diff --git a/.flake8 b/.flake8
new file mode 100644
index 00000000..b33811f1
--- /dev/null
+++ b/.flake8
@@ -0,0 +1,2 @@
+[flake8]
+ignore = E501, W503 
\ No newline at end of file
diff --git a/.github/workflows/python-package.yml b/.github/workflows/python-package.yml
new file mode 100644
index 00000000..865b1408
--- /dev/null
+++ b/.github/workflows/python-package.yml
@@ -0,0 +1,40 @@
+# This workflow will install Python dependencies, run tests and lint with a variety of Python versions
+# For more information see: https://docs.github.com/en/actions/automating-builds-and-tests/building-and-testing-python
+
+name: Python package
+
+on:
+  push:
+    branches: [ "master" ]
+  pull_request:
+    branches: [ "master" ]
+
+jobs:
+  build:
+
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.8", "3.9", "3.10", "3.11", "3.12", "3.13", "3.14"]
+
+    steps:
+    - uses: actions/checkout@v4
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v3
+      with:
+        python-version: ${{ matrix.python-version }}
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        python -m pip install flake8 pytest
+        if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
+    - name: Lint with flake8
+      run: |
+        # stop the build if there are Python syntax errors or undefined names
+        flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
+        # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
+        flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
+    - name: Test with pytest
+      run: |
+        pytest
diff --git a/.gitignore b/.gitignore
index d8961065..b532e65e 100644
--- a/.gitignore
+++ b/.gitignore
@@ -14,4 +14,5 @@ nosetests.xml
 .idea
 .cache
 /.noseids
-/.venv
\ No newline at end of file
+/.venv
+/poetry.lock
\ No newline at end of file
diff --git a/Makefile b/Makefile
index 012e4b78..9caf08a5 100644
--- a/Makefile
+++ b/Makefile
@@ -4,7 +4,7 @@ PY := .venv/bin/python
 PIP := .venv/bin/pip
 PEP8 := .venv/bin/pep8
 NOSE := .venv/bin/nosetests
-TWINE := twine
+TWINE := .venv/bin/twine
 
 # ###########
 # Tests rule!
@@ -24,7 +24,7 @@ all: setup develop
 venv: .venv/bin/python
 
 setup: venv
-	$(PIP) install -r requirements-dev.txt
+	$(PIP) install -r requirements-dev.txt | grep -v "already satisfied" || true
 
 .venv/bin/python:
 	test -d .venv || which python3 && python3 -m venv .venv || virtualenv .venv
@@ -45,11 +45,16 @@ develop: .venv/lib/python*/site-packages/readability-lxml.egg-link
 .PHONY: clean_all
 clean_all: clean_venv
 
+.PHONY: build
+build:
+	poetry build
+
 # ###########
 # Deploy
 # ###########
 .PHONY: dist
 dist:
+	$(PY) -m pip install wheel
 	$(PY) setup.py sdist bdist_wheel
 	$(TWINE) check dist/*
 
