UUID JSON export from Mongo DB

The 'ugly' string is base64 encoded.
We get the MongoDB file with a copy task from the and write it to a blob storage. Afterwards we import it with databricks and decode the base64.

The function we use:

def base64ToGUID(b64str):
    try:
      s = bitstring.BitArray(bytes=base64.b64decode(b64str)).hex
      def rev2(s_):
        def chunks(n):
          for i in range(0, len(s_), n):
              yield s_[i:i+n]
        return "".join(list(chunks(2))[::-1])
      return "-".join([rev2(s[:8]),rev2(s[8:][:4]),rev2(s[12:][:4]),s[16:][:4],s[20:]])
    except:
      return None

We read the JSON to a spark dataframe, convert it to a pandas dataframe and apply the decoding:

# Create pandas dataframe
df_pandas = df_spark.toPandas()

#iterate over rows to decode the columns
for index, row in df_pandas.iterrows():
   row["row_name_1"] = base64ToGUID(row["row_name_1"])
   row["row_name_2"] = base64ToGUID(row["row_name_2"])
   row["row_name_n"] = base64ToGUID(row["row_name_n"])

Hope this helps.