diff --git a/.flake8 b/.flake8 new file mode 100644 index 0000000..7e9e9b5 --- /dev/null +++ b/.flake8 @@ -0,0 +1,3 @@ +[flake8] +select = E,W,F +max-line-length = 95 diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml new file mode 100644 index 0000000..1ec5855 --- /dev/null +++ b/.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 diff --git a/.gitignore b/.gitignore index f43714d..583d8a4 100644 --- a/.gitignore +++ b/.gitignore @@ -9,7 +9,6 @@ __pycache__/ build dist -*.whl *.egg-info .tox .workflow diff --git a/.travis.yml b/.travis.yml deleted file mode 100644 index 516b989..0000000 --- a/.travis.yml +++ /dev/null @@ -1,14 +0,0 @@ -language: python -install: pip install tox -script: tox -v -e $TOX_ENV -- --verbose -dist: xenial -python: - - "3.7" -sudo: false -env: - - TOX_ENV=py37-coverage - - TOX_ENV=py37pep8 - - TOX_ENV=py37 -notifications: - email: - - development@policystat.com diff --git a/README.md b/README.md new file mode 100644 index 0000000..ad2c40a --- /dev/null +++ b/README.md @@ -0,0 +1,73 @@ +# HTML Tree Diff + +Structure aware diff of XML and HTML documents. + +The intended use is to concisely show the edits that have been made in a +document, so that authors of html content can review their work. + + +## What do we mean by "HTML Tree Diff"? + +* **HTML:** The inputs to the diff function are HTML documents +* **Tree:** It considers the full XML tree structure of the inputs, not just text based changes. +* **Diff:** The output is human-readable HTML, using `` and `` tags to show the changes. + + +## Command line interface + +You can execute `htmltreediff.cli` directly as a python module, passing it html files to diff: + +``` +$ python -m htmltreediff.cli one.html two.html +

+ + one + + + two + +

+``` + + +## 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('

...one...

', '

...two...

', pretty=True)) +

+ ... + + one + + + two + + ... +

+``` + +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 very quick brown fox jumpsfoxes jump over the lazy dog. +``` + + +## Running the unit tests + +Use `pytest` to run the tests with coverage: + +```shell +$ pip install -r requirements/testing.txt +$ pytest -v +``` \ No newline at end of file diff --git a/README.rst b/README.rst deleted file mode 100644 index 16bffc5..0000000 --- a/README.rst +++ /dev/null @@ -1,74 +0,0 @@ -============== -HTML Tree Diff -============== - -.. image:: https://travis-ci.org/PolicyStat/htmltreediff.svg?branch=master - :target: https://travis-ci.org/PolicyStat/htmltreediff - -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 and 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 -

- - one - - - two - -

- - -Python API ----------- - -You can also use htmltreediff from within a python program as a library. - -For HTML Changes:: - - >>> from htmltreediff import diff - >>> print(diff('

...one...

', '

...two...

', pretty=True)) -

- ... - - one - - - two - - ... -

- -And also for text-only changes:: - - >>> print(diff( - ... 'The quick brown fox jumps over the lazy dog.', - ... 'The very quick brown foxes jump over the dog.', - ... plaintext=False, - ... )) - The very quick brown fox jumpsfoxes jump over the lazy dog. - - -Running the unit tests ----------------------- - -The unit test suite requires the packages ``nose`` and ``coverage`` to run. Just run the ``run_tests.sh`` script, and all the tests will run, with code coverage. Code coverage should always be at 100%. diff --git a/htmltreediff/diff_core.py b/htmltreediff/diff_core.py index d279961..124877b 100644 --- a/htmltreediff/diff_core.py +++ b/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): diff --git a/htmltreediff/test_html.py b/htmltreediff/test_html.py deleted file mode 100644 index f0d3bab..0000000 --- a/htmltreediff/test_html.py +++ /dev/null @@ -1,868 +0,0 @@ -from textwrap import dedent - -from nose.tools import assert_equal - -from htmltreediff.html import diff -from htmltreediff.tests import assert_html_equal -from htmltreediff.changes import distribute -from htmltreediff.html import add_class_to_empty_del_tags, fix_lists, fix_tables -from htmltreediff.util import ( - parse_minidom, - minidom_tostring, - remove_insignificant_text_nodes, - remove_xml_declaration, - get_location, -) -from htmltreediff.test_util import collapse - - -# Preprocessing - -preprocessing_cases = [ - ( - 'empty document', - '', - '', - '' - ), - ( - 'tail text', - '

one

tail', - '

one

tail', - '

one

tail', - ), - ( - 'ignore comments', - '
', - '
', - '
', - ), - ( - 'ignore style tags', - '', - '', - '', - ), - ( - 'style tag in a block of text', - '

xxxyyy

', - '

xxxyyy

', - '

xxxyyy

', - ), - ( - 'ignore font tags', - '', - '', - '', - ), - ( - 'ignore comment tags', - '', - '', - '', - ), - ( - 'illegal text nodes inside tables are not removed', - ''' - - illegal text - - - - - -
stuff
- ''', - ' illegal text
stuff
', - ' illegal text
stuff
', # noqa - ), -] - - -def test_preprocessing(): - for description, old_html, target, target_raw, in preprocessing_cases: - def test(): - dom = parse_minidom(old_html) - assert_equal(minidom_tostring(dom), target) - assert_equal(remove_xml_declaration(dom.toxml()), target_raw) - test.description = description - yield test - - -def test_remove_insignificant_text_nodes(): - html = dedent(''' - - - -

- one two three -

- - - - -
stuff
- - - ''') - target_html = ('

one two three

' - '
stuff
') - - dom = parse_minidom(html) - remove_insignificant_text_nodes(dom) - html = minidom_tostring(dom) - assert_equal(html, target_html) - - # Check that it is idempotent. - dom = parse_minidom(html) - remove_insignificant_text_nodes(dom) - html = minidom_tostring(dom) - assert_equal(html, target_html) - - -def test_remove_insignificant_text_nodes_nbsp(): - html = dedent(''' - - - - - - - AAA - - -
  