@@ -57,6 +62,12 @@ dist:
 upload:
 	$(TWINE) upload dist/*
 
-.PHONY: version_update
-version_update:
-	$(EDITOR) setup.py
+.PHONY: bump
+bump:
+	$(EDITOR) readability/__init__.py
+	$(eval VERSION := $(shell grep "__version__" readability/__init__.py | cut -d'"' -f2))
+	# fix first occurrence of version in pyproject.toml
+	sed -i '0,/version = ".*"/s//version = "$(VERSION)"/' pyproject.toml
+	git commit -m "Bump version to $(VERSION)" pyproject.toml readability/__init__.py
+	git tag $(VERSION)
+	git push --tags
diff --git a/README.md b/README.md
new file mode 100644
index 00000000..e09a515a
--- /dev/null
+++ b/README.md
@@ -0,0 +1,67 @@
+[![PyPI version](https://img.shields.io/pypi/v/readability-lxml.svg)](https://pypi.python.org/pypi/readability-lxml)
+
+# python-readability
+
+Given an HTML document, extract and clean up the main body text and title.
+
+This is a Python port of a Ruby port of [arc90's Readability project](https://web.archive.org/web/20130519040221/http://www.readability.com/).
+
+## Installation
+
+It's easy using `pip`, just run:
+
+```bash
+$ pip install readability-lxml
+```
+
+As an alternative, you may also use conda to install, just run:
+
+```bash
+$ conda install -c conda-forge readability-lxml
+```
+
+## Usage
+
+```python
+>>> import requests
+>>> from readability import Document
+
+>>> response = requests.get('http://example.com')
+>>> doc = Document(response.content)
+>>> doc.title()
+'Example Domain'
+
+>>> doc.summary()
+"""<html><body><div><body id="readabilityBody">\n<div>\n    <h1>Example Domain</h1>\n
+<p>This domain is established to be used for illustrative examples in documents. You may
+use this\n    domain in examples without prior coordination or asking for permission.</p>
+\n    <p><a href="http://www.iana.org/domains/example">More information...</a></p>\n</div>
+\n</body>\n</div></body></html>"""
+```
+
+## Change Log
+- 0.8.4 Better CJK support, thanks @cdhigh
+- 0.8.3.1 Support for python 3.8 - 3.13
+- 0.8.3 We can now save all images via keep_all_images=True (default is to save 1 main image), thanks @botlabsDev
+- 0.8.2 Added article author(s) (thanks @mattblaha)
+- 0.8.1 Fixed processing of non-ascii HTMLs via regexps.
+- 0.8 Replaced XHTML output with HTML5 output in summary() call.
+- 0.7.1 Support for Python 3.7 . Fixed a slowdown when processing documents with lots of spaces.
+- 0.7 Improved HTML5 tags handling. Fixed stripping unwanted HTML nodes (only first matching node was removed before).
+- 0.6 Finally a release which supports Python versions 2.6, 2.7, 3.3 - 3.6
+- 0.5 Preparing a release to support Python versions 2.6, 2.7, 3.3 and 3.4
+- 0.4 Added Videos loading and allowed more images per paragraph
+- 0.3 Added Document.encoding, positive\_keywords and negative\_keywords
+
+## Licensing
+
+This code is under [the Apache License 2.0](http://www.apache.org/licenses/LICENSE-2.0) license.
+
+## Thanks to
+
+- Latest [readability.js](https://github.com/MHordecki/readability-redux/blob/master/readability/readability.js)
+- Ruby port by starrhorne and iterationlabs
+- [Python port](https://github.com/gfxmonk/python-readability) by gfxmonk
+- [Decruft effort](https://web.archive.org/web/20110214150709/https://www.minvolai.com/blog/decruft-arc90s-readability-in-python/) to move to lxml
+- "BR to P" fix from readability.js which improves quality for smaller texts
+- Github users contributions.
diff --git a/README.rst b/README.rst
deleted file mode 100644
index 9b0a8b71..00000000
--- a/README.rst
+++ /dev/null
@@ -1,76 +0,0 @@
-.. image:: https://travis-ci.org/buriy/python-readability.svg?branch=master
-    :target: https://travis-ci.org/buriy/python-readability
-.. image:: https://img.shields.io/pypi/v/readability-lxml.svg
-    :target: https://pypi.python.org/pypi/readability-lxml
-
-python-readability
-==================
-
-Given an HTML document, extract and clean up the main body text and title.
-
-This is a Python port of a Ruby port of `arc90's Readability
-project <https://web.archive.org/web/20130519040221/http://www.readability.com/>`__.
-
-Installation
-------------
-
-It's easy using ``pip``, just run:
-
-.. code-block:: bash
-
-    $ pip install readability-lxml
-
-As an alternative, you may also use conda to install, just run:
-
-.. code-block:: bash
-
-    $ conda install -c conda-forge readability-lxml 
-
-Usage
------
-
-.. code-block:: python
-
-    >>> import requests
-    >>> from readability import Document
-
-    >>> response = requests.get('http://example.com')
-    >>> doc = Document(response.content)
-    >>> doc.title()
-    'Example Domain'
-
-    >>> doc.summary()
-    """<html><body><div><body id="readabilityBody">\n<div>\n    <h1>Example Domain</h1>\n
-    <p>This domain is established to be used for illustrative examples in documents. You may
-    use this\n    domain in examples without prior coordination or asking for permission.</p>
-    \n    <p><a href="http://www.iana.org/domains/example">More information...</a></p>\n</div>
-    \n</body>\n</div></body></html>"""
-
-Change Log
-----------
-
--  0.8.2 Added article author(s) (thanks @mattblaha)
--  0.8.1 Fixed processing of non-ascii HTMLs via regexps.
--  0.8 Replaced XHTML output with HTML5 output in summary() call.
--  0.7.1 Support for Python 3.7 . Fixed a slowdown when processing documents with lots of spaces.
--  0.7 Improved HTML5 tags handling. Fixed stripping unwanted HTML nodes (only first matching node was removed before).
--  0.6 Finally a release which supports Python versions 2.6, 2.7, 3.3 - 3.6
--  0.5 Preparing a release to support Python versions 2.6, 2.7, 3.3 and 3.4
--  0.4 Added Videos loading and allowed more images per paragraph
--  0.3 Added Document.encoding, positive\_keywords and negative\_keywords
-
-Licensing
----------
-
-This code is under `the Apache License
-2.0 <http://www.apache.org/licenses/LICENSE-2.0>`__ license.
-
-Thanks to
----------
-
--  Latest `readability.js <https://github.com/MHordecki/readability-redux/blob/master/readability/readability.js>`__
--  Ruby port by starrhorne and iterationlabs
--  `Python port <https://github.com/gfxmonk/python-readability>`__ by gfxmonk
--  `Decruft effort <https://web.archive.org/web/20110214150709/https://www.minvolai.com/blog/decruft-arc90s-readability-in-python/>` to move to lxml
--  "BR to P" fix from readability.js which improves quality for smaller texts
--  Github users contributions.
diff --git a/pyproject.toml b/pyproject.toml
new file mode 100644
index 00000000..b377690a
--- /dev/null
+++ b/pyproject.toml
@@ -0,0 +1,25 @@
+[tool.poetry]
+name = "readability-lxml"
+version = "0.8.4.1"
+description = "fast html to text parser (article readability tool) with python 3 support"
+authors = ["Yuri Baburov <burchik@gmail.com>"]
+license = "Apache License 2.0"
+readme = "README.md"
+packages = [
+    { include = "readability" },
+]
+
+[tool.poetry.dependencies]
+python = ">=3.8.2,<3.15"
+chardet = "^5.2.0"
+cssselect = [
+    { version = "~1.2", markers = "python_version < '3.9'" },
+    { version = "~1.3", markers = "python_version >= '3.9'" }
+]
+lxml = {extras = ["html-clean"], version = "^5.4.0"}
+lxml-html-clean = {markers = "python_version < \"3.11\"", version = "^0.4.2"}
+
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"
diff --git a/readability/__init__.py b/readability/__init__.py
index 6a263bf7..b36f021d 100644
--- a/readability/__init__.py
+++ b/readability/__init__.py
@@ -1,3 +1,3 @@
-__version__ = "0.8.2"
+__version__ = "0.8.4.1"
 
