search_sorted silently allows comparisons between different types

[cmdlineluser]

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({
    "foo": [1, 4, 7],
    "bar": [2, 3, 5]
})

df.with_columns(idx = pl.col("foo").search_sorted("bar"))

Log output

No response

Issue description

I just noticed this as previously (it seems around <= 0.20.23) passing a string to search_sorted was parsed as a column name.

It now seems to be parsed as a literal, but there does not seem to be a type check.

So in this case, we end up getting a "valid" result from searching an int column with a string, which seems unintended?

df.with_columns(
    expected = pl.col("foo").search_sorted(pl.col("bar")),
    actual = pl.col("foo").search_sorted("bar")
)

# shape: (3, 4)
# ┌─────┬─────┬──────────┬────────┐
# │ foo ┆ bar ┆ expected ┆ actual │
# │ --- ┆ --- ┆ ---      ┆ ---    │
# │ i64 ┆ i64 ┆ u32      ┆ u32    │
# ╞═════╪═════╪══════════╪════════╡
# │ 1   ┆ 2   ┆ 1        ┆ 3      │
# │ 4   ┆ 3   ┆ 1        ┆ 3      │
# │ 7   ┆ 5   ┆ 2        ┆ 3      │
# └─────┴─────┴──────────┴────────┘

Expected behavior

SchemaError invalid dtype?

Installed versions

--------Version info---------
Polars:               1.4.1
Index type:           UInt32
Platform:             macOS-13.6.1-arm64-arm-64bit
Python:               3.12.2 (main, Feb  6 2024, 20:19:44) [Clang 15.0.0 (clang-1500.1.0.2.5)]

----Optional dependencies----
adbc_driver_manager:  <not installed>
cloudpickle:          <not installed>
connectorx:           <not installed>
deltalake:            <not installed>
fastexcel:            <not installed>
fsspec:               <not installed>
gevent:               <not installed>
great_tables:         <not installed>
hvplot:               <not installed>
matplotlib:           <not installed>
nest_asyncio:         <not installed>
numpy:                1.26.4
openpyxl:             <not installed>
pandas:               2.2.1
pyarrow:              15.0.2
pydantic:             <not installed>
pyiceberg:            <not installed>
sqlalchemy:           <not installed>
torch:                <not installed>
xlsx2csv:             <not installed>
xlsxwriter:           <not installed>

[deanm0000]

I don't know if that was a lucky guess or you're being coy with "seems around" but 0.20.23 is exactly the first version where that behavior changed/broke.

Here's the PR
7ae1e58

[ritchie46]

Ouch.. that's a sneaky one.