taskrabbit/elasticsearch-dump

const base = require('./base.js')
const util = require('util')
const fs = require('fs')
const { parse, format } = require('fast-csv')
const StreamSplitter = require('../splitters/csvStreamSplitter.js')
const { PassThrough } = require('stream')
const zlib = require('zlib')
const _ = require('lodash')
const { fromCsvUrl } = require('../is-url.js')
const jsonParser = require('../jsonparser.js')

class csv extends base {
  constructor (parent, file, options) {
    super(parent, file, options)
    this._reading = false

    // csv headers
    this._customHeaders = parent.options.csvCustomHeaders ? _.split(parent.options.csvCustomHeaders, ',') : []
    this._headers = this._customHeaders.length > 0 ? this._customHeaders : parent.options.csvFirstRowAsHeaders

    // csv custom columns
    this._csvIdColumn = this.parent.options.csvIdColumn
    this._csvIndexColumn = this.parent.options.csvIndexColumn
    this._csvTypeColumn = this.parent.options.csvTypeColumn
    this._csvIgnoreAutoColumns = this.parent.options.csvIgnoreAutoColumns

    this._csvConfigs = this.parent.options.csvConfigs

    // csv url parse
    this._fileObj = fromCsvUrl(this.file)

    this.streamSplitter = new StreamSplitter(this._fileObj.path, parent.options, this)
  }

  flatten (json) {
    return _.mapValues(json, (v) => jsonParser.stringify(v, this.parent))
  }

  unflatten (json) {
    return _.mapValues(json, (v) => jsonParser.parse(v, this.parent))
  }

  async setupGet (offset) {
    this.bufferedData = []
    this.stream = parse(this._csvConfigs || {
      headers: this._headers,
      delimiter: this.parent.options.csvDelimiter,
      renameHeaders: this.parent.options.csvRenameHeaders,
      ignoreEmpty: this.parent.options.csvIgnoreEmpty,
      skipLines: this.parent.options.csvSkipLines,
      skipRows: this.parent.options.csvSkipRows,
      maxRows: this.parent.options.csvMaxRows,
      trim: this.parent.options.csvTrim,
      rtrim: this.parent.options.csvRTrim,
      ltrim: this.parent.options.csvLTrim,
      discardUnmappedColumns: this.parent.options.csvDiscardUnmappedColumns,
      quote: this.parent.options.csvQuoteChar,
      escape: this.parent.options.csvEscapeChar
    })

    if (!this.elementsToSkip) { this.elementsToSkip = offset }

    // converts the csv -> json object into
    // elasticdump object
    this.stream.transform(r => {
      const data = this.parent.options.csvHandleNestedData ? this.unflatten(r) : r

      const o = {
        _source: data
      }

      if (this._csvIdColumn) { o._id = data[this._csvIdColumn] }
      if (this._csvIndexColumn) { o._index = data[this._csvIndexColumn] }
      if (this._csvTypeColumn) { o._type = data[this._csvTypeColumn] }

      return o
    })

    this.metaStream = fs.createReadStream(this._fileObj.path)

    this.__setupStreamEvents()
    this._throughStream = new PassThrough()
    if (this.parent.options.fsCompress) {
      this._throughStream = zlib.createGunzip()
    }

    try {
      this.metaStream.pipe(this._throughStream).pipe(this.stream)
    } catch (err) {
      throw new Error(util.inspect({ msg: 'Pipe failed', err }))
    }
  }

  __handleData (data) {
    let lineCounter = 0
    data.forEach(elem => {
      let targetElem = elem
      if (elem._source) {
        targetElem = _.defaults(
          {
            ...elem._source
          },
          this._csvIgnoreAutoColumns
            ? {}
            : {
                [this._csvIdColumn || '@id']: elem._id,
                [this._csvIndexColumn || '@index']: elem._index,
                [this._csvTypeColumn || '@type']: elem._type
              })
      }

      this.log(targetElem)
      lineCounter++
    })
    return lineCounter
  }

  createCsvStream () {
    const _csvStream = format(this._csvConfigs || {
      headers: this._headers,
      writeHeaders: this.parent.options.csvWriteHeaders,
      delimiter: this.parent.options.csvDelimiter,
      rowDelimiter: this.parent.options.csvRowDelimiter,
      quote: this.parent.options.csvQuoteChar,
      escape: this.parent.options.csvEscapeChar,
      includeEndRowDelimiter: this.parent.options.csvIncludeEndRowDelimiter
    })

    // simple flat for csv
    if (this.parent.options.csvHandleNestedData) {
      _csvStream.transform(this.flatten.bind(this))
    }
    return _csvStream
  }

  // accept arr, callback where arr is an array of objects
  // return (error, writes)
  set (data, limit, offset, callback) {
    const error = null
    let lineCounter = 0

    if (!this.stream) {
      // TODO: add options to append the file
      if (!this.parent.options.overwrite && fs.existsSync(this._fileObj.path)) {
        return callback(new Error(`File \`${this.file}\` already exists, quitting`))
      } else if (this.shouldSplit) {
        // do nothing !
      } else {
        const _csvStream = this.createCsvStream()

        let _throughStream = new PassThrough()
        if (this.parent.options.fsCompress) {
          _throughStream = zlib.createGzip()
        }

        _csvStream.pipe(_throughStream).pipe(fs.createWriteStream(this._fileObj.path))
        this.stream = _csvStream
      }
    }

    if (data.length === 0) {
      if (this.shouldSplit) {
        this.streamSplitter.end()
      } else {
        this.stream.on('close', () => {
          delete this.stream
          return callback(null, lineCounter)
        })

        this.stream.end()
      }
    } else {
      lineCounter += this.__handleData(data)
      process.nextTick(() => callback(error, lineCounter))
    }
  }

  log (line) {
    if (this.shouldSplit) {
      this.streamSplitter.write(line)
    } else {
      this.stream.write(line)
    }
  }
}

module.exports = {
  csv
}