Feat: extract folder structure for datasets (Issue #125)

[zoidy]

Description

Use playwright to scrape each dataset's page and extract the necessary folder structure. Then recreate that folder structure when downloading files via the API.

See #125

Documentation Update

• I have updated README.md and other relevant documentation
• No documentation update is needed

Implementation Notes

This method uses playwright.

• create a new folder called browsers in the mamba environment
• add a new environment variable to the environment to install browsers into that directory conda env config vars set PLAYWRIGHT_BROWSERS_PATH=<path to envs folder>/envs/rebach/browsers
• Install playwright into the environment following the docs mamba install pytest-playwright
• Install Chromium and dependencies following the docs. playwright install --with-deps --no-shell chromium

The data structure is contained in a <script> tag. This tag is stripped of any Javascript to leave a JSON string that is converted to a Python dict via json.loads(). The actual directory information is in a Python dict the form

{'<fileid1>` : `path/to/file1`, '<fileid2>` : `path/to/file2`}

where <fileid> is the Figshare file id. If the file is in the root, the value corresponding to the given file id will be an empty string.

[HafeezOJ]

Implementation: The folder structure is obtained from the article version details API response. Therefore, Playwright or any external package is no longer needed.