 from .readability import Document
diff --git a/readability/cleaners.py b/readability/cleaners.py
index 69825c6b..e0b07260 100644
--- a/readability/cleaners.py
+++ b/readability/cleaners.py
@@ -1,6 +1,9 @@
 # strip out a set of nuisance html attributes that can mess up rendering in RSS feeds
 import re
-from lxml.html.clean import Cleaner
+try:
+    from lxml.html.clean import Cleaner
+except ImportError:
+    from lxml_html_clean import Cleaner
 
 bad_attrs = ["width", "height", "style", "[-a-z]*color", "background[-a-z]*", "on*"]
 single_quoted = "'[^']+'"
diff --git a/readability/encoding.py b/readability/encoding.py
index c95cc14d..08332df0 100644
--- a/readability/encoding.py
+++ b/readability/encoding.py
@@ -1,9 +1,8 @@
 import re
 try:
-    import cchardet
+    import cchardet as chardet
 except ImportError:
     import chardet
-import sys
 
 
 RE_CHARSET = re.compile(r'<meta.*?charset=["\']*(.+?)["\'>]', flags=re.I)
diff --git a/readability/htmls.py b/readability/htmls.py
index 87299f5a..d99a9f53 100644
--- a/readability/htmls.py
+++ b/readability/htmls.py
@@ -110,27 +110,34 @@ def shorten_title(doc):
             if e.text_content():
                 add_match(candidates, e.text_content(), orig)
 
