{
  "id": 115752,
  "title": "How to get long answer text?",
  "url": "/competitions/tensorflow2-question-answering/discussion/115752",
  "author_name": "",
  "post_date": "2019-11-05T03:54:55.334702800Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>hi there,\nhow to get long answer candidate texts by using the document tokens in long answer candidates?\nthanks.</p>",
  "messages": [
    {
      "id": "665493",
      "postDate": "11/05/2019 03:54:55",
      "content": "<p>hi there,\nhow to get long answer candidate texts by using the document tokens in long answer candidates?\nthanks.</p>",
      "rawMarkdown": "hi there,\nhow to get long answer candidate texts by using the document tokens in long answer candidates?\nthanks.",
      "votes": null
    },
    {
      "id": "665610",
      "postDate": "11/05/2019 07:10:50",
      "content": "<p>If you split the document_text into an array using spaces, then use the start and end values in the candidate data, that should give you the corresponding candidate text</p>",
      "rawMarkdown": "If you split the document_text into an array using spaces, then use the start and end values in the candidate data, that should give you the corresponding candidate text",
      "votes": null
    },
    {
      "id": "665806",
      "postDate": "11/05/2019 12:18:14",
      "content": "<p>great. thanks, alex. it worked</p>",
      "rawMarkdown": "great. thanks, alex. it worked",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 665610,
      "author_name": "afederation",
      "author_url": "",
      "post_date": "11/05/2019 07:10:50",
      "content": "<p>If you split the document_text into an array using spaces, then use the start and end values in the candidate data, that should give you the corresponding candidate text</p>",
      "votes": null,
      "replies": [
        {
          "id": 665806,
          "author_name": "willbe",
          "author_url": "",
          "post_date": "11/05/2019 12:18:14",
          "content": "<p>great. thanks, alex. it worked</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "665493": "hi there,\nhow to get long answer candidate texts by using the document tokens in long answer candidates?\nthanks.",
    "665610": "If you split the document_text into an array using spaces, then use the start and end values in the candidate data, that should give you the corresponding candidate text",
    "665806": "great. thanks, alex. it worked"
  },
  "source": "meta"
}