### Install polars-whichlang

Source: https://github.com/rmalouf/polars_whichlang/blob/main/README.md

Install the library using pip. This is the only setup required.

```bash
pip install polars-whichlang
```

--------------------------------

### Detect Language in LazyFrame

Source: https://context7.com/rmalouf/polars_whichlang/llms.txt

Apply `detect_lang` to a LazyFrame for efficient processing within Polars' query engine. The syntax is identical to eager DataFrames.

```python
# --- Lazy LazyFrame usage ---
lazy_result = (
    df.lazy()
    .with_columns(detect_lang("text").alias("lang"))
    .collect()
)
print(lazy_result)
```

--------------------------------

### detect_lang(text)

Source: https://context7.com/rmalouf/polars_whichlang/llms.txt

Detects the language of a string column and returns an ISO 639-3 three-letter language code. This function operates element-wise and integrates natively into Polars' execution engine, supporting both eager and lazy APIs. It accepts a column name, a Polars expression, or a Polars Series as input and returns a Polars expression of dtype String. Raises ComputeError if the input column is not of string dtype.

```APIDOC
## detect_lang(text)

### Description
Registers and calls the compiled Rust plugin function `detect_lang` on a Polars string expression. The `text` argument accepts a column name (string), a `pl.Expr`, or a `pl.Series`. Returns a `pl.Expr` of dtype `String` containing ISO 639-3 three-letter language codes for each row. The function is element-wise and integrates natively into Polars' execution engine, making it compatible with both eager and lazy APIs. Raises `ComputeError` if the input column is not of string dtype.

### Parameters
#### Path Parameters
- **text** (string | pl.Expr | pl.Series) - Required - The input column or expression containing strings to analyze.

### Request Example
```python
import polars as pl
from polars_whichlang import detect_lang

df = pl.DataFrame({
    "id": [1, 2, 3, 4, 5, 6],
    "text": [
        "This is a test.",            # English
        "Đây là một bài kiểm tra.",   # Vietnamese
        "Dies ist ein Test",          # German
        "这是一个测试",                 # Mandarin
        "이건 테스트야",               # Korean
        "Это тест",                   # Russian
    ],
})

result = df.with_columns(
    detect_lang("text").alias("lang")
)
print(result)

# --- Lazy LazyFrame usage ---
lazy_result = (
    df.lazy()
    .with_columns(detect_lang("text").alias("lang"))
    .collect()
)
print(lazy_result)

# --- Using a Polars expression directly ---
result_expr = df.with_columns(
    detect_lang(pl.col("text")).alias("lang")
)

# --- Filtering to only English rows ---
english_only = (
    df.lazy()
    .filter(detect_lang("text") == pl.lit("eng"))
    .collect()
)
print(english_only)

# --- Error handling: non-string column raises ComputeError ---
from polars.exceptions import ComputeError

numeric_df = pl.DataFrame({"val": [1, 2, 3]})
try:
    numeric_df.with_columns(detect_lang("val").alias("lang"))
except ComputeError as e:
    print(f"Error: {e}")  # ComputeError: input column must be of type String
```

### Response
#### Success Response (pl.Expr)
- **lang** (String) - A Polars expression that, when evaluated, will contain the ISO 639-3 three-letter language code for each input string.
```

--------------------------------

### Handle Non-String Column Errors

Source: https://context7.com/rmalouf/polars_whichlang/llms.txt

Attempting to use `detect_lang` on a non-string column will raise a `ComputeError`. Ensure the input column is of string dtype.

```python
# --- Error handling: non-string column raises ComputeError ---
from polars.exceptions import ComputeError

numeric_df = pl.DataFrame({"val": [1, 2, 3]})
try:
    numeric_df.with_columns(detect_lang("val").alias("lang"))
except ComputeError as e:
    print(f"Error: {e}")  # ComputeError: input column must be of type String
```

--------------------------------

### Detect Language in Eager DataFrame

Source: https://context7.com/rmalouf/polars_whichlang/llms.txt

Use `detect_lang` within `with_columns` to add a language code column to an eager DataFrame. The input can be a column name or a Polars expression.

```python
import polars as pl
from polars_whichlang import detect_lang

# --- Eager DataFrame usage ---
df = pl.DataFrame({
    "id": [1, 2, 3, 4, 5, 6],
    "text": [
        "This is a test.",            # English
        "Đây là một bài kiểm tra.",   # Vietnamese
        "Dies ist ein Test",          # German
        "这是一个测试",                 # Mandarin
        "이건 테스트야",               # Korean
        "Это тест",                   # Russian
    ],
})

result = df.with_columns(
    detect_lang("text").alias("lang")
)
print(result)
```

--------------------------------

### Use Polars Expression with detect_lang

Source: https://context7.com/rmalouf/polars_whichlang/llms.txt

Pass a Polars column expression directly to `detect_lang` for more complex query constructions.

```python
# --- Using a Polars expression directly ---
result_expr = df.with_columns(
    detect_lang(pl.col("text")).alias("lang")
)
```

--------------------------------

### Filter Rows by Detected Language

Source: https://context7.com/rmalouf/polars_whichlang/llms.txt

Use `detect_lang` within a `filter` operation to select rows based on their identified language code.

```python
# --- Filtering to only English rows ---
english_only = (
    df.lazy()
    .filter(detect_lang("text") == pl.lit("eng"))
    .collect()
)
print(english_only)
```

--------------------------------

### Detect Language in Polars DataFrame

Source: https://github.com/rmalouf/polars_whichlang/blob/main/README.md

Use the `detect_lang` function from `polars_whichlang` to add a language detection column to a Polars DataFrame. Ensure the text column is of string type.

```python
import polars as pl
from polars_whichlang import detect_lang

df = pl.DataFrame(
    {
        "index": [1, 2, 3, 4],
        "text": [
            "This is a test. ", 
            "Đây là một bài kiểm tra.", 
            "Dies ist ein Test", 
            "这是一个测试"
        ],
    }
)

df.with_columns(detect_lang('text').alias('lang'))
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.