### Install dependencies

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Install the project and configure pre-commits.

```bash
make install
```

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

### Create a new branch

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Start a new branch for your changes.

```bash
git checkout -b my-new-feature-branch
```

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

### Install autoimport

Source: https://github.com/lyz-code/autoimport/blob/main/README.md

Use pip to install the autoimport package into your Python environment.

```bash
pip install autoimport
```

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

### Example source code for autoimport

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

An example of Python source code containing missing and unused imports.

```python
import requests


def hello(names: Tuple[str]) -> None:
    for name in names:
        print(f"Hi {name}!")


os.getcwd()
```

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

### Build documentation

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Build the static documentation site locally.

```bash
make docs
```

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

### Set up virtual environment

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Configure a virtual environment for running tests.

```bash
virtualenv -p `which python3.7` env
source env/bin/activate
```

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

### Clone the repository

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Initial steps to clone your fork and enter the directory.

```bash
git clone git@github.com:<your username>/autoimport.git
cd autoimport
```

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

### Report version information

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Use this command to provide version details when opening an issue.

```bash
python -c "import autoimport.version; print(autoimport.version.version_info())"
```

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

### Specifying Custom Config File

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

Use the --config-file flag to point to a specific configuration file.

```bash
$: autoimport --config-file ~/.autoimport.toml file.py
```

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

### Run tests and linting

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Execute the default test suite and linting. Use the ARGS variable to pass arguments to pytest.

```bash
make
```

```bash
make test ARGS='-k test_markdownlint_passes'
```

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

### Format code

Source: https://github.com/lyz-code/autoimport/blob/main/docs/contributing.md

Run formatting tools to fix code style and imports.

```bash
make format
```

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

### Run autoimport via command line

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

Execute autoimport on specific files or directories.

```bash
$: autoimport file.py
$: autoimport dir/
```

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

### Configuration via pyproject.toml

Source: https://context7.com/lyz-code/autoimport/llms.txt

Autoimport can be configured through `pyproject.toml` to define custom common statements and other settings. Configuration is loaded automatically from the project directory.

```APIDOC
## Configuration via pyproject.toml

Autoimport can be configured through `pyproject.toml` to define custom common statements and other settings. Configuration is loaded automatically from the project directory.

### Configuration Options

- **disable_move_to_top** (bool) - Optional - If True, imports inside functions are left in place. Defaults to `false`.

### Common Statements

The `[tool.autoimport.common_statements]` section allows defining custom mappings for module aliases or frequently used objects.

#### Example `pyproject.toml`

```toml
# pyproject.toml

[tool.autoimport]
# Disable moving imports to top (not recommended, but useful for circular imports)
disable_move_to_top = false

# Define custom common statements
[tool.autoimport.common_statements]
"np" = "import numpy as np"
"pd" = "import pandas as pd"
"plt" = "import matplotlib.pyplot as plt"
"tf" = "import tensorflow as tf"
"torch" = "import torch"
"FooBar" = "from baz_qux import FooBar"
"MyModel" = "from myproject.models import MyModel"
```
```

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

### Configuring Common Statements

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

Define custom common statements in pyproject.toml or a standalone configuration file.

```toml
# pyproject.toml

[tool.autoimport.common_statements]
"np" = "import numpy as np"
"FooBar" = "from baz_qux import FooBar"
```

```toml
# .autoimport.toml

[common_statements]
"np" = "import numpy as np"
"FooBar" = "from baz_qux import FooBar"
```

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

### Configure pre-commit for autoimport

Source: https://github.com/lyz-code/autoimport/blob/main/docs/editor_integration.md

Add these lines to the .pre-commit-config.yaml file to enable autoimport as a pre-commit hook.

```yaml
repos:
  - repo: https://github.com/lyz-code/autoimport/
    rev: master
    hooks:
      - id: autoimport
```

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

### Use Custom Configuration File with Autoimport CLI

Source: https://context7.com/lyz-code/autoimport/llms.txt

Specify a custom configuration file for autoimport to use when processing Python files.

```bash
autoimport --config-file ~/.autoimport.toml file.py
```

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

### Run autoimport as a library

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

Use the fix_files function to process files or directories programmatically.

```python
from autoimport import fix_files

fix_files(["file.py", "dir/"])
```

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

### Ignore __init__.py Files with Autoimport CLI

Source: https://context7.com/lyz-code/autoimport/llms.txt

Configure autoimport to skip processing __init__.py files, which often contain intentionally unused imports.

```bash
autoimport --ignore-init-modules src/
```