- ''') - dom = parse_minidom(html) - remove_insignificant_text_nodes(dom) - html = minidom_tostring(dom) - assert_equal( - html, - ('' - ' AAA
'), - ) - - -# Post-processing - -def test_other_node_type_inserted(): - changes = diff( - u'

foo

', - u'

foo bar

', - ) - assert_equal( - changes, - '

foo bar

', - ) - - -def test_non_printing_characters(): - changes = diff( - '', - '
\x1Ffoo\x21

\x00

bar

', - ) - assert_equal( - changes, - '

foo!

bar

' - ) - - -def test_cutoff(): - changes = diff( - '

totally

', - '

different

', - cutoff=0.2, - ) - assert_equal( - changes, - '

The differences from the previous version are too large to show ' - 'concisely.

', - ) - - -def test_html_diff_pretty(): - cases = [ - ( - 'Simple Addition', - '

one

', - '

one

two

', - dedent(''' -

one

- -

two

- - ''').strip(), - ), - ] - for test_name, old_html, new_html, pretty_changes in cases: - def test(): - changes = diff(old_html, new_html, cutoff=0.0, pretty=True) - assert_equal(pretty_changes, changes) - test.description = 'test_html_diff_pretty - %s' % test_name - yield test - - -def test_distribute(): - cases = [ - ('
  • A
  • B
    • ', - '
  • A
  • B
    • '), - ] - for original, distributed in cases: - def test(original, distributed): - original = parse_minidom(original) - distributed = parse_minidom(distributed) - node = get_location(original, [0]) - distribute(node) - assert_html_equal( - minidom_tostring(original), - minidom_tostring(distributed), - ) - yield test, original, distributed - - -def test_get_location(): - html = '
  • A
  • B
    • ' - original = parse_minidom(html) - try: - get_location(original, [10]) - raise AssertionError('ValueError not raised') - except ValueError: - pass - - -def test_fix_lists(): - cases = [ - ( - 'simple list item insert', - ''' -
      -
    1. one
      2. -
    2. two
      3.  -
    - ''', - ''' -
      -
    1. one
      2. -
    2. two
      3. -
    - ''' - ), - ( - 'multiple list item insert', - ''' -
      -
    1. one
      2. - -
    2. two
      3. -
    3. three
      4. -       -
    - ''', - ''' -
      -
    1. one
      2. -
    2. two
      3. -
    3. three
      4. -
    - ''' - ), - ( - 'simple list item delete afterward', - ''' -
      -
    1. one
      2. -
    2. one and a half
      3.  -
    - ''', - ''' -
      -
    1. one
      2. -
    2. one and a half
      3. -
    - ''' - ), - ( - 'simple list item delete first', - ''' -
      -
    1. one half
      2.  -
    2. one
      3. -
    - ''', - ''' -
      -
    1. one half
      2. -
    2. one
      3. -
    - ''' - ), - ( - 'multiple list item delete first', - ''' -
      - -
    1. one third
      2. -
    2. two thirds
      3. -       -
    3. one
      4. -
    - ''', - ''' -
      -
    1. one third
      2. -
    2. two thirds
      3. -
    3. one
      4. -
    - ''' - ), - ( - 'insert and delete separately', - ''' -
      -
    1. one
      2. -
    2. two
      3.  -
    3. three
      4. -
    4. three point five
      5.  -
    5. four
      6. -
    - ''', - ''' -
      -
    1. one
      2. -
    2. two
      3. -
    3. three -
    4. three point five
      5. -
    5. four
      6. -
    - ''' - ), - ( - 'multiple list item delete', - ''' -
      -
    1. one
      2. - -
    2. two
      3. -
    3. three
      4. -       -
    - ''', - ''' -
      -
    1. one
      2. -
    2. two
      3. -
    3. three
      4. -
    - ''' - ), - ( - 'delete only list item', - ''' -
      - -
    1. one
      2. -       -
    - ''', - ''' -
      -
    1. one
      2. -
    - ''' - ), - ( - 'LI full content change does not add another LI', - ''' -
      - -
    1. AAA
      2. -       - -
    2. BBB
      3. -       -
    - ''', - ''' -
      -
    1. AAABBB
      2. -
    - ''' - ), - ( - 'LI full content change keeps attrs', - ''' -
      - -
    1. AAA
      2. -       - -
    2. BBB
      3. -       -
    - ''', - ''' -
      -
    1. AAABBB
      2. -
    - ''' - ), - ( - 'LI changes markup internalization fix not done if next tag is not an insert', # noqa - ''' -
      - -
    1. AAA
      2. -       -
    2. BBB
      3. - -
    3. CCC
      4. -       -
    - ''', - ''' -
      -
    1. - AAA -
      2. -
    2. BBB
      3. -
    3. CCC
      4. -
    - ''', - ), - ( - 'LI changes markup internalization fix not done if next tag is not an insert', # noqa - ''' -
      - -
    1. AAA
      2. -       -
    2. BBB
      3. - -
    3. CCC
      4. -       -
    - ''', - ''' -
      -
    1. - AAA -
      2. -
    2. BBB
      3. -
    3. CCC
      4. -
    - ''', - ), - ( - 'LI after del must be ins', - ''' -
      - -
    1. AAA
      2. -       - -
    2. BBB
      3. -       - -
    3. CCC
      4. -       -
    - ''', - ''' -
      -
    1. - AAA -
      2. -
    2. BBBCCC
      3. -
    - ''', - ), - ( - 'LI changes markup internalization fix not performed if next tags child is not li', # noqa - ''' -
      - -
    1. AAA
      2. -       - - BBB - -
    - ''', - ''' -
      -
    1. - AAA -
      2. - - BBB - -
    - ''', - ), - ( - 'LI changes markup internalization fix not performed if next tags is text', # noqa - ''' -
      - -
    1. AAA
      2. -       - - BBB - -
    - ''', - ''' -
      -
    1. - AAA -
      2. - - BBB - -