+    cjk = re.compile('[\u4e00-\u9fff]+')
+
     if candidates:
         title = sorted(candidates, key=len)[-1]
     else:
         for delimiter in [" | ", " - ", " :: ", " / "]:
             if delimiter in title:
                 parts = orig.split(delimiter)
-                if len(parts[0].split()) >= 4:
-                    title = parts[0]
+                p0 = parts[0]
+                pl = parts[-1]
+                if (len(p0.split()) >= 4) or (len(p0) >= 4 and cjk.search(p0)):
+                    title = p0
                     break
-                elif len(parts[-1].split()) >= 4:
-                    title = parts[-1]
+                elif (len(pl.split()) >= 4) or (len(pl) >= 4 and cjk.search(pl)):
+                    title = pl
                     break
         else:
             if ": " in title:
-                parts = orig.split(": ")
-                if len(parts[-1].split()) >= 4:
-                    title = parts[-1]
+                p1 = orig.split(": ")[-1]
+                if (len(p1.split()) >= 4) or (len(p1) >= 4 and cjk.search(p1)):
+                    title = p1
                 else:
                     title = orig.split(": ", 1)[1]
 
-    if not 15 < len(title) < 150:
+    if cjk.search(title):
+        if not (4 <= len(title) < 100):  # Allow length >= 4, cap at 100
+            return orig
+    elif not 15 < len(title) < 150:
         return orig
 
     return title
diff --git a/readability/readability.py b/readability/readability.py
index c86e7d17..c5739056 100755
--- a/readability/readability.py
+++ b/readability/readability.py
@@ -42,11 +42,11 @@
     "divToPElementsRe": re.compile(
         r"<(a|blockquote|dl|div|img|ol|p|pre|table|ul)", re.I
     ),
-    #'replaceBrsRe': re.compile(r'(<br[^>]*>[ \n\r\t]*){2,}',re.I),
-    #'replaceFontsRe': re.compile(r'<(\/?)font[^>]*>',re.I),
-    #'trimRe': re.compile(r'^\s+|\s+$/'),
-    #'normalizeRe': re.compile(r'\s{2,}/'),
-    #'killBreaksRe': re.compile(r'(<br\s*\/?>(\s|&nbsp;?)*){1,}/'),
+    # 'replaceBrsRe': re.compile(r'(<br[^>]*>[ \n\r\t]*){2,}',re.I),
+    # 'replaceFontsRe': re.compile(r'<(\/?)font[^>]*>',re.I),
+    # 'trimRe': re.compile(r'^\s+|\s+$/'),
+    # 'normalizeRe': re.compile(r'\s{2,}/'),
+    # 'killBreaksRe': re.compile(r'(<br\s*\/?>(\s|&nbsp;?)*){1,}/'),
     "videoRe": re.compile(r"https?:\/\/(www\.)?(youtube|vimeo)\.com", re.I),
     # skipFootnoteLink:      /^\s*(\[?[a-z0-9]{1,2}\]?|^|edit|citation needed)\s*$/i,
 }
@@ -210,12 +210,13 @@ def get_clean_html(self):
         """
         return clean_attributes(tounicode(self.html, method="html"))
 
-    def summary(self, html_partial=False):
+    def summary(self, html_partial=False, keep_all_images=False):
         """
         Given a HTML file, extracts the text of the article.
 
         :param html_partial: return only the div of the document, don't wrap
                              in html and body tags.
+        :param keep_all_images: Keep all images in summary.
 
         Warning: It mutates internal DOM representation of the HTML document,
         so it is better to call other API methods before this one.
@@ -257,7 +258,7 @@ def summary(self, html_partial=False):
                         article = self.html.find("body")
                         if article is None:
                             article = self.html
-                cleaned_article = self.sanitize(article, candidates)
+                cleaned_article = self.sanitize(article, candidates, keep_all_images)
 
                 article_length = len(cleaned_article or "")
                 retry_length = self.retry_length
@@ -502,7 +503,7 @@ def reverse_tags(self, node, *tag_names):
         for tag_name in tag_names:
             yield from reversed(node.findall(".//%s" % tag_name))
 