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

### Configure pre-commit hook

Source: https://context7.com/lyz-code/autoimport/llms.txt

Integrate autoimport into your pre-commit workflow to automatically fix imports before commits.

```yaml
# .pre-commit-config.yaml

repos:
  - repo: https://github.com/lyz-code/autoimport/
    rev: master
    hooks:
      - id: autoimport
        # Optional: exclude certain files
        exclude: ^tests/fixtures/
        # Optional: add arguments
        args: [--ignore-init-modules]
```

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

### Fix Files with Configuration

Source: https://context7.com/lyz-code/autoimport/llms.txt

Apply `fix_files` with a custom configuration, such as defining common statements for import resolution.

```python
# With configuration
config = {
    "common_statements": {
        "np": "import numpy as np"
    }
}
file = click.File("r+").convert("analysis.py", None, None)
fix_files((file,), config=config)
```

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

### Autoimport Configuration in pyproject.toml

Source: https://context7.com/lyz-code/autoimport/llms.txt

Configure autoimport settings, including disabling import moving and defining custom common statements, within the `pyproject.toml` file.

```toml
# pyproject.toml

[tool.autoimport]
# Disable moving imports to top (not recommended, but useful for circular imports)
disable_move_to_top = false

# Define custom common statements
[tool.autoimport.common_statements]
"np" = "import numpy as np"
"pd" = "import pandas as pd"
"plt" = "import matplotlib.pyplot as plt"
"tf" = "import tensorflow as tf"
"torch" = "import torch"
"FooBar" = "from baz_qux import FooBar"
"MyModel" = "from myproject.models import MyModel"
```

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

### Fix Multiple Files and Directories with Autoimport CLI

Source: https://context7.com/lyz-code/autoimport/llms.txt

Process multiple Python files and directories simultaneously using the autoimport command.

```bash
autoimport file1.py file2.py mypackage/
```

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

### Read from Stdin and Output to Stdout with Autoimport CLI

Source: https://context7.com/lyz-code/autoimport/llms.txt

Pipe Python code to autoimport for processing and receive the fixed code on standard output.

```bash
cat file.py | autoimport -
```

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

### Command Line Interface

Source: https://context7.com/lyz-code/autoimport/llms.txt

The `autoimport` command-line tool can be used to process individual Python files or entire directories recursively to fix import statements.

```APIDOC
## Command Line Interface

The `autoimport` command processes Python files to fix import statements. It accepts individual files or directories (which are processed recursively for all `.py` files).

### Usage Examples

```bash
# Fix a single file
autoimport file.py

# Fix all Python files in a directory recursively
autoimport src/

# Fix multiple files and directories
autoimport file1.py file2.py mypackage/

# Use a custom configuration file
autoimport --config-file ~/.autoimport.toml file.py

# Ignore __init__.py files (which often have intentionally "unused" imports)
autoimport --ignore-init-modules src/

# Keep unused imports (don't remove them)
autoimport --keep-unused-imports file.py

# Read from stdin and output to stdout
cat file.py | autoimport -
```
```

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

### Use built-in common statements

Source: https://context7.com/lyz-code/autoimport/llms.txt

Autoimport automatically recognizes and maps common imports without requiring manual configuration.

```python
from autoimport import fix_code

# BeautifulSoup from bs4
source = "soup = BeautifulSoup(html, 'html.parser')"
print(fix_code(source))
# from bs4 import BeautifulSoup
#
#
# soup = BeautifulSoup(html, 'html.parser')

# Mock and patch from unittest.mock
source = """
mock_obj = Mock()
with patch('module.function'):
    pass
"""
print(fix_code(source))
# from unittest.mock import Mock
# from unittest.mock import patch
#
#
# mock_obj = Mock()
# with patch('module.function'):
#     pass

# Path from pathlib
source = "config_path = Path('/etc/config')"
print(fix_code(source))
# from pathlib import Path
#
#
# config_path = Path('/etc/config')
```

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

### Resulting source code after autoimport

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

The source code after running autoimport, showing corrected imports.

```python
import os
from typing import Tuple


def hello(names: Tuple[str]) -> None:
    for name in names:
        print(f"Hi {name}!")


os.getcwd()
```

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

### Recursively Fix Python Files in Directory with Autoimport CLI

Source: https://context7.com/lyz-code/autoimport/llms.txt

Apply autoimport to all Python files within a specified directory and its subdirectories.

```bash
autoimport src/
```