    - ''', - ), - ] - for test_name, changes, fixed_changes in cases: - changes = collapse(changes) - fixed_changes = collapse(fixed_changes) - - def test(): - changes_dom = parse_minidom(changes) - fix_lists(changes_dom) - assert_html_equal(minidom_tostring(changes_dom), fixed_changes) - test.description = 'test_fix_lists - %s' % test_name - yield test - - -def test_fix_tables(): - cases = [ - ( - 'add a table row', - ''' - - - -
    A
    B
    - ''', - ''' - - - -
    A
    B
    - ''' - ), - ( - 'tbody inside ins is distributed', - ''' - - -
    A
    - ''', - ''' - - -
    A
    - ''' - ), - ( - 'tbody inside del is distributed', - ''' - - -
    A
    - ''', - ''' - - -
    A
    - ''' - ), - ( - 'thead inside ins is distributed', - ''' - - - -
    Header
    Data
    - ''', - ''' - - - -
    Header
    Data
    - ''' - ), - ( - 'thead inside del is distributed', - ''' - - - -
    Header
    Data
    - ''', - ''' - - - -
    Header
    Data
    - ''' - ), - ( - 'tfoot inside ins is distributed', - ''' - - - -
    Data
    Footer
    - ''', - ''' - - - -
    Data
    Footer
    - ''' - ), - ( - 'tfoot inside del is distributed', - ''' - - - -
    Data
    Footer
    - ''', - ''' - - - -
    Data
    Footer
    - ''' - ), - ( - 'tbody del and ins pair is internalized', - ''' - - - -
    old data
    new data
    - ''', - ''' - - -
    old datanew data
    - ''' - ), - ( - 'thead del and ins pair is internalized', - ''' - - - - -
    old header
    new header
    data
    - ''', - ''' - - - -
    old headernew header
    data
    - ''' - ), - ( - 'tfoot del and ins pair is internalized', - ''' - - - - -
    data
    old footer
    new footer
    - ''', - ''' - - - -
    data
    old footernew footer
    - ''' - ), - ( - 'tr del and ins pair is internalized', - ''' - - - - - -
    old row
    new row
    - ''', - ''' - - - - -
    old rownew row
    - ''' - ), - ( - 'remove ins and del tags at the wrong level of the table', - ''' - - - - - - - - - - - - - - - -
    A
    - ''', - ''' - - - - - - - - -
    A
    - ''', - ), - ] - for test_name, changes, fixed_changes in cases: - changes = collapse(changes) - fixed_changes = collapse(fixed_changes) - - def test(): - changes_dom = parse_minidom(changes, strict_xml=True) - fix_tables(changes_dom) - assert_html_equal(minidom_tostring(changes_dom), fixed_changes) - test.description = 'test_fix_tables - %s' % test_name - yield test - - -def test_diff_focused_on_changed_cells_when_colgroup_added(): - old_html = ( - '' - '' - '' - '
    Alphaunchanged
    Betaold value
    ' - ) - new_html = ( - '' - '' - '' - '
    Alphaunchanged
    Betanew value
    ' - ) - expected = ( - '' - '' - '' - '
    Alphaunchanged
    Betaoldnew value
    ' - ) - assert_equal(diff(old_html, new_html), expected) - - -def test_similar_rows_not_misaligned_with_colgroup(): - """ - When a colgroup is added and rows share boilerplate text, pairwise - fuzzy matching should align each old row to its positional counterpart - rather than misaligning due to non-transitive text similarity. - - The text lengths here are tuned to trigger SequenceMatcher misalignment - in the old code path; do not shorten them. - """ - old_html = ( - '' - '' # noqa E501 - '' - '' - '
    Alphashared setup textRate: checkLong notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkRinse.
    Gammashared setup textRate: checkRinse.
    ' - ) - new_html = ( - '' - '' # noqa E501 - '' # noqa E501 - '' # noqa E501 - '
    Alphashared setup textRate: changed1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: changed2Rinse.
    Gammashared setup textRate: changed3Rinse.
    ' - ) - expected = ( - '' - '' # noqa E501 - '' # noqa E501 - '' # noqa E501 - '
    Alphashared setup textRate: checkchanged1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkchanged2Rinse.
    Gammashared setup textRate: checkchanged3Rinse.
    ' - ) - assert_equal(diff(old_html, new_html), expected) - - -def test_similar_rows_not_misaligned_without_colgroup(): - """ - Same non-transitive fuzzy equality problem as above, but triggered - without a colgroup -- every row has a small change so no exact matches - exist and all rows go through fuzzy matching. - - The text lengths here are tuned to trigger SequenceMatcher misalignment - in the old code path; do not shorten them. - """ - old_html = ( - '' - '' # noqa E501 - '' - '' - '
    Alphashared setup textRate: checkLong notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkRinse.
    Gammashared setup textRate: checkRinse.
    ' - ) - new_html = ( - '' - '' # noqa E501 - '' # noqa E501 - '' # noqa E501 - '
    Alphashared setup textRate: changed1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: changed2Rinse.
    Gammashared setup textRate: changed3Rinse.
    ' - ) - expected = ( - '' - '' # noqa E501 - '' # noqa E501 - '' # noqa E501 - '
    Alphashared setup textRate: checkchanged1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkchanged2Rinse.
    Gammashared setup textRate: checkchanged3Rinse.
    ' - ) - assert_equal(diff(old_html, new_html), expected) - - -def test_add_class_to_empty_del_tags(): - cases = [ - ( - 'empty del tag', - '', - '', - ), - ( - 'del tag with space', - ' ', - ' ', - ), - ( - 'del tag with child', - '

    ', - '

