Update in Python Pandas not working as expected

[Vanan]

Hi,

I'm trying to update python dataframes, but it is not working as expected.

 

Frame1

 

description|cusip|value1|value2|value3|VALUE4

this_is_file3|ATG0104|30|14|45|500

this_is_file1|ATG0102|10|12|13|200

this_is_file11|ATG0101|20|30|40|105

this_is_file2|ATG0103|20|22|23|300

 

 

Frame2

 

description|CUSIP|VALUE3|value2|value4|value1

this_is_file2_updated|ATG0102|34.40|10.21|100.14|90.12

this_is_file1_updated|ATG0101|12.50|11.34|13.76|104.2

this_is_file3_updated|ATG0103|10.13|9.12|14.34|405.14

 

 

Result after updating(value1, value2, value3) of Frame1 with Frame2 based on Key(cusip/CUSIP)

 

description|CUSIP|value1|value2|VALUE3|VALUE4

this_is_file1_updated|ATG0101|104.2|11.34|12.5|200

this_is_file2_updated|ATG0102|90.12|10.21|34.4|105

this_is_file3_updated|ATG0103|405.14|9.12|10.13|300

this_is_file3|ATG0104|30.0|14.0|45.0|500

 

I'm indexing the frames by Key(cusip/CUSIP) and then performing the update on columns (value1, value2 and VALUE3).

 

As you can observe, the value of last column(VALUE4) is not aligned properly with the keys. What might be the issue?

 

 

The following is the code that I'm using:

def replace(self, dataobj, k_options, K_options, d_options, D_options,S_options):

        print('Before renaming column names');

        print(list(self.frame.columns.values));

        for j in range(0, len(k_options)):

            self.frame.rename(columns = {k_options[j]: K_options[j]}, inplace=True);

        for i in range(0, len(d_options)):

            self.frame.rename(columns = {d_options[i]: D_options[i]}, inplace=True);

        print('After renaming column names');

        print(list(self.frame.columns.values));

        self.frame.set_index(K_options); 

        dataobj.frame.set_index(K_options);

        print('Other Frame: ', dataobj.frame);

        self.frame.update(dataobj.frame, join='left', overwrite=True);

        print('Updated Frame: ', self.frame);

        time.sleep(2000);

        self.frame.to_csv(S_options[0]+'_new.csv', sep='|',index=None, na_values=[""]);

        print('Updated');

 

Inputs:

k_options     ['cusip']

K_options    ['CUSIP']

d_options     "value1|value2|value3"

D_options    "value1|value2|VALUE3"

Thanks,

Vanan.