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

### Basic Fix Code Usage

Source: https://context7.com/lyz-code/autoimport/llms.txt

Demonstrates the basic functionality of `fix_code` to manage imports in a Python source string. Requires `Tuple` from `typing`.

```python
from autoimport import fix_code

# Basic usage - fix missing and unused imports
source = """
import requests

def hello(names: Tuple[str]) -> None:
    for name in names:
        print(f"Hi {name}!")

os.getcwd()
"""

fixed = fix_code(source)
print(fixed)
# Output:
# import os
# from typing import Tuple
# 
# 
# def hello(names: Tuple[str]) -> None:
#     for name in names:
#         print(f"Hi {name}!")
#
```

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

### Fix Multiple Files In Place

Source: https://context7.com/lyz-code/autoimport/llms.txt

Use `fix_files` to modify multiple Python files directly. Requires `click` for file handling.

```python
from autoimport import fix_files
import click

# Fix multiple files in place
file1 = click.File("r+").convert("module1.py", None, None)
file2 = click.File("r+").convert("module2.py", None, None)

fix_files((file1, file2))
# Files are modified in place
```

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

### Disabling Move to Top

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

Set disable_move_to_top to true in the configuration to prevent moving imports.

```toml
[tool.autoimport]
disable_move_to_top = true
```

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

### Keep Unused Imports with Autoimport CLI

Source: https://context7.com/lyz-code/autoimport/llms.txt

Instruct autoimport to retain unused import statements instead of removing them.

```bash
autoimport --keep-unused-imports file.py
```

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

### Fix Code with Custom Common Statements

Source: https://context7.com/lyz-code/autoimport/llms.txt

Utilize `fix_code` with a custom configuration dictionary to define common statements like `np` for `numpy`.

```python
from autoimport import fix_code

# With custom common statements configuration
source_with_numpy = """
data = np.array([1, 2, 3])
"""

config = {
    "common_statements": {
        "np": "import numpy as np",
        "pd": "import pandas as pd"
    }
}
fixed = fix_code(source_with_numpy, config=config)
print(fixed)
# Output:
# import numpy as np
# 
# 
# data = np.array([1, 2, 3])
```

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

### Ignore lines with noqa or fmt comments

Source: https://context7.com/lyz-code/autoimport/llms.txt

Use specific comments to prevent autoimport from moving or removing import statements.

```python
# Prevent an import from being moved to the top
a = 1

from os import getcwd  # noqa: autoimport
getcwd()

# Prevent inline import/debug statement from being moved
def debug_function():
    import pdb; pdb.set_trace()  # fmt: skip
    return 'debugging'

# Keep an import that appears unused but is used dynamically
from mymodule import handler  # noqa: autoimport
# handler is loaded dynamically via getattr()
```

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

### Fix Code with Disabled Move to Top

Source: https://context7.com/lyz-code/autoimport/llms.txt

Configure `fix_code` using a dictionary to disable the automatic moving of import statements to the top of the file.

```python
from autoimport import fix_code

# Disable moving imports to top
source_with_inline = """
import requests

requests.get('https://example.com')

def test():
    import os
    os.getcwd()
"""

config = {"disable_move_to_top": True}
fixed = fix_code(source_with_inline, config=config)
# Imports inside functions are left in place
```

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

### Preventing Import Movement

Source: https://github.com/lyz-code/autoimport/blob/main/docs/index.md

Use # noqa: autoimport or # fmt: skip to keep an import statement in its current position.

```python
a = 1

from os import getcwd  # noqa: autoimport

getcwd()
```

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

### Fix Files Keeping Unused Imports

Source: https://context7.com/lyz-code/autoimport/llms.txt

Use `fix_files` with `keep_unused_imports=True` to preserve unused imports, particularly useful for `__init__.py` files.

```python
# With keep_unused_imports for __init__.py
init_file = click.File("r+").convert("mypackage/__init__.py", None, None)
fix_files((init_file,), keep_unused_imports=True)
```

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

### Fix Single Python File with Autoimport CLI

Source: https://context7.com/lyz-code/autoimport/llms.txt

Use the autoimport command-line tool to fix import statements in a single Python file.

```bash
autoimport file.py
```

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

### fix_files Function

Source: https://context7.com/lyz-code/autoimport/llms.txt

The `fix_files` function processes multiple file objects, fixing their import statements in place. When reading from stdin, it returns the fixed code as a string instead of modifying files.

```APIDOC
## fix_files Function

The `fix_files` function processes multiple file objects, fixing their import statements in place. When reading from stdin, it returns the fixed code as a string instead of modifying files.

### Parameters

- **files** (iterable of file objects) - Required - An iterable of file objects to process.
- **config** (dict) - Optional - A dictionary containing configuration options. See `Configuration via pyproject.toml` for details.
- **keep_unused_imports** (bool) - Optional - If True, unused imports will not be removed.

### Request Example

