zfill with leading + differs from Python stdlib and from pandas #22983
Description
Checks
- I have checked that this issue has not already been reported.
- I have confirmed this bug exists on the latest version of Polars.
Reproducible example
In [1]: import polars as pl
In [2]: import pandas as pd
In [3]: '-1'.zfill(3)
Out[3]: '-01'
In [4]: pd.Series(['-1']).str.zfill(3)
Out[4]:
0 -01
dtype: object
In [5]: pl.Series(['-1']).str.zfill(3)
Out[5]:
shape: (1,)
Series: '' [str]
[
"-01"
]
In [6]: '+1'.zfill(3)
Out[6]: '+01'
In [7]: pd.Series(['+1']).str.zfill(3)
Out[7]:
0 +01
dtype: object
In [8]: pl.Series(['+1']).str.zfill(3)
Out[8]:
shape: (1,)
Series: '' [str]
[
"0+1"
]Log output
Issue description
If there's a leading '-', then pandas, Polars, and Python agree
If there's a leading '+', then pandas and Python pad to its right, whereas Polars pads to its left
Expected behavior
I think
In [8]: pl.Series(['+1']).str.zfill(3)
Out[8]:
shape: (1,)
Series: '' [str]
[
"0+1"
]
?
Installed versions
In [9]: pl.show_versions()
--------Version info---------
Polars: 1.29.0
Index type: UInt32
Platform: Linux-5.15.167.4-microsoft-standard-WSL2-x86_64-with-glibc2.35
Python: 3.12.5 (main, Aug 14 2024, 05:08:31) [Clang 18.1.8 ]
LTS CPU: False
----Optional dependencies----
Azure CLI <not installed>
adbc_driver_manager <not installed>
altair 5.5.0
azure.identity <not installed>
boto3 <not installed>
cloudpickle 3.1.1
connectorx <not installed>
deltalake <not installed>
fastexcel <not installed>
fsspec 2025.3.2
gevent <not installed>
google.auth 2.38.0
great_tables 0.16.1
matplotlib 3.9.2
numpy 2.2.6
openpyxl <not installed>
pandas 2.2.3
polars_cloud <not installed>
pyarrow 19.0.0
pydantic <not installed>
pyiceberg <not installed>
sqlalchemy <not installed>
torch <not installed>
xlsx2csv <not installed>
xlsxwriter <not installed>