    ', - ), - ( - 'del tag with spaces and characters', - ' abc ', - ' abc ', - ), - - ] - for test_name, test_input, expected_result in cases: - def test(): - dom = parse_minidom(test_input, strict_xml=True) - add_class_to_empty_del_tags(dom) - assert_html_equal(minidom_tostring(dom), expected_result) - test.description = 'test_add_class_to_empty_del_tags - %s' % test_name - yield test diff --git a/htmltreediff/test_text.py b/htmltreediff/test_text.py deleted file mode 100644 index b0718c5..0000000 --- a/htmltreediff/test_text.py +++ /dev/null @@ -1,107 +0,0 @@ -# coding: utf8 -from nose.tools import assert_equal - -from htmltreediff.html import diff -from htmltreediff.text import split_text - - -def test_text_split(): - cases = [ - ('word', - ['word']), - ('two words', - ['two', ' ', 'words']), - ('abcdef12', - ['abcdef', '12']), - ('entity"s', - ['entity', '"', 's']), - ('stuff stuff', - ['stuff', ' ', 'stuff']), - ( - 'Stuff with an ampersand A&B stuff. Stuff with a semicolon; more stuff.', # noqa - [ - 'Stuff', ' ', 'with', ' ', 'an', ' ', 'ampersand', ' ', 'A', - '&', 'B', ' ', 'stuff', '.', ' ', 'Stuff', ' ', 'with', ' ', - 'a', ' ', 'semicolon', ';', ' ', 'more', ' ', 'stuff', '.', - ], - ), - ("we're excited", - ["we're", " ", "excited"]), - ('dial 1-800-555-1234', - ['dial', ' ', '1-800-555-1234']), - ('Effective 1/2/2003', - ['Effective', ' ', '1/2/2003']), - (u'über français', - [u'über', u' ', u'français']), - (u'em dashes \u2013 \u2013', - [u'em', u' ', u'dashes', u' ', u'\u2013', u' ', u'\u2013']), - ] - for text, target in cases: - def test(): - assert_equal(split_text(text), target) - yield test - - -def test_text_diff(): - cases = [ - ( - 'sub-word changes', - 'The quick brown fox jumps over the lazy dog.', - 'The very quick brown foxes jump over the dog.', - 'The very quick brown fox jumpsfoxes jump over the lazy dog.', # noqa - ), - ( - 'special characters', - 'Assume that A < B, and A & B = {}', - 'If we assume that A < B, and A & B = {}', - 'AssumeIf we assume that A < B, and A & B = {}', # noqa - ), - ( - 'contractions', - "we were excited", - "we're excited", - "we werewe're excited", - ), - ( - 'dates', - 'Effective 1/2/2003', - 'Effective 3/4/2005', - 'Effective 1/2/20033/4/2005', - ), - ( - 'text diff with <', - 'x', - '<', - 'x<', - ), - ( - 'text diff with >', - 'x', - '>', - 'x>', - ), - ( - 'text diff with &', - 'x', - '&', - 'x&', - ), - ( - 'do not remove newlines unless necessary', - 'one two three\nfour six', - 'one three\nfour five six', - 'one two three\nfour five six', - ), - # long text diff is broken - # ( - # 'long text diff', - # open('htmltreediff/fixtures/long_diff/before.txt').read(), - # open('htmltreediff/fixtures/long_diff/after.txt').read(), - # open('htmltreediff/fixtures/long_diff/diff.html').read(), - # ), - ] - for description, old, new, changes in cases: - def test(): - assert_equal(diff(old, new, plaintext=True), changes) - test.description = 'test_text_diff - %s' % description - yield test diff --git a/htmltreediff/tests/__init__.py b/htmltreediff/tests/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/htmltreediff/test_cli.py b/htmltreediff/tests/test_cli.py similarity index 65% rename from htmltreediff/test_cli.py rename to htmltreediff/tests/test_cli.py index e5cba8b..b46f288 100644 --- a/htmltreediff/test_cli.py +++ b/htmltreediff/tests/test_cli.py @@ -1,10 +1,8 @@ import sys import tempfile -from six import StringIO +from io import StringIO from textwrap import dedent -from nose.tools import assert_equal - from htmltreediff.cli import main @@ -21,14 +19,12 @@ def test_main(): try: sys.stdout = stream = StringIO() main(argv=('', f1.name, f2.name)) - assert_equal( - stream.getvalue(), - dedent(''' -

