modernize, drop old python, upgrade lxml

.flake8

@@ -0,0 +1,3 @@
+[flake8]
+select = E,W,F
+max-line-length = 95

.github/workflows/ci.yml

@@ -0,0 +1,70 @@
+name: CI
+
+on:
+  pull_request:
+  workflow_dispatch:
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - python-version: "3.8"
+            # 4.2.5 support is still required but building from source is slow
+            # and has missing dependencies, so we use a prebuilt wheel
+            lxml-spec: "vendor/lxml-4.2.5-cp38-cp38-linux_x86_64.whl"
+          # lxml v4
+          - python-version: "3.8"
+            lxml-spec: "lxml>=4,<5"
+          - python-version: "3.10"
+            lxml-spec: "lxml>=4,<5"
+          - python-version: "3.11"
+            lxml-spec: "lxml>=4,<5"
+          - python-version: "3.12"
+            lxml-spec: "lxml>=4,<5"
+          # lxml v5
+          - python-version: "3.8"
+            lxml-spec: "lxml>=5,<6"
+          - python-version: "3.10"
+            lxml-spec: "lxml>=5,<6"
+          - python-version: "3.11"
+            lxml-spec: "lxml>=5,<6"
+          - python-version: "3.12"
+            lxml-spec: "lxml>=5,<6"
+          # lxml v6
+          - python-version: "3.8"
+            lxml-spec: "lxml>=6,<7"
+          - python-version: "3.10"
+            lxml-spec: "lxml>=6,<7"
+          - python-version: "3.11"
+            lxml-spec: "lxml>=6,<7"
+          - python-version: "3.12"
+            lxml-spec: "lxml>=6,<7"
+          # lxml latest
+          - python-version: "3.12"
+            lxml-spec: "lxml"
+    steps:
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
+      - uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1 # v6.3.0
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install dependencies
+        run: |
+          pip install --upgrade pip
+          pip install "${{ matrix.lxml-spec }}"
+          pip install -r requirements/default.txt -r requirements/testing.txt
+          pip install -e .
+      - name: Run tests
+        run: pytest
+
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
+      - uses: actions/setup-python@ece7cb06caefa5fff74198d8649806c4678c61a1 # v6.3.0
+        with:
+          python-version: "3.12"
+      - run: pip install flake8
+      - run: flake8 htmltreediff

.gitignore

@@ -9,7 +9,6 @@ __pycache__/
 
 build
 dist
-*.whl
 *.egg-info
 .tox
 .workflow

README.md

@@ -0,0 +1,73 @@
+# HTML Tree Diff
+
+Structure aware diff of XML and HTML documents.
+
+The intended use is to concisely show the edits that have been made in a
+document, so that authors of html content can review their work.
+
+
+## What do we mean by "HTML Tree Diff"?
+
+* **HTML:** The inputs to the diff function are HTML documents
+* **Tree:** It considers the full XML tree structure of the inputs, not just text based changes.
+* **Diff:** The output is human-readable HTML, using `<ins>` and `<del>` tags to show the changes.
+
+
+## Command line interface
+
+You can execute `htmltreediff.cli` directly as a python module, passing it html files to diff:
+
+```
+$ python -m htmltreediff.cli one.html two.html
+<h1>
+  <del>
+    one
+  </del>
+  <ins>
+    two
+  </ins>
+</h1>
+```
+
+
+## Python API
+
+You can also use htmltreediff from within a python program as a library.
+
+For HTML Changes:
+
+```python
+>>> from htmltreediff import diff
+>>> print(diff('<h1>...one...</h1>', '<h1>...two...</h1>', pretty=True))
+<h1>
+  ...
+  <del>
+    one
+  </del>
+  <ins>
+    two
+  </ins>
+  ...
+</h1>
+```
+
+And also for text-only changes:
+
+```python
+>>> print(diff(
+...     'The quick brown fox jumps over the lazy dog.',
+...     'The very quick brown foxes jump over the dog.',
+...     plaintext=False,
+... ))
+The <ins>very </ins>quick brown <del>fox jumps</del><ins>foxes jump</ins> over the<del> lazy</del> dog.
+```
+
+
+## Running the unit tests
+
+Use `pytest` to run the tests with coverage:
+
+```shell
+$ pip install -r requirements/testing.txt
+$ pytest -v
+```

htmltreediff/diff_core.py

@@ -1,8 +1,6 @@
 import difflib
 from xml.dom import Node
 
-import six
-
 from htmltreediff.lcs import matching_blocks_from_hashes
 from htmltreediff.text import is_text_junk
 from htmltreediff.util import (
@@ -277,7 +275,7 @@ def get_opcodes(matching_blocks):
 
 
 def _is_junk(hashable_node):
-    if isinstance(hashable_node, six.string_types):
+    if isinstance(hashable_node, str):
         return is_text_junk(hashable_node)
     # Nodes with no text or just whitespace are junk.
     for descendant in walk_dom(hashable_node.node):