import json

import difflib

def diff_json_datasets(dataset1_path, dataset2_path, max_lines=100):

"""

Diffs two JSON datasets, limiting the number of lines shown.

Args:

dataset1_path (str): Path to the first JSON dataset file.

dataset2_path (str): Path to the second JSON dataset file.

max_lines (int): Maximum number of lines to display in the diff.

"""

try:

with open(dataset1_path, 'r') as f1:

data1 = json.load(f1)

with open(dataset2_path, 'r') as f2:

data2 = json.load(f2)

except FileNotFoundError as e:

print(f"Error: File not found: {e}")

return

except json.JSONDecodeError as e:

print(f"Error: Invalid JSON: {e}")

return

# Convert data to strings for diffing

str1 = json.dumps(data1, indent=4, sort_keys=True)

str2 = json.dumps(data2, indent=4, sort_keys=True)

# Use difflib to find differences

diff = difflib.unified_diff(str1.splitlines(), str2.splitlines(), fromfile=dataset1_path, tofile=dataset2_path)

# Limit the number of lines displayed

for line in diff:

if len(line) > max_lines:

break

print(line)

if __name__ == '__main__':

# Example usage: Replace with your file paths

diff_json_datasets('dataset1.json', 'dataset2.json', max_lines=50)