Adding multiple columns with loc fails

[impact27]

Code Sample, a copy-pastable example if possible

import pandas as pd

df = pd.DataFrame(index=[1, 2], columns = list('ab'))

# Single assignment on existing column
df.loc[1, 'a'] = 1

# Multiple assignment on existing column
df.loc[2, ['a', 'b']] = [1, 2]

# Single assignment on new column
df.loc[3, 'c'] = 3

# Multiple assignment on single new column
df.loc[[1, 2], 'l'] = [6, 7]

# Multiple assignment on multiple new column
# Surprisingly fails
try:
    df.loc[3, ['d', 'e']] = [6, 7]
    print(df)
except KeyError:
    print("Error")

Problem description

The behaviour of loc is surprising:

• When assigning, new columns and index are created as required.
• You can assign multiple index and columns
• BUT, if you try to create a column or index while doing multiple assignment, new columns and index are NOT created as required, and you get a KeyError.

Futhermore, using the indexing operators [] directly will create the required columns.

Expected Output

     a    b    c    l    d    e
1    1  NaN  NaN  6.0  NaN  NaN
2    1    2  NaN  7.0  NaN  NaN
3  NaN  NaN  3.0  NaN  6.0  7.0

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.7.2.final.0
python-bits: 64
OS: Darwin
OS-release: 18.2.0
machine: x86_64
processor: i386
byteorder: little
LC_ALL: en_US.UTF-8
LANG: en_US.UTF-8
LOCALE: en_US.UTF-8

pandas: 0.24.1
pytest: 4.2.0
pip: 19.0.2
setuptools: 40.8.0
Cython: 0.28.5
numpy: 1.15.2
scipy: 1.2.0
pyarrow: None
xarray: None
IPython: 7.0.1
sphinx: 1.8.1
patsy: None
dateutil: 2.7.3
pytz: 2018.5
blosc: None
bottleneck: None
tables: None
numexpr: None
feather: None
matplotlib: 3.0.2
openpyxl: None
xlrd: None
xlwt: None
xlsxwriter: None
lxml.etree: 4.2.5
bs4: None
html5lib: 1.0.1
sqlalchemy: None
pymysql: None
psycopg2: None
jinja2: 2.10
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None
gcsfs: None

[TomAugspurger]

Thanks. You might need to look into _convert_indexer in pandas/core/indexing.py to check why is_setter implies raise_missing. It seems like this should work but I may be missing something.

For clarity, could you edit down your issue to include just the

df = ...
df.loc[3, ['d', 'e']] = [6, 7]

and your expected output?

[mroeschke]

This looks to work on master now. Could use a test

In [11]: import pandas as pd
    ...:
    ...: df = pd.DataFrame(index=[1, 2], columns = list('ab'))
    ...:
    ...: # Single assignment on existing column
    ...: df.loc[1, 'a'] = 1
    ...:
    ...: # Multiple assignment on existing column
    ...: df.loc[2, ['a', 'b']] = [1, 2]
    ...:
    ...: # Single assignment on new column
    ...: df.loc[3, 'c'] = 3
    ...:
    ...: # Multiple assignment on single new column
    ...: df.loc[[1, 2], 'l'] = [6, 7]
    ...:
    ...: # Multiple assignment on multiple new column
    ...: # Surprisingly fails
    ...: try:
    ...:     df.loc[3, ['d', 'e']] = [6, 7]
    ...:     print(df)
    ...: except KeyError:
    ...:     print("Error")
    ...:
     a    b    c    l    d    e
1    1  NaN  NaN  6.0  NaN  NaN
2    1    2  NaN  7.0  NaN  NaN
3  NaN  NaN  3.0  NaN  6.0  7.0