How to pass column name into parameter in pandas

Question

I am trying to create a xml from a dataframe . Issue that I am facing i want pass column into paramter and its not working.

    ORDER_NO        1175        1304          1421  
   7283630          2019-12-04  2019-12-10   2019-12-12 
   7283650          2019-12-25  NaN          2019-12-20

My code

header = """<ORD>{}</ORD>"""
body ="""
<osi:ORDSTSINF types:STSCDE="{}">
<DTM>{}</DTM>
</osi:ORDSTSINF>
<osi:ORDSTSINF types:STSCDE="{}">
<DTM>{}</DTM>
</osi:ORDSTSINF>
<osi:ORDSTSINF types:STSCDE="{}">
<DTM>{}</DTM>
</osi:ORDSTSINF>"""

for row in df.itertuples():
    with open(f'{row[1]}.xml', 'w') as f:
        f.write(header.format(row[1]))
        f.write(body.format(col[2], row[2], col[3],row[3],col[4],row[4]))

I want to pass then columnname wherever STS={} is coming.

Expected output

<ORD> 7283630</ORD>
<osi:ORDSTSINF types:STSCDE="1175">
<DTM>2019-12-04</DTM>
<osi:ORDSTSINF types:STSCDE="1304">
<DTM>2019-12-10</DTM>
<osi:ORDSTSINF types:STSCDE="1421">
<DTM>22019-12-12</DTM>

How can this be done in python?

jezrael · Accepted Answer · 2020-02-21 06:43:50Z

1

Because python count from 0 in indexing subtract 1 from selecting columns names like cols[2] to cols[1] for all values:

cols = df.columns
for row in df.itertuples():
    with open(f'{row[1]}.xml', 'w') as f:
        f.write(header.format(row[1]))
        f.write(body.format(cols[1], row[2], cols[2],row[3],cols[3],row[4]))

<ORD>7283630</ORD>
<osi:ORDSTSINF types:STSCDE="1175">
<DTM>2019-12-04</DTM>
</osi:ORDSTSINF>
<osi:ORDSTSINF types:STSCDE="1304">
<DTM>2019-12-10</DTM>
</osi:ORDSTSINF>
<osi:ORDSTSINF types:STSCDE="1421">
<DTM>2019-12-12</DTM>
</osi:ORDSTSINF>

EDIT:

More dynamic solution with loop:

header = """<ORD>{}</ORD>"""
body ="""
<osi:ORDSTSINF types:STSCDE="{}">
<DTM>{}</DTM>"""

cols = df.columns
for row in df.itertuples():
    with open(f'{row[1]}.xml', 'w') as f:
        f.write(header.format(row[1]))
        for c, r in zip(row[2:], cols[1:]):
            f.write(body.format(r, c))

EDIT1: For mit missing values add notna:

header = """<ORD>{}</ORD>"""
body ="""
<osi:ORDSTSINF types:STSCDE="{}">
<DTM>{}</DTM>"""

cols = df.columns
for row in df.itertuples():
    with open(f'{row[1]}.xml', 'w') as f:
        f.write(header.format(row[1]))
        for c, r in zip(row[2:], cols[1:]):
            if pd.notna(c):
                f.write(body.format(r, c))

edited Feb 21, 2020 at 6:43

answered Feb 20, 2020 at 7:17

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

aeapen Over a year ago

@jezrael, Thanks it works . Is there another easy way of writing body statement. Here I based on number of columns ,I am duplicating statement . Instead canwe automatically create number body lines needed based on columns and pass the value

aeapen Over a year ago

I have more question,for record 2 number in data frame value of column 1304 is NaN. Then I want remove body being generated for 1304 column and keep the rest. In general wherever NaN is occurring that part has excluded from the body.Can that be done?

jezrael Over a year ago

@aeapen so in output is 5 times <osi:ORDSTSINF types:STSCDE=? Now I am offline, on phone only, so tomorrow should be added solution.

aeapen Over a year ago

I dont want <osi:ORDSTSINF types:STSCDE="1304"> <DTM>NaN</DTM> to be created but for 1175 &1421 the XML itshould be generated. In Short , I want the code skip the XML generation when row values for STS_CDE is NaN.

aeapen Over a year ago

for record 1 in the data frame ,xml should generated for all status codes but for record 2 only 1175 and 1421 should be created avoiding NaN

Collectives™ on Stack Overflow

How to pass column name into parameter in pandas

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related