CODEWITHSUNDEEP
apache-sparkpysparkapache-spark-sql
Arvinth
Arvinth

Reputation: 70

Error while removing empty spaces in Spark dataframe - PySpark

I am reading csv file into spark dataframe. The csv has empty spaces " " in many columns, i want to remove these empty spaces. There are 500 columns in csv, So i cannot specific columns manually in my code

Sample data:

  ADVANCE_TYPE  CHNG_DT    BU_IN
     A          20190718    1
                20190728    2 
                20190714     
     B          20190705     
                20190724    4

Code:

    from pyspark.sql.functions import col,when,regexp_replace,trim

    df_csv = spark.read.options(header='true').options(delimiter=',').options(inferSchema='true').options(nullValue="None").csv("test41.csv")  


    for col_name in df_csv.columns:
       df_csv = df_csv.select(trim(col(col_name)))

But these code is not removing empty spaces. Please help!

Upvotes: 0

Views: 296

Answers (1)

notNull
notNull

Reputation: 31480

You can apply trim on all the required columns using list comprehension.

Example:

df=spark.createDataFrame([("   ","12343","   ","9  ","   0")])

#finding length of each column
expr=[length(col(col_name)).name('length'+ col_name) for col_name in df.columns]

df.select(expr).show()
#+--------+--------+--------+--------+--------+
#|length_1|length_2|length_3|length_4|length_5|
#+--------+--------+--------+--------+--------+
#|       3|       5|       3|       3|       4|
#+--------+--------+--------+--------+--------+

#trim on all the df columns
expr=[trim(col(col_name)).name(col_name) for col_name in df.columns]

df1=df.select(expr)
df1.show()
#+---+-----+---+---+---+
#| _1|   _2| _3| _4| _5|
#+---+-----+---+---+---+
#|   |12343|   |  9|  0|
#+---+-----+---+---+---+

#length on df1 columns
expr=[length(col(col_name)).name('length'+ col_name) for col_name in df.columns]
df1.select(expr).show()
#+--------+--------+--------+--------+--------+
#|length_1|length_2|length_3|length_4|length_5|
#+--------+--------+--------+--------+--------+
#|       0|       5|       0|       1|       1|
#+--------+--------+--------+--------+--------+

Upvotes: 2

Related Questions