Rasa NLU training data migrator from JSON format to Markdown

rasa-json-to-md.py

#!/usr/local/bin/python3

import json
from pprint import pprint
from collections import defaultdict
from pathlib import Path


def main():
	path = Path("./data.json").resolve()
	print(path)

	with path.open() as f:
		content = json.load(f)

	lookup = {
		"intents": defaultdict(list),
		"synonyms": defaultdict(list)
	}

	content = content['rasa_nlu_data']['common_examples']
	for item in content:
		intent = item['intent']
		text = item['text']

		for entity in reversed(item['entities']):
			text = text.replace(entity['value'], f"[{entity['value']}]({entity['entity']})")

		lookup['intents'][intent].append(text)

	for intent, words in lookup['intents'].items():
		print(f"## intent:{intent}")
		[print(f"- {text}") for text in words]
		print("")


if __name__ == '__main__':
	main()