-    def sanitize(self, node, candidates):
+    def sanitize(self, node, candidates, keep_all_images=False):
         MIN_LEN = self.min_text_length
         for header in self.tags(node, "h1", "h2", "h3", "h4", "h5", "h6"):
             if self.class_weight(header) < 0 or self.get_link_density(header) > 0.33:
@@ -563,8 +564,8 @@ def sanitize(self, node, candidates):
                 to_remove = False
                 reason = ""
 
-                # if el.tag == 'div' and counts["img"] >= 1:
-                #    continue
+                if keep_all_images and el.tag == 'div' and counts["img"] >= 1:
+                    continue
                 if counts["p"] and counts["img"] > 1 + counts["p"] * 1.3:
                     reason = "too many images (%s)" % counts["img"]
                     to_remove = True
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 4731fa9d..996bbfc0 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -1,6 +1,3 @@
-lxml
-chardet
 nose
-pep8
-coverage
-wrapt-timeout-decorator
+twine
+flake8
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
deleted file mode 100644
index d6e1198b..00000000
--- a/requirements.txt
+++ /dev/null
@@ -1 +0,0 @@
--e .
diff --git a/setup.py b/setup.py
index 26894d47..dfb0846f 100755
--- a/setup.py
+++ b/setup.py
@@ -9,14 +9,8 @@
      "cchardet",
 ]
 
-test_deps = [
-    # Test timeouts
-    "wrapt-timeout-decorator",
-]
-
 extras = {
     'speed': speed_deps,
-    'test': test_deps,
 }
 
 # Adapted from https://github.com/pypa/pip/blob/master/setup.py
@@ -43,8 +37,8 @@ def find_version(*file_paths):
     author_email="burchik@gmail.com",
     description="fast html to text parser (article readability tool) with python 3 support",
     test_suite="tests.test_article_only",
-    long_description=open("README.rst").read(),
-    long_description_content_type='text/x-rst',
+    long_description=open("README.md").read(),
+    long_description_content_type="text/markdown",
     license="Apache License 2.0",
     url="http://github.com/buriy/python-readability",
     packages=["readability"],
@@ -54,7 +48,6 @@ def find_version(*file_paths):
         "lxml-html-clean; python_version < '3.11'",
         "cssselect"
     ],
-    tests_require=test_deps,
     extras_require=extras,
     classifiers=[
         "Environment :: Web Environment",
@@ -72,6 +65,7 @@ def find_version(*file_paths):
         "Programming Language :: Python :: 3.11",
         "Programming Language :: Python :: 3.12",
         "Programming Language :: Python :: 3.13",
+        "Programming Language :: Python :: 3.14",
         "Programming Language :: Python :: Implementation :: PyPy",
     ],
 )
diff --git a/tests/samples/summary-keep-all-images.sample.html b/tests/samples/summary-keep-all-images.sample.html
new file mode 100644
index 00000000..127683fc
--- /dev/null
+++ b/tests/samples/summary-keep-all-images.sample.html
@@ -0,0 +1,29 @@
+<!DOCTYPE html>
+<html lang="en">
+<head></head>
+<body>
+<h2>
+    <span>
+        H2 Headline H2 Headline H2 Headline H2 Headline H2 Headline H2 Headline H2 Headline H2 Headline H2 Headline H2 Headline
+    </span>
+</h2>
+<p>
+    <spa>
+        Text Text Text Text Text Text Text Text Text Text
+    </spa>
+</p>
+<div>
+    <span>
+        <a>
+            <img src="data:image/png;base64,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"
+            />
+         </a>
+    </span>
+</div>
+<p>
+    <spa>
+        Text Text Text Text Text Text Text Text Text Text
+    </spa>
+</p>
+</body>
+</html>
\ No newline at end of file
diff --git a/tests/test_article_only.py b/tests/test_article_only.py
index c5592cfb..fe322121 100644
--- a/tests/test_article_only.py
+++ b/tests/test_article_only.py
@@ -1,8 +1,34 @@
 import os
+import time
 import unittest
 
 from readability import Document
-from wrapt_timeout_decorator import *
+from functools import wraps
+
+
+class TimeoutException(Exception):
+    """Exception raised when a function exceeds its time limit."""
+    pass
+
+
+def timeout(seconds):
+    """Decorator to enforce a timeout on function execution."""
+    def decorator(func):
+        @wraps(func)
+        def wrapper(*args, **kwargs):
+            start_time = time.perf_counter()
+            result = func(*args, **kwargs)
+            end_time = time.perf_counter()
+            elapsed_time = end_time - start_time
+            if elapsed_time > seconds:
+                raise TimeoutException(
+                    f"Function '{func.__name__}' exceeded time limit of {seconds} seconds "
+                    f"with an execution time of {elapsed_time:.4f} seconds"
+                )
+            return result
+        return wrapper
+    return decorator
+
 
