XML to JSON Converter
This utility converts XML files into JSON format. It parses the XML structure using xml.etree.ElementTree and recursively transforms it into a Python dictionary, which is then serialized to JSON.
Modules Used:
xml.etree.ElementTree: To parse the XML file.- json: To write the JSON output.
- argparse: To handle command-line arguments.
The Code
Save this as xml2json.py.
import xml.etree.ElementTree as ET
import json
import argparse
import sys
def element_to_dict(element):
"""
Recursively converts an ElementTree Element to a dictionary.
"""
node = {}
# 1. Attributes
if element.attrib:
node.update(("@" + k, v) for k, v in element.attrib.items())
# 2. Text Content
if element.text and element.text.strip():
node["#text"] = element.text.strip()
# 3. Children
for child in element:
child_data = element_to_dict(child)
# Handle multiple children with the same tag (create list)
if child.tag in node:
if isinstance(node[child.tag], list):
node[child.tag].append(child_data)
else:
node[child.tag] = [node[child.tag], child_data]
else:
node[child.tag] = child_data
# Simplification: If node only has text, return the text directly
if len(node) == 1 and "#text" in node:
return node["#text"]
return node
def convert_xml_to_json(xml_file, json_file):
try:
# Parse XML
tree = ET.parse(xml_file)
root = tree.getroot()
# Convert to dict (wrap root tag)
data = {root.tag: element_to_dict(root)}
# Write JSON
with open(json_file, 'w', encoding='utf-8') as f:
json.dump(data, f, indent=4)
print(f"Successfully converted '{xml_file}' to '{json_file}'")
except ET.ParseError as e:
print(f"Error parsing XML: {e}")
except FileNotFoundError:
print(f"Error: File '{xml_file}' not found.")
except Exception as e:
print(f"An error occurred: {e}")
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="XML to JSON Converter")
parser.add_argument("input", help="Input XML file path")
parser.add_argument("output", help="Output JSON file path")
args = parser.parse_args()
convert_xml_to_json(args.input, args.output)Usage
python xml2json.py data.xml data.json