convert gentle word alignment to w3c annotations

transform.py

import json

def main():
  with open("word_aligned.json", "r") as f:
      aligned = json.load(f)
      transform(aligned["words"])

def transform(words):
    enumerated = [{"word":i, "position":x} for x,i in enumerate(words)]
    annotations = {"items": list(map(transform_annotation, enumerated))}
    print(json.dumps(annotations, indent=4))

def transform_annotation(word):
    if "alignedWord" in word["word"]:

        return {
            "id":"http://example.org/podcast/annotation/{0}".format(word["position"]),
            "type":"Annotation",
            "motivation":"painting",
            "body": {
                "id":"http://example.org/podcast/annotation/{0}/word".format(word["position"]),
                "type": "Text",
                "value": word["word"]["alignedWord"]
            },
        "target":"http://example.org/podcast/canvas#t={0},{1}".format(word["word"]["start"], word["word"]["end"])
            }

if __name__== "__main__":
  main()