Fixing "SyntaxError: Expected double-quoted property name in JSON at line X": RFC 8259 vs JS Object Literals — DevX Tool Blog

1. The Anatomy of "Expected double-quoted property name in JSON"

Few errors cause more immediate friction during API integration than the parser exception:

SyntaxError: Expected double-quoted property name in JSON at line 1 column 2 (char 1)
  or
JSON.parse: expected property name or '}' at line 1 column 2 of the JSON data

In modern web development, engineers work simultaneously with Python dictionaries, JavaScript object literals, YAML, and JSON. While JavaScript and Python permit single quotes ('key') or completely unquoted keys ({ host: "127.0.0.1" }), the JSON grammar specified in ECMA-404 and IETF RFC 8259 strictly mandates that every object key must be an ASCII double-quoted string literal.

2. Why ECMAScript Object Literals Are Not JSON

A frequent misunderstanding among frontend developers is assuming JSON is a subset of JavaScript syntax. Although JSON stands for JavaScript Object Notation, standard JSON grammar imposes much stricter lexical rules:

• Forbidden Single Quotes: { 'status': 'active' } is invalid JSON. Both keys and string values must use ".
• Forbidden Bare Identifiers: { status: "active" } works seamlessly in JavaScript code but fails instantly when parsed via JSON.parse().
• Forbidden Trailing Commas: { "id": 101, } breaks strictly compliant JSON parsers across Go, Python, and Rust.
• Forbidden Hexadecimal or Binary Literals: Numbers like 0xFF or 0b1010 cannot be parsed as valid JSON.

3. The Most Common Source: Stringified Python Dictionaries

When building backend microservices with Flask, Django, or FastAPI, developers often accidentally write:

# WRONG: Python str() outputs single-quoted string representation
user_payload = {"user_id": 42, "role": "admin"}
raw_string = str(user_payload)
# Result: "{'user_id': 42, 'role': 'admin'}" -> FAILS in JSON.parse()

To output compliant JSON from Python, always serialize with the native json standard library:

import json

# CORRECT: Standard RFC 8259 compliant JSON
json_string = json.dumps(user_payload)
# Result: '{"user_id": 42, "role": "admin"}' -> Passes all parsers

4. Automated Correction: Regex & AST Sanitizers

If your system ingests external configuration or AI-generated output that contains unquoted or single-quoted keys, you can sanitize it before passing it to strict tokenizers.

TypeScript / JavaScript Sanitizer Utility

/**
 * Transforms relaxed JavaScript object literal strings into valid JSON
 */
export function repairJsonString(raw: string): string {
  return raw
    // Replace unquoted keys: { key: "val" } -> { "key": "val" }
    .replace(/([{,]s*)([a-zA-Z0-9_$-]+)s*:/g, '$1"$2":')
    // Replace single-quoted keys and string values with double quotes
    .replace(/'([^'\\]*(\\.[^'\\]*)*)'/g, '"$1"')
    // Remove illegal trailing commas before closing braces/brackets
    .replace(/,s*([}\]])/g, '$1');
}

Validating With Python ast.literal_eval

If you must read a string generated by Python's repr() or str() safely without executing arbitrary bytecode, use ast.literal_eval:

import ast
import json

def convert_py_dict_string_to_json(raw_str: str) -> str:
    # Safely parses Python literals without eval() security vulnerabilities
    py_dict = ast.literal_eval(raw_str)
    return json.dumps(py_dict)

5. Production Best Practices

To prevent unquoted property name errors in CI/CD pipelines and web applications:

1. Enforce Content-Type Headers: Never parse responses as JSON if the upstream server returns text/plain or text/html.
2. Employ Local Verification: Before committing configurations or mocking API fixtures, pass the payload through our local, zero-telemetry JSON Formatter & Validator to ensure perfect conformance with ECMA-404.