 SAMPLES = os.path.join(os.path.dirname(__file__), "samples")
 
@@ -100,7 +126,7 @@ def test_correct_cleanup(self):
         assert not "aside" in s
 
     # Many spaces make some regexes run forever
-    @timeout(3, use_signals=False)
+    @timeout(3)
     def test_many_repeated_spaces(self):
         long_space = " " * 1000000
         sample = "<html><body><p>foo" + long_space + "</p></body></html>"
@@ -123,6 +149,7 @@ def test_utf8_kanji(self):
         sample = load_sample("utf-8-kanji.sample.html")
         doc = Document(sample)
         res = doc.summary()
+        assert 0 < len(res) < 10000
 
     def test_author_present(self):
         sample = load_sample("the-hurricane-rubin-carter-denzel-washington.html")
@@ -133,3 +160,74 @@ def test_author_absent(self):
         sample = load_sample("si-game.sample.html")
         doc = Document(sample)
         assert '[no-author]' == doc.author()
+
+    def test_keep_images_present(self):
+        sample = load_sample("summary-keep-all-images.sample.html")
+
+        doc = Document(sample)
+
+        assert "<img" in doc.summary(keep_all_images=True)
+
+    def test_keep_images_absent(self):
+        sample = load_sample("summary-keep-all-images.sample.html")
+
+        doc = Document(sample)
+
+        assert "<img" not in doc.summary(keep_all_images=False)
+
+    def test_keep_images_absent_by_defautl(self):
+        sample = load_sample("summary-keep-all-images.sample.html")
+
+        doc = Document(sample)
+
+        assert "<img" not in doc.summary()
+
+    def test_cjk_summary(self):
+        """Check we can extract CJK text correctly."""
+        html = """
+        <html>
+            <head>
+                <title>这是标题</title>
+            </head>
+            <body>
+                <div>一些无关紧要的内容</div>
+                <div class="article-content">
+                    <h1>主要文章标题</h1>
+                    <p>这是主要内容的第一段。</p>
+                    <p>これはコンテンツの第2段落です。</p>
+                    <p>이것은 콘텐츠의 세 번째 단락입니다.</p>
+                    <p>This is the fourth paragraph.</p>
+                </div>
+                <div>More irrelevant stuff</div>
+            </body>
+        </html>
+        """
+        doc = Document(html)
+        summary = doc.summary()
+        # Check that the main CJK content is present in the summary
+        self.assertTrue("这是主要内容的第一段" in summary)
+        self.assertTrue("これはコンテンツの第2段落です" in summary)
+        self.assertTrue("이것은 콘텐츠의 세 번째 단락입니다" in summary)
+        # Check that irrelevant content is mostly gone
+        self.assertFalse("一些无关紧要的内容" in summary)
+
+    def test_shorten_title_delimiter_bug(self):
+        """Test that shorten_title handles delimiters correctly when the last part is valid.
+
+        This specifically targets a potential bug where 'p1' might be used instead of 'pl'.
+        """
+        html = """
+        <html>
+            <head>
+                <title>Short Part | これは長いです</title>
+            </head>
+            <body>
+                <div>Content</div>
+            </body>
+        </html>
+        """
+        doc = Document(html)
+        # With the bug, this call might raise NameError: name 'p1' is not defined
+        # With the fix, it should correctly return the last part.
+        short_title = doc.short_title()
+        self.assertEqual(short_title, "これは長いです")
diff --git a/tox.ini b/tox.ini
index 926fda50..73a97fd0 100644
--- a/tox.ini
+++ b/tox.ini
@@ -5,7 +5,7 @@
 
 [tox]
 envlist =
-    py{38,39,310,311,312,313,py3}, doc
+    py{38,39,310,311,312,313,314,py3}, doc
 skip_missing_interpreters =
     True