+---+----+----------------------------+
| id| seq|                  seq_detail|
+---+----+----------------------------+
|  a|   1|               {'a':'apple'}|
|  b|   1|               {'a':'apple'}|
|  c| 1 2| [{'a':'apple'},{'b':'ben'}]|
|  d|   2|                 {'b':'ben'}|
|  e|   1|               {'a':'apple'}|
|  f|   3|                 {'c':'cat'}|
|  g|   1|               {'a':'apple'}|
|  h|   1|               {'a':'apple'}|
+---+----+----------------------------+

+----+-------------------------+-----------+---------+
| seq|                       id| seq_detail| id_count|
+----+-------------------------+-----------+---------+
| 1_1|['a', 'b', 'c', 'e', 'g']|      apple|        5|
| 1_2|                      'h'|      apple|        1|
| 2_1|               ['c', 'd']|        ben|        2|
| 3_1|                      'f'|        cat|        1|
+---+----+---------------------+-----------+---------+

from pyspark.sql import SparkSession


spark = SparkSession.builder
    .master('local[*]')
    .appName("sample")
    .config("spark.some.config.option", "some-value")
    .getOrCreate()

for i in range(1,4):
    df = spark.read.parquet('./parquet')
    df = df. filter(df.seq.contains(str(i)))
    df.show()
    if df.count() == 0:
        pass
    else:
        df = df.collect()
        data = [x.asdict() for x in df]