#!/usr/bin/env bash

SOURCE_BUCKET="s3://sme-data-decompressed"
TARGET_BUCKET="s3://sme-data-archive"

DATES=()
DATES+=("2020-12-02")
DATES+=("2020-12-03")
DATES+=("2020-12-04")
DATES+=("2020-12-05")
DATES+=("2020-12-06")

COUNTRIES=()
COUNTRIES+=("US")
COUNTRIES+=("AU")
COUNTRIES+=("BR")
COUNTRIES+=("CA")
COUNTRIES+=("DE")
COUNTRIES+=("ES")
COUNTRIES+=("GB")
COUNTRIES+=("IT")
COUNTRIES+=("MX")

set -x

for d in "${DATES[@]}"; do
  for c in "${COUNTRIES[@]}"; do
    FILE_NAME="streams_${d//-/}_${c}.json"
    ARCHIVE_NAME="${FILE_NAME//.json/}.gz"
    if ! aws s3 ls "${TARGET_BUCKET}/spotify/streams/v4/report_date=${d}/report_licensor=sme/${ARCHIVE_NAME}";
    then
      aws s3 cp "${SOURCE_BUCKET}/spotify/streams/v4/report_date=${d}/report_licensor=sme/${FILE_NAME}" ./
      gzip "${FILE_NAME}"
      mv "${FILE_NAME}.gz" "${ARCHIVE_NAME}"
      aws s3 cp "${ARCHIVE_NAME}" "${TARGET_BUCKET}/spotify/streams/v4/report_date=${d}/report_licensor=sme/"
      rm "${ARCHIVE_NAME}"
    else
      echo "File ${ARCHIVE_NAME} already exists in bucket ${TARGET_BUCKET}"
    fi
  done
done