```python
from autoimport import fix_files
import click

# Fix multiple files in place
# Assuming module1.py and module2.py exist and are writable
# file1 = click.File("r+").convert("module1.py", None, None)
# file2 = click.File("r+").convert("module2.py", None, None)
# fix_files((file1, file2))

# With configuration
config = {
    "common_statements": {
        "np": "import numpy as np"
    }
}
# Assuming analysis.py exists and is writable
# file = click.File("r+").convert("analysis.py", None, None)
# fix_files((file,), config=config)

# With keep_unused_imports for __init__.py
# Assuming mypackage/__init__.py exists and is writable
# init_file = click.File("r+").convert("mypackage/__init__.py", None, None)
# fix_files((init_file,), keep_unused_imports=True)
```
```

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

### Handle TYPE_CHECKING blocks

Source: https://context7.com/lyz-code/autoimport/llms.txt

Autoimport preserves imports inside TYPE_CHECKING blocks to prevent circular import issues.

```python
from autoimport import fix_code

source = """
import os
from typing import TYPE_CHECKING

if TYPE_CHECKING:
    from .model import Book

os.getcwd()

def read_book(book: Book):
    pass
"""

fixed = fix_code(source)
# The TYPE_CHECKING block is preserved as-is
print(fixed)
# import os
# from typing import TYPE_CHECKING
#
# if TYPE_CHECKING:
#     from .model import Book
#
#
# os.getcwd()
#
#
# def read_book(book: Book):
#     pass
```

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

### Fix Code Keeping Unused Imports

Source: https://context7.com/lyz-code/autoimport/llms.txt

Use the `keep_unused_imports=True` option with `fix_code` to prevent the removal of unused import statements.

```python
from autoimport import fix_code

# Keep unused imports (useful for __init__.py files)
source_with_unused = """
import gzip
import hashlib

csv_writer = csv.DictWriter(filename, fieldnames=["name", "age"])
gzip.open(filename, 'wb')
"""

fixed = fix_code(source_with_unused, keep_unused_imports=True)
print(fixed)
# Output:
# import gzip
# import hashlib
# 
# import csv
# 
# 
# csv_writer = csv.DictWriter(filename, fieldnames=["name", "age"])
# gzip.open(filename, 'wb')
```

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

### fix_code Function

Source: https://context7.com/lyz-code/autoimport/llms.txt

The `fix_code` function is the core library function that fixes Python source code by adding missing imports, removing unused imports, and moving import statements to the top of the file. It returns the corrected source code as a string.

```APIDOC
## fix_code Function

The `fix_code` function is the core library function that fixes Python source code by adding missing imports, removing unused imports, and moving import statements to the top of the file. It returns the corrected source code as a string.

### Parameters

- **source** (str) - Required - The Python source code as a string.
- **config** (dict) - Optional - A dictionary containing configuration options. See `Configuration via pyproject.toml` for details.
- **keep_unused_imports** (bool) - Optional - If True, unused imports will not be removed.

### Request Example

```python
from autoimport import fix_code

# Basic usage - fix missing and unused imports
source = """
import requests

def hello(names: Tuple[str]) -> None:
    for name in names:
        print(f"Hi {name}!")

os.getcwd()
"""

fixed = fix_code(source)
print(fixed)

# With custom common statements configuration
source_with_numpy = """
data = np.array([1, 2, 3])
"""

config = {
    "common_statements": {
        "np": "import numpy as np",
        "pd": "import pandas as pd"
    }
}
fixed = fix_code(source_with_numpy, config=config)
print(fixed)

# Keep unused imports (useful for __init__.py files)
source_with_unused = """
import gzip
import hashlib

csv_writer = csv.DictWriter(filename, fieldnames=["name", "age"])
gzip.open(filename, 'wb')
"""

fixed = fix_code(source_with_unused, keep_unused_imports=True)
print(fixed)

# Disable moving imports to top
source_with_inline = """
import requests

requests.get('https://example.com')

def test():
    import os
    os.getcwd()
"""

config = {"disable_move_to_top": True}
fixed = fix_code(source_with_inline, config=config)
```

### Response Example

```python
# Output for basic usage:
# import os
# from typing import Tuple
# 
# 
def hello(names: Tuple[str]) -> None:
#     for name in names:
#         print(f"Hi {name}!")
# 
# os.getcwd()

# Output for custom common statements:
# import numpy as np
# 
# 
# data = np.array([1, 2, 3])

# Output for keep_unused_imports=True:
# import gzip
# import hashlib
# 
# import csv
# 
# 
# csv_writer = csv.DictWriter(filename, fieldnames=["name", "age"])
# gzip.open(filename, 'wb')
```
```

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