requests-html is not installed where your script runs. The PyPI name is `requests-html` (with a dash) but the import is `requests_html` (with an underscore). Run `python -m pip install requests-html` and the import works.

## The error

```text
ModuleNotFoundError: No module named 'requests_html'
```

## Fix it

1. **Install requests-html**

```
python -m pip install requests-html
```
Expected: `Successfully installed requests-html-...` plus its dependencies.

2. **Confirm**

```
python -c "from requests_html import HTMLSession; print('ok')"
```
Expected: prints `ok`.

## When this applies

- any `from requests_html import ...` fails naming `requests_html`
- a fresh env following a tutorial

## When it does NOT apply

- `cannot import name 'search' from 'parse'` means requests-html IS installed but its dependency broke
- the lxml_html_clean ImportError is the lxml 5.2 split

## Compatibility

requests-html 0.10.x. Note the project is unmaintained since 2021, so expect dependency friction on modern Python (the parse and lxml skills in this set cover the two big ones).

## Why it happens

The distribution name uses a dash (`requests-html`) while the import uses an underscore (`requests_html`). People who guess the pip name from the import name (`pip install requests_html`) get nothing, and people who skip the install get this ModuleNotFoundError.

## Edge cases

- `pip install requests_html` (underscore) fails outright on PyPI. The dash form is the only one that exists.
- Because the project is unmaintained, a fresh install on Python 3.12+ may immediately hit the `parse` or `lxml_html_clean` errors. Those have their own fixes.