beautifulsoup featurenotfound lxml parser not installed error fails

Fixes BeautifulSoup raising FeatureNotFound when code asks for the lxml parser but lxml is not installed in the running environment. Use when html.parser works but lxml fails. Not for malformed HTML that parses but returns empty results.

Confirm the failing call passes lxml as the parser argument to BeautifulSoup.

Install lxml through the interpreter that runs the code, using python -m pip install lxml.

Rerun a small parse test on a short HTML string with the lxml parser.

If you cannot install packages in that environment, change the parser argument to html.parser and rerun.

Add lxml to your project requirements file and restart the app or notebook kernel.

Resolved from the public problem: https://vectle.com/posts/pstU6n81XPTpckyTT11rZJmQ

Problem query: beautifulsoup featurenotfound lxml parser not installed error fails

TL;DR:

Not specified by the original author.

BeautifulSoup does not ship its own HTML parser. When you write BeautifulSoup(html, 'lxml') and lxml is not installed, you get FeatureNotFound instead of a soup. Install it with pip install lxml and the same line works. If you cannot install lxml, switch the second argument to 'html.parser' (built into Python, no install needed) and move on.

The error

bs4.FeatureNotFound: Couldn't find a tree builder with the features you requested: lxml

Fix it

1. Install lxml

pip install lxml

Expected: pip reports Successfully installed lxml-....

2. Re-run your parse line unchanged

python -c "from bs4 import BeautifulSoup; print(BeautifulSoup('hi', 'lxml').get_text())"

Expected: it prints hi. The line that raised now works because the requested tree builder exists.

3. Alternative when you cannot install anything

Change the parser argument to the built-in one:

soup = BeautifulSoup(html, 'html.parser')

Expected: parsing works with no new installs. Slightly slower and more lenient than lxml, fine for most scraping.

When this applies

• the traceback is bs4.FeatureNotFound and names lxml
• you passed 'lxml', 'lxml-xml', or 'xml' as the parser argument
• a library you use (pandas read_html, for example) asks for lxml under the hood

When it does NOT apply

• ModuleNotFoundError: No module named 'bs4' is a different problem (install beautifulsoup4)
• FeatureNotFound naming html5lib means install html5lib instead
• an ImportError on the BeautifulSoup name is the stub/shadowing problem

Compatibility

All beautifulsoup4 4.x versions. lxml wheels exist for CPython 3.8+ on Windows, macOS, and Linux; on odd platforms you may need the libxml2 system libraries to build it.

Variant phrasings

Couldn't find a tree builder with the features you requested: html5lib

Same error, different parser. Fix is pip install html5lib, or switch to 'html.parser'.

Couldn't find a tree builder with the features you requested: lxml-xml

You asked for the XML tree builder. Same fix: pip install lxml.

Why it happens

BeautifulSoup is a tree API, not a parser. The second argument to the constructor picks which installed parsing library does the actual work. If that library is not installed, BeautifulSoup cannot build the tree and raises FeatureNotFound. The message names exactly the feature set it looked for, so the fix is installing the named parser or picking one you have.

Edge cases

• In a venv, make sure you pip-install into the same environment that runs the script.
• On Alpine/musl Linux there is no lxml wheel, so pip builds from source and needs libxml2-dev and libxslt-dev system packages.
• If lxml is installed but the error persists, you have two Pythons and installed into the wrong one. python -m pip install lxml targets the interpreter you run.

Stack Overflow question 24398302 is the canonical thread for this exact FeatureNotFound string: the asker passed 'lxml' without having lxml installed, and the accepted answer is pip install lxml (with apt-get install of the XML dev libraries on Linux where no wheel applied). The same string recurs in project trackers such as nsepy#238 with the same resolution. Source: https://stackoverflow.com/questions/24398302

• The traceback ends in FeatureNotFound naming lxml as the requested feature. - The same code works on your machine but fails in a container, CI job, or fresh virtualenv. - The failure happens at BeautifulSoup construction, before any selector or network code runs.

Reproduce in the failing interpreter. Run a two-line script that imports bs4 and parses a tiny HTML string with the lxml parser, using the same python command your job uses. Expected: the same FeatureNotFound error, confirming the environment and not your page content is at fault.

Check whether lxml imports at all. Run python -c with an import of lxml.etree. Success check: if the import fails, lxml is missing or broken in this environment regardless of what pip list in another environment claims.

Install or repair lxml in this environment. Install lxml with the pip belonging to the same interpreter, or in a container rebuild the image after adding lxml to the requirements file. Expected: the import check from step 2 now passes.

Or drop the lxml dependency deliberately. If you cannot install packages where the code runs, change the parser argument from lxml to html.parser. Success check: the script runs with no extra install, accepting slightly slower and less forgiving parsing.

Re-run the real scrape on one page. Fetch and parse a single known page first. Expected: the soup object builds and your selectors return results, instead of failing at parser setup before any extraction happens.

Resolved from the public problem: https://vectle.com/posts/pstpPZp9ApDsgyJhky7iKvfg

Problem query: bs4.exceptions.featurenotfound couldn't find a tree builder with the features you requested lxml error

TL;DR: BeautifulSoup raises FeatureNotFound when you ask for a tree builder it cannot load, and with the lxml parser that almost always means lxml is not installed (or not importable) in the Python environment running the script. Either install lxml into that exact environment or change the parser argument to html.parser, which ships with Python. Verify by parsing a small page in the same interpreter that runs your scraper.