Cannot add constant field to empty structs

[jankislinger]

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

import polars as pl

df = pl.DataFrame({"s": [{}, {}, {}]})
df.with_columns(
    pl.col("s").struct.with_fields(a=pl.lit(42))
)

Log output

Traceback (most recent call last):
  File "/.../report.py", line 4, in <module>
    df.with_columns(
    ~~~~~~~~~~~~~~~^
        pl.col("s").struct.with_fields(a=pl.lit(42))
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/.../.venv/lib/python3.13/site-packages/polars/dataframe/frame.py", line 9586, in with_columns
    return self.lazy().with_columns(*exprs, **named_exprs).collect(_eager=True)
           ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
  File "/.../.venv/lib/python3.13/site-packages/polars/lazyframe/frame.py", line 2056, in collect
    return wrap_df(ldf.collect(callback))
                   ~~~~~~~~~~~^^^^^^^^^^
polars.exceptions.InvalidOperationError: Series s, length 1 doesn't match the DataFrame height of 3

If you want expression: col("s").with_fields([dyn int: 42.alias("a")]) to be broadcasted, ensure it is a scalar (for instance by adding '.first()').

Issue description

When a struct column has no fields, it cannot add extra constant field. Adding pl.lit(42).first() didn't help. It seems like the with_fields is trying to create a series of length 1.

>>> df.select(
>>>     pl.col("s").struct.with_fields(a=pl.lit(42))
>>> )
shape: (1, 1)
┌───────────┐
│ s         │
│ ---       │
│ struct[1] │
╞═══════════╡
│ {42}      │
└───────────┘

I tried forcing some non-empty data type to the struct which fixes the issue.

>>> df = pl.DataFrame({"s": pl.Series([{}, {}, {}], dtype=pl.Struct([pl.Field("_", pl.Boolean)]))})
>>> df.with_columns(
>>>     pl.col("s").struct.with_fields(a=pl.lit(42).first())
>>> )
shape: (3, 1)
┌───────────┐
│ s         │
│ ---       │
│ struct[2] │
╞═══════════╡
│ {null,42} │
│ {null,42} │
│ {null,42} │
└───────────┘

Expected behavior

I would expect to create a column with the same field value in each element.

shape: (3, 1)
┌───────────┐
│ s         │
│ ---       │
│ struct[1] │
╞═══════════╡
│ {42}      │
│ {42}      │
│ {42}      │
└───────────┘

Installed versions

--------Version info---------
Polars:              1.21.0
Index type:          UInt32
Platform:            Linux-6.8.0-52-generic-x86_64-with-glibc2.39
Python:              3.13.1 (main, Jan  5 2025, 05:33:47) [Clang 19.1.6 ]
LTS CPU:             False
----Optional dependencies----
Azure CLI            <not installed>
adbc_driver_manager  <not installed>
altair               <not installed>
azure.identity       <not installed>
boto3                <not installed>
cloudpickle          <not installed>
connectorx           <not installed>
deltalake            <not installed>
fastexcel            <not installed>
fsspec               <not installed>
gevent               <not installed>
google.auth          <not installed>
great_tables         <not installed>
matplotlib           <not installed>
numpy                <not installed>
openpyxl             <not installed>
pandas               <not installed>
pyarrow              19.0.0
pydantic             <not installed>
pyiceberg            <not installed>
sqlalchemy           <not installed>
torch                <not installed>
xlsx2csv             <not installed>
xlsxwriter           <not installed>

[mcrumiller]

As a temporary workaround, you can use pl.repeat:

import polars as pl

df = pl.DataFrame({"s": [{}, {}, {}]})
df.with_columns(
    pl.col("s").struct.with_fields(
        a=pl.repeat(42, pl.len())
    )
)
# shape: (3, 1)
# ┌───────────┐
# │ s         │
# │ ---       │
# │ struct[1] │
# ╞═══════════╡
# │ {42}      │
# │ {42}      │
# │ {42}      │
# └───────────┘

[jankislinger]

that is better workaround than what I have, thanks!