Convert XML to JSON in Python While Preserving Inline Text Order

XML can mix text and nested tags within a single paragraph. If your output needs to display that paragraph with markup, preserve the order of the strings and child elements during conversion.

Worked example

Convert a paragraph with nested emphasis

import json
import xml.etree.ElementTree as ET

source = "<p>Ship <strong>three</strong> cases today.</p>"

def encode(element):
    parts = [element.text] if element.text is not None else []
    for child in element:
        parts.append(encode(child))
        if child.tail is not None:
            parts.append(child.tail)
    return {"tag": element.tag, "parts": parts}

print(json.dumps(encode(ET.fromstring(source)), indent=2))

Steps

  1. Save the example as mixed.py and run python3 mixed.py. ElementTree stores text before the first child as element.text and the text immediately after a child as that child's tail.
  2. Use an array of alternating text strings and nested element objects for the output. The example builds parts recursively and attaches each tail immediately after its child.
  3. Test a paragraph with no child and one with two inline tags. Check whether consumers can display the full sentence while preserving the emphasis order.

What to check

  • The parts array should contain 'Ship ', the strong node, and ' cases today.' in that order. Stripping the spaces from each part can join words when the sentence is reconstructed.
  • This mapping preserves inline text for the sample. XML comments, attributes, and namespace resolution need their own explicit rules when they matter.

References