json-result-formatting

JSON Result Formatting

Overview

Formatting retrieved data into standardized JSON output with answer lists and token tracking.

Use Cases

• Writing query results to JSON files
• Formatting answers as lists regardless of count
• Tracking API token consumption
• Creating consistent output files for downstream processing

Required Output Format

{
    "q1": {"answer": ["xxx"], "tokens": 123},
    "q2": {"answer": ["xxx", "yyy"], "tokens": 456},
    "q3": {"answer": [], "tokens": 789}
}

Code Examples

Initialize Result Container

import json

result = {
    "q1": {"answer": [], "tokens": 0},
    "q2": {"answer": [], "tokens": 0},
    "q3": {"answer": [], "tokens": 0}
}

Add Single Answer

def add_answer(result, question_key, answer_items, tokens=0):
    """Add answer as list (always list format)"""
    # Ensure answer_items is a list
    if isinstance(answer_items, str):
        answer_items = [answer_items]
    elif not isinstance(answer_items, list):
        answer_items = list(answer_items)

    result[question_key] = {
        "answer": answer_items,
        "tokens": int(tokens)
    }
    return result

# Usage
result = add_answer(result, "q1", ["eid_1e9356f5"], tokens=150)
result = add_answer(result, "q2", employee_ids_list, tokens=200)

Write to JSON File

import json

def write_result_file(result, filepath):
    """Write result to JSON file"""
    with open(filepath, 'w') as f:
        json.dump(result, f, indent=4)
    print(f"Results written to {filepath}")

# Usage
write_result_file(result, '/root/answer.json')

Track Token Usage (Without API)

import json

# For local data processing, estimate tokens
def estimate_tokens_from_text(text):
    """Rough estimation: ~4 characters per token"""
    return len(text) // 4

# Better: track actual API usage
def track_tokens(usage_dict):
    """Track from API response"""
    if hasattr(usage_dict, 'input_tokens'):
        return usage_dict.input_tokens + usage_dict.output_tokens
    return 0