Get Document as Text

//native

/**

 * Get Document as Text

 * Get a Google Doc's title and its body as plain text (paragraphs and table cells, in reading order). Only the first tab of a multi-tab document is read.

 */

export async function main(auth: RT.Gdocs, document_id: string) {

  const response = await fetch(

    `https://docs.googleapis.com/v1/documents/${document_id}`,

    {

      headers: {

        Authorization: `Bearer ${auth.token}`,

        Accept: "application/json",

      },

    }

  )

  if (!response.ok) {

    throw new Error(`${response.status} ${await response.text()}`)

  }

  const doc = await response.json()

  return {

    document_id: doc.documentId,

    title: doc.title,

    text: toText(doc.body?.content ?? []),

  }

}

function toText(content: any[]): string {

  let text = ""

  for (const el of content) {

    for (const e of el.paragraph?.elements ?? [])

      text += e.textRun?.content ?? ""

    for (const row of el.table?.tableRows ?? [])

      for (const cell of row.tableCells ?? [])

        text += toText(cell.content ?? [])

    if (el.tableOfContents) text += toText(el.tableOfContents.content ?? [])

  }

  return text

}