    one

    - -

    two

    -     - ''').strip() + '\n', - ) + expected = dedent(''' +

    one

    + +

    two

    +     + ''').strip() + '\n' + assert stream.getvalue() == expected finally: sys.stdout = old_stdout diff --git a/htmltreediff/tests.py b/htmltreediff/tests/test_core.py similarity index 93% rename from htmltreediff/tests.py rename to htmltreediff/tests/test_core.py index 979eed8..071374b 100644 --- a/htmltreediff/tests.py +++ b/htmltreediff/tests/test_core.py @@ -1,28 +1,27 @@ -# coding: utf8 - from pprint import pformat from xml.dom import Node -from nose.tools import assert_equal +import pytest from htmltreediff.html import diff from htmltreediff.util import ( parse_minidom, parse_text, minidom_tostring, - html_equal, is_text, ) -from htmltreediff.test_util import ( - reverse_edit_script, - reverse_changes_html, + +from htmltreediff.tests.util import ( + assert_html_equal, + assert_html_not_equal, + assert_strip_changes, + collapse, get_edit_script, html_patch, - strip_changes_old, - strip_changes_new, - remove_attributes, - collapse, parse_cases, + remove_attributes, + reverse_changes_html, + reverse_edit_script, ) # since the test cases get automatically reversed, only include insert cases, @@ -1206,29 +1205,6 @@ def reverse_cases(cases): all_test_cases = (test_cases + reverse_test_cases + one_way_test_cases + insane_test_cases) -def assert_html_equal(a_html, b_html): - assert html_equal(a_html, b_html), ( - u'These html documents are not equal:\n%r\n====\n%r' % ( - a_html, - b_html, - ) - ) - - -def assert_html_not_equal(a_html, b_html): - assert not html_equal(a_html, b_html), ( - u'These html documents should not be equal:\n%r\n====\n%r' % ( - a_html, - b_html, - ) - ) - - -def assert_strip_changes(old_html, new_html, changes): - assert_html_equal(old_html, strip_changes_old(changes)) - assert_html_equal(new_html, strip_changes_new(changes)) - - def test_parse_comments(): assert_html_equal( minidom_tostring(parse_minidom('')), @@ -1248,7 +1224,7 @@ def test_parse_text(): text = 'test one two < & > ;' dom = parse_text(text) root = dom.documentElement - assert_equal(len(root.childNodes), 1) + assert len(root.childNodes) == 1 child = root.childNodes[0] assert is_text(child) assert child.nodeValue == text @@ -1311,57 +1287,42 @@ def test_remove_attributes(): assert_html_equal(remove_attributes(html), stripped_html) -def test_edit_script(): - # edit script output does not reverse easily, don't test the reverse cases - for case in parse_cases(test_cases + one_way_test_cases): - if not case.edit_script: - continue +def _edit_script_cases(): + return [ + case for case in parse_cases(test_cases + one_way_test_cases) + if case.edit_script + ] + - def test(): - actual_edit_script = get_edit_script(case.old_html, case.new_html) - assert_equal( - case.edit_script, - actual_edit_script, - ('These edit scripts do not match:\n%s\n!=\n%s' - % (pformat(case.edit_script), pformat(actual_edit_script))), - ) - test.description = 'test_edit_script - %s' % case.name - yield test +@pytest.mark.parametrize('case', _edit_script_cases(), ids=lambda c: c.name) +def test_edit_script(case): + actual_edit_script = get_edit_script(case.old_html, case.new_html) + assert case.edit_script == actual_edit_script, ( + 'These edit scripts do not match:\n' + f'{pformat(case.edit_script)}\n!=\n{pformat(actual_edit_script)}' + ) -def test_html_patch(): - for case in parse_cases(all_test_cases): - # check that applying the diff gives back the same new_html - def test(): - edit_script = [] - edit_script = get_edit_script(case.old_html, case.new_html) - edited_html = html_patch(case.old_html, edit_script) - assert_html_equal( - remove_attributes(edited_html), - remove_attributes(case.new_html), - ) - test.description = 'test_html_patch - %s' % case.name - yield test +@pytest.mark.parametrize('case', list(parse_cases(all_test_cases)), ids=lambda c: c.name) +def test_html_patch(case): + edit_script = get_edit_script(case.old_html, case.new_html) + edited_html = html_patch(case.old_html, edit_script) + assert_html_equal( + remove_attributes(edited_html), + remove_attributes(case.new_html), + ) -def test_cases_sanity(): - # check that removing the ins and del markup gives the original - sane_cases = (test_cases + reverse_test_cases + one_way_test_cases) - for case in parse_cases(sane_cases): - def test(): - assert_strip_changes( - case.old_html, - case.new_html, - case.target_changes, - ) - test.description = 'test_cases_sanity - %s' % case.name - yield test +@pytest.mark.parametrize( + 'case', + list(parse_cases(test_cases + reverse_test_cases + one_way_test_cases)), + ids=lambda c: c.name, +) +def test_cases_sanity(case): + assert_strip_changes(case.old_html, case.new_html, case.target_changes) -def test_html_diff(): - for case in parse_cases(all_test_cases): - def test(): - changes = diff(case.old_html, case.new_html, cutoff=0.0) - assert_html_equal(changes, case.target_changes) - test.description = 'test_html_diff - %s' % case.name - yield test +@pytest.mark.parametrize('case', list(parse_cases(all_test_cases)), ids=lambda c: c.name) +def test_html_diff(case): + changes = diff(case.old_html, case.new_html, cutoff=0.0) + assert_html_equal(changes, case.target_changes) diff --git a/htmltreediff/test_diff_core.py b/htmltreediff/tests/test_diff_core.py similarity index 73% rename from htmltreediff/test_diff_core.py rename to htmltreediff/tests/test_diff_core.py index 9f5f77a..90da59b 100644 --- a/htmltreediff/test_diff_core.py +++ b/htmltreediff/tests/test_diff_core.py @@ -1,8 +1,5 @@ -# coding: utf-8 from unittest.mock import patch -from nose.tools import assert_equal - from htmltreediff.diff_core import ( _has_fuzzy_hash_collisions, fuzzy_match_blocks, @@ -26,40 +23,37 @@ def get_dom_nodes(html): def test_has_fuzzy_hash_collisions_no_children(): - assert_equal(_has_fuzzy_hash_collisions([]), False) + assert _has_fuzzy_hash_collisions([]) is False def test_all_distinct_tags_dont_have_collisions(): - assert_equal(_has_fuzzy_hash_collisions(get_dom_nodes('

    a

    b

    ')), False) + assert _has_fuzzy_hash_collisions(get_dom_nodes('

    a

    b

    ')) is False def test_same_tag_twice_has_collision(): - assert_equal(_has_fuzzy_hash_collisions(get_dom_nodes('

    a

    b

    ')), True) + assert _has_fuzzy_hash_collisions(get_dom_nodes('

    a

    b

    ')) is True def test_text_nodes_ignored_in_collision_check(): - assert_equal(_has_fuzzy_hash_collisions(get_dom_nodes('hello

