playwrightnb

This is a Python module that lets you use sync mode Playwright interactively, inside a Jupyter notebook.

To use it:

from playwrightnb import get_page
page = get_page()
page.goto('http://example.org');
# ... do things with `page`...
page.stop()

Unlike non-jupyter usage, you don't use a context manager, but instead use stop when you're done to close the browser session (or to be more precise -- internally we still use a context manager, but we patch it to not auto-close). get_page only returns the page object, not the browser or playwright objects. You should still be able to perform most common web scraping tasks effectively, such as:

• Navigating to pages using page.goto()
• Waiting for elements to load with page.wait_for_selector()
• Extracting data from the page using methods like page.query_selector(), page.query_selector_all(), page.text_content(), etc.
• Interacting with elements using page.click(), page.type(), page.hover(), and more
• Evaluating JavaScript with page.evaluate()

However, there are a few capabilities you might miss out on without direct access to the browser or playwright objects:

1. Launching multiple browser contexts or pages simultaneously
2. Configuring browser-specific settings or permissions
3. Accessing browser-level methods like browser.new_page() or browser.close()

For most scraping scenarios, having the page object alone should suffice. But if you need more advanced control, use the following code as a starting point (which is identical to the source of get_page):

with sync_playwright() as p:
    browser = p.chromium.launch(*args, **kw)
    page = browser.new_context().new_page()
    page.stop = p.stop