {"secret": "data"}

Full patterns: permission-patterns.md

Parameter Handling

Parameters arrive as strings from HTTP requests. ALWAYS convert explicitly.

@frappe.whitelist()
def calculate(amount, quantity, items=None):
    amount = float(amount)          # ALWAYS cast numeric params
    quantity = int(quantity)
    if isinstance(items, str):      # ALWAYS parse JSON strings
        items = frappe.parse_json(items)
    return amount * quantity

Access all request parameters via frappe.form_dict:

@frappe.whitelist()
def dynamic_handler():
    all_params = frappe.form_dict
    customer = frappe.form_dict.get("customer")

Type Annotations [v15+]

Frappe v15+ validates type annotations automatically at request time via Pydantic:

@frappe.whitelist()
def get_orders(customer: str, limit: int = 10, active: bool = True) -> dict:
    # Frappe auto-validates: limit MUST be convertible to int
    return {"orders": frappe.get_all("Sales Order", limit=limit)}

force_types and require_type_annotated_api_methods [v15+]

• @frappe.whitelist(force_types=True) — EVERY parameter MUST have a type annotation
• App-level enforcement via hooks.py: require_type_annotated_api_methods = 1
• Missing annotations raise FrappeTypeError

Full details: parameter-handling.md

Client Calls

frappe.call(): Standalone APIs

// Promise-based (ALWAYS prefer this)
frappe.call({
    method: 'myapp.api.get_summary',
    args: { customer: 'CUST-001' },
    freeze: true,
    freeze_message: __('Loading...')
}).then(r => {
    console.log(r.message);  // return value is in r.message
}).catch(err => {
    frappe.show_alert({ message: __('Error'), indicator: 'red' });
});

frm.call(): Controller Methods

frm.call('calculate_taxes', { include_shipping: true })
    .then(r => frm.set_value('tax_amount', r.message.tax_amount));

REST API (External Clients)

# Token auth (ALWAYS use for external integrations)
curl -H "Authorization: token api_key:api_secret" \
     -H "Content-Type: application/json" \
     -X POST https://site.com/api/method/myapp.api.create_order \
     -d '{"customer": "CUST-001"}'

Full patterns: client-calls.md

Error Handling

@frappe.whitelist()
def process_order(order_id):
    if not order_id:
        frappe.throw(_("Order ID required"), frappe.ValidationError)

    if not frappe.has_permission("Sales Order", "write", order_id):
        frappe.throw(_("Not permitted"), frappe.PermissionError)

    try:
        result = heavy_operation(order_id)
        return {"success": True, "data": result}
    except Exception:
        frappe.log_error(frappe.get_traceback(), "process_order")
        frappe.throw(_("Operation failed. Contact support."))

Full patterns: error-handling.md

Response Patterns

# Return value auto-wraps as {"message": <return_value>}
@frappe.whitelist()
def get_data():
    return {"key": "value"}   # Client receives: {"message": {"key": "value"}}

# Custom HTTP status
@frappe.whitelist()
def create_item(data):
    doc = frappe.get_doc(data).insert()
    frappe.local.response["http_status_code"] = 201
    return {"name": doc.name}

# File download
@frappe.whitelist()
def download_report(name):
    content = generate_pdf(name)
    frappe.response.filename = f"{name}.pdf"
    frappe.response.filecontent = content
    frappe.response.type = "download"

Full patterns: response-patterns.md

Rate Limiting [v14+]

from frappe.rate_limiter import rate_limit

@frappe.whitelist(allow_guest=True)
@rate_limit(limit=5, seconds=60)  # 5 requests per 60 seconds per IP
def public_endpoint():
    return {"status": "ok"}

rate_limit signature:

rate_limit(key=None, limit=5, seconds=86400, methods="ALL", ip_based=True)

ALWAYS apply @rate_limit on allow_guest=True endpoints to prevent abuse.

Version Differences

Critical Rules

1. NEVER skip permission checks — @frappe.whitelist() only confirms login, not authorization
2. NEVER use user input in raw SQL — ALWAYS use parameterized queries or ORM
3. NEVER leak stack traces — log with frappe.log_error(), show generic messages
4. ALWAYS validate input types — parameters arrive as strings from HTTP
5. ALWAYS apply @rate_limit on guest endpoints — prevents abuse
6. NEVER use ignore_permissions=True without a preceding role check
7. ALWAYS use JSON.stringify() for complex JS args — arrays and objects

Full anti-patterns: anti-patterns.md

Security Checklist

For EVERY whitelisted method, verify:

• Permission check present (frappe.has_permission() or frappe.only_for())
• Input validated (types, ranges, formats)
• SQL queries parameterized (NEVER string interpolation)
• Error messages contain no internal details
• allow_guest=True only with explicit reason + rate limiting
• ignore_permissions=True only with preceding role check
• HTTP methods restricted where possible
• Response contains only necessary fields (no sensitive data leaks)

Reference Files