    a

    b

    ')), False) + assert _has_fuzzy_hash_collisions(get_dom_nodes('hello

    a

    b

    ')) is False # --- build_pairwise_match_matrix --- def test_build_pairwise_match_matrix_no_matches(): - assert_equal( - build_pairwise_match_matrix(['a', 'b'], ['c', 'd']), - [[False, False], [False, False]], - ) + expected = [[False, False], [False, False]] + assert build_pairwise_match_matrix(['a', 'b'], ['c', 'd']) == expected def test_build_pairwise_match_matrix_some_matches(): - assert_equal( - build_pairwise_match_matrix(['a', 'b'], ['b', 'c']), - [[False, False], [True, False]], - ) + expected = [[False, False], [True, False]] + assert build_pairwise_match_matrix(['a', 'b'], ['b', 'c']) == expected def test_build_pairwise_match_matrix_empty(): - assert_equal(build_pairwise_match_matrix([], []), []) + expected = [] + assert build_pairwise_match_matrix([], []) == expected # --- compute_longest_common_subsequence_lengths_table --- @@ -68,13 +62,13 @@ def test_build_pairwise_match_matrix_empty(): def test_lcs_lengths_no_matches(): match_matrix = [[False, False], [False, False]] lcs_lengths = compute_longest_common_subsequence_lengths_table(match_matrix) - assert_equal(lcs_lengths[0][0], 0) + assert lcs_lengths[0][0] == 0 def test_lcs_lengths_diagonal_matches_accumulate(): match_matrix = [[True, False], [False, True]] lcs_lengths = compute_longest_common_subsequence_lengths_table(match_matrix) - assert_equal(lcs_lengths[0][0], 2) + assert lcs_lengths[0][0] == 2 # --- traceback_longest_common_subsequence_matched_pairs --- @@ -83,19 +77,16 @@ def test_lcs_lengths_diagonal_matches_accumulate(): def test_traceback_no_matches_returns_no_pairs(): match_matrix = [[False, False], [False, False]] lcs_lengths = compute_longest_common_subsequence_lengths_table(match_matrix) - assert_equal( - traceback_longest_common_subsequence_matched_pairs(match_matrix, lcs_lengths), - [], - ) + result = traceback_longest_common_subsequence_matched_pairs(match_matrix, lcs_lengths) + assert result == [] def test_traceback_two_diagonal_matches_returns_correct_pairs(): match_matrix = [[True, False], [False, True]] lcs_lengths = compute_longest_common_subsequence_lengths_table(match_matrix) - assert_equal( - traceback_longest_common_subsequence_matched_pairs(match_matrix, lcs_lengths), - [(0, 0), (1, 1)], - ) + result = traceback_longest_common_subsequence_matched_pairs( + match_matrix, lcs_lengths) + assert result == [(0, 0), (1, 1)] def test_traceback_skips_unmatched_old_node_to_reach_best_lcs(): @@ -103,64 +94,56 @@ def test_traceback_skips_unmatched_old_node_to_reach_best_lcs(): # old[0] does not match; old[1] does: result should be [(1, 0)], not [] match_matrix = [[False], [True]] lcs_lengths = compute_longest_common_subsequence_lengths_table(match_matrix) - assert_equal( - traceback_longest_common_subsequence_matched_pairs(match_matrix, lcs_lengths), - [(1, 0)], - ) + result = traceback_longest_common_subsequence_matched_pairs( + match_matrix, lcs_lengths) + assert result == [(1, 0)] # --- group_consecutive_pairs_into_blocks --- def test_group_no_pairs_produces_only_sentinel(): - assert_equal( - group_consecutive_pairs_into_blocks([], 2, 3), - [(2, 3, 0)], - ) + assert group_consecutive_pairs_into_blocks([], 2, 3) == [(2, 3, 0)] def test_group_consecutive_pairs_merged_into_one_block(): - assert_equal( - group_consecutive_pairs_into_blocks([(0, 0), (1, 1)], 2, 2), - [(0, 0, 2), (2, 2, 0)], - ) + assert group_consecutive_pairs_into_blocks([(0, 0), (1, 1)], 2, 2) == [ + (0, 0, 2), (2, 2, 0)] def test_group_non_consecutive_pairs_become_separate_blocks(): - assert_equal( - group_consecutive_pairs_into_blocks([(0, 0), (2, 2)], 3, 3), - [(0, 0, 1), (2, 2, 1), (3, 3, 0)], - ) + assert group_consecutive_pairs_into_blocks([(0, 0), (2, 2)], 3, 3) == [ + (0, 0, 1), (2, 2, 1), (3, 3, 0)] # --- fuzzy_match_blocks --- def test_fuzzy_match_blocks_empty_old(): - assert_equal(fuzzy_match_blocks([], get_dom_nodes('

    hello

    ')), [(0, 1, 0)]) + assert fuzzy_match_blocks([], get_dom_nodes('

    hello

    ')) == [(0, 1, 0)] def test_fuzzy_match_blocks_empty_new(): - assert_equal(fuzzy_match_blocks(get_dom_nodes('

    hello

    '), []), [(1, 0, 0)]) + assert fuzzy_match_blocks(get_dom_nodes('

    hello

    '), []) == [(1, 0, 0)] def test_fuzzy_match_blocks_similar_text_same_tag_matches(): old = get_dom_nodes('

    Hello world

    ') new = get_dom_nodes('

    Hello earth

    ') - assert_equal(fuzzy_match_blocks(old, new), [(0, 0, 1), (1, 1, 0)]) + assert fuzzy_match_blocks(old, new) == [(0, 0, 1), (1, 1, 0)] def test_fuzzy_match_blocks_different_tag_does_not_match(): old = get_dom_nodes('

    Hello world

    ') new = get_dom_nodes('

    Hello world

    ') - assert_equal(fuzzy_match_blocks(old, new), [(1, 1, 0)]) + assert fuzzy_match_blocks(old, new) == [(1, 1, 0)] def test_fuzzy_match_blocks_unmatched_node_excluded_from_blocks(): # old[0] and old[1] fuzzy-match new[0] and new[1]; old[2] has no match old = get_dom_nodes('

    Hello world

    Foo bar

    Other

    ') new = get_dom_nodes('

    Hello earth

    Foo bar

    ') - assert_equal(fuzzy_match_blocks(old, new), [(0, 0, 2), (3, 2, 0)]) + assert fuzzy_match_blocks(old, new) == [(0, 0, 2), (3, 2, 0)] # --- table-context restriction for fuzzy_match_blocks --- diff --git a/htmltreediff/tests/test_html.py b/htmltreediff/tests/test_html.py new file mode 100644 index 0000000..a2b2f05 --- /dev/null +++ b/htmltreediff/tests/test_html.py @@ -0,0 +1,843 @@ +from textwrap import dedent + +import pytest + +from htmltreediff.changes import distribute +from htmltreediff.html import add_class_to_empty_del_tags, diff, fix_lists, fix_tables +from htmltreediff.util import ( + parse_minidom, + minidom_tostring, + remove_insignificant_text_nodes, + remove_xml_declaration, + get_location, +) + +from htmltreediff.tests.util import assert_html_equal, collapse + + +# Preprocessing + +preprocessing_cases = [ + ( + 'empty document', + '', + '', + '' + ), + ( + 'tail text', + '

    one

    tail', + '

    one

    tail', + '

    one

    tail', + ), + ( + 'ignore comments', + '
    ', + '
    ', + '
    ', + ), + ( + 'ignore style tags', + '', + '', + '', + ), + ( + 'style tag in a block of text', + '

    xxxyyy

    ', + '

    xxxyyy

    ', + '

    xxxyyy

    ', + ), + ( + 'ignore font tags', + '', + '', + '', + ), + ( + 'ignore comment tags', + '', + '', + '', + ), + ( + 'illegal text nodes inside tables are not removed', + ''' + + illegal text + + + + + +
    stuff
    + ''', + ' illegal text
    stuff
    ', + ' illegal text
    stuff
    ', # noqa E501 + ), +] + + +@pytest.mark.parametrize( + 'description,old_html,target,target_raw', + preprocessing_cases, + ids=[case[0] for case in preprocessing_cases], +) +def test_preprocessing(description, old_html, target, target_raw): + dom = parse_minidom(old_html) + assert minidom_tostring(dom) == target + assert remove_xml_declaration(dom.toxml()) == target_raw + + +def test_remove_insignificant_text_nodes(): + html = dedent(''' + + + +

    + one two three +

    + + + + +
    stuff
    + + + ''') + target_html = ('

    one two three

    ' + '
    stuff
    ') + + dom = parse_minidom(html) + remove_insignificant_text_nodes(dom) + html = minidom_tostring(dom) + assert html == target_html + + # Check that it is idempotent. + dom = parse_minidom(html) + remove_insignificant_text_nodes(dom) + html = minidom_tostring(dom) + assert html == target_html + + +def test_remove_insignificant_text_nodes_nbsp(): + html = dedent(''' + + + + + + + AAA + + +
      
    + ''') + dom = parse_minidom(html) + remove_insignificant_text_nodes(dom) + html = minidom_tostring(dom) + assert html == ( + '' + ' AAA
    ' + ) + + +# Post-processing + +def test_other_node_type_inserted(): + changes = diff( + u'

    foo

    ', + u'

    foo bar

    ', + ) + assert changes == '

    foo bar

    ' + + +def test_non_printing_characters(): + changes = diff( + '', + '
    \x1Ffoo\x21

    \x00

    bar

    ', + ) + assert changes == '

    foo!

    bar

    ' + + +def test_cutoff(): + changes = diff( + '

    totally

    ', + '

    different

    ', + cutoff=0.2, + ) + assert changes == ( + '

    The differences from the previous version are too large to show ' + 'concisely.

    ') + + +cases = [ + ( + 'Simple Addition', + '

    one

    ', + '

    one

    two

    ', + dedent(''' +

    one

    + +

    two

    +     + ''').strip(), + ), +] + + +@pytest.mark.parametrize( + 'test_name,old_html,new_html,pretty_changes', + cases, + ids=[case[0] for case in cases], +) +def test_html_diff_pretty(test_name, old_html, new_html, pretty_changes): + changes = diff(old_html, new_html, cutoff=0.0, pretty=True) + assert pretty_changes == changes + + +@pytest.mark.parametrize('original,distributed', [ + ('
  • A
  • B
    • ', + '
  • A
  • B
    • '), +]) +def test_distribute(original, distributed): + original = parse_minidom(original) + distributed = parse_minidom(distributed) + node = get_location(original, [0]) + distribute(node) + assert_html_equal( + minidom_tostring(original), + minidom_tostring(distributed), + ) + + +def test_get_location(): + html = '
  • A
  • B
    • ' + original = parse_minidom(html) + try: + get_location(original, [10]) + raise AssertionError('ValueError not raised') + except ValueError: + pass + + +cases = [ + ( + 'simple list item insert', + ''' +
      +
    1. one
      2. +
    2. two
      3.  +
    + ''', + ''' +
      +
    1. one
      2. +
    2. two
      3. +
    + ''' + ), + ( + 'multiple list item insert', + ''' +
      +
    1. one
      2. + +
    2. two
      3. +
    3. three
      4. +       +
    + ''', + ''' +
      +
    1. one
      2. +
    2. two
      3. +
    3. three
      4. +
    + ''' + ), + ( + 'simple list item delete afterward', + ''' +
      +
    1. one
      2. +
    2. one and a half
      3.  +
    + ''', + ''' +
      +
    1. one
      2. +
    2. one and a half
      3. +
    + ''' + ), + ( + 'simple list item delete first', + ''' +
      +
    1. one half
      2.  +
    2. one
      3. +
    + ''', + ''' +
      +
    1. one half
      2. +
    2. one
      3. +
    + ''' + ), + ( + 'multiple list item delete first', + ''' +
      + +
    1. one third
      2. +
    2. two thirds
      3. +       +
    3. one
      4. +
    + ''', + ''' +
      +
    1. one third
      2. +
    2. two thirds
      3. +
    3. one
      4. +
    + ''' + ), + ( + 'insert and delete separately', + ''' +
      +
    1. one
      2. +
    2. two
      3.  +
    3. three
      4. +
    4. three point five
      5.  +
    5. four
      6. +
    + ''', + ''' +
      +
    1. one
      2. +
    2. two
      3. +
    3. three
      4. +
    4. three point five
      5. +
    5. four
      6. +
    + ''' + ), + ( + 'multiple list item delete', + ''' +
      +
    1. one
      2. + +
    2. two
      3. +
    3. three
      4. +       +
    + ''', + ''' +
      +
    1. one
      2. +
    2. two
      3. +
    3. three
      4. +
    + ''' + ), + ( + 'delete only list item', + ''' +
      + +
    1. one
      2. +       +
    + ''', + ''' +
      +
    1. one
      2. +
    + ''' + ), + ( + 'LI full content change does not add another LI', + ''' +
      + +
    1. AAA
      2. +       + +
    2. BBB
      3. +       +
    + ''', + ''' +
      +
    1. AAABBB
      2. +
    + ''' + ), + ( + 'LI full content change keeps attrs', + ''' +
      + +
    1. AAA
      2. +       + +
    2. BBB
      3. +       +
    + ''', + ''' +
      +
    1. AAABBB
      2. +
    + ''' + ), + ( + 'LI changes markup internalization fix not done if next tag is not an insert', + ''' +
      + +
    1. AAA
      2. +       +
    2. BBB
      3. + +
    3. CCC
      4. +       +
    + ''', + ''' +
      +
    1. + AAA +
      2. +
    2. BBB
      3. +
    3. CCC
      4. +
    + ''', + ), + ( + 'LI after del must be ins', + ''' +
      + +
    1. AAA
      2. +       + +
    2. BBB
      3. +       + +
    3. CCC
      4. +       +
    + ''', + ''' +
      +
    1. + AAA +
      2. +
    2. BBBCCC
      3. +
    + ''', + ), + ( + 'LI changes markup internalization fix not performed if next tags child is not li', # noqa E501 + ''' +
      + +
    1. AAA
      2. +       + + BBB + +
    + ''', + ''' +
      +
    1. + AAA +
      2. + + BBB + +
    + ''', + ), + ( + 'LI changes markup internalization fix not performed if next tags is text', + ''' +
      + +
    1. AAA
      2. +       + + BBB + +
    + ''', + ''' +
      +
    1. + AAA +
      2. + + BBB + +
    + ''', + ), +] + + +@pytest.mark.parametrize( + 'test_name,changes,fixed_changes', + cases, + ids=[case[0] for case in cases], +) +def test_fix_lists(test_name, changes, fixed_changes): + changes = collapse(changes) + fixed_changes = collapse(fixed_changes) + changes_dom = parse_minidom(changes) + fix_lists(changes_dom) + assert_html_equal(minidom_tostring(changes_dom), fixed_changes) + + +cases = [ + ( + 'add a table row', + ''' + + + +
    A
    B
    + ''', + ''' + + + +
    A
    B
    + ''' + ), + ( + 'tbody inside ins is distributed', + ''' + + +
    A
    + ''', + ''' + + +
    A
    + ''' + ), + ( + 'tbody inside del is distributed', + ''' + + +
    A
    + ''', + ''' + + +
    A
    + ''' + ), + ( + 'thead inside ins is distributed', + ''' + + + +
    Header
    Data
    + ''', + ''' + + + +
    Header
    Data
    + ''' + ), + ( + 'thead inside del is distributed', + ''' + + + +
    Header
    Data
    + ''', + ''' + + + +
    Header
    Data
    + ''' + ), + ( + 'tfoot inside ins is distributed', + ''' + + + +
    Data
    Footer
    + ''', + ''' + + + +
    Data
    Footer
    + ''' + ), + ( + 'tfoot inside del is distributed', + ''' + + + +
    Data
    Footer
    + ''', + ''' + + + +
    Data
    Footer
    + ''' + ), + ( + 'tbody del and ins pair is internalized', + ''' + + + +
    old data
    new data
    + ''', + ''' + + +
    old datanew data
    + ''' + ), + ( + 'thead del and ins pair is internalized', + ''' + + + + +
    old header
    new header
    data
    + ''', + ''' + + + +
    old headernew header
    data
    + ''' + ), + ( + 'tfoot del and ins pair is internalized', + ''' + + + + +
    data
    old footer
    new footer
    + ''', + ''' + + + +
    data
    old footernew footer
    + ''' + ), + ( + 'tr del and ins pair is internalized', + ''' + + + + + +
    old row
    new row
    + ''', + ''' + + + + +
    old rownew row
    + ''' + ), + ( + 'remove ins and del tags at the wrong level of the table', + ''' + + + + + + + + + + + + + + + +
    A
    + ''', + ''' + + + + + + + + +
    A
    + ''', + ), +] + + +@pytest.mark.parametrize( + 'test_name,changes,fixed_changes', + cases, + ids=[case[0] for case in cases], +) +def test_fix_tables(test_name, changes, fixed_changes): + changes = collapse(changes) + fixed_changes = collapse(fixed_changes) + changes_dom = parse_minidom(changes, strict_xml=True) + fix_tables(changes_dom) + assert_html_equal(minidom_tostring(changes_dom), fixed_changes) + + +def test_diff_focused_on_changed_cells_when_colgroup_added(): + old_html = ( + '' + '' + '' + '
    Alphaunchanged
    Betaold value
    ' + ) + new_html = ( + '' + '' + '' + '
    Alphaunchanged
    Betanew value
    ' + ) + expected = ( + '' + '' + '' + '
    Alphaunchanged
    Betaoldnew value
    ' + ) + assert diff(old_html, new_html) == expected + + +def test_similar_rows_not_misaligned_with_colgroup(): + """ + When a colgroup is added and rows share boilerplate text, pairwise + fuzzy matching should align each old row to its positional counterpart + rather than misaligning due to non-transitive text similarity. + + The text lengths here are tuned to trigger SequenceMatcher misalignment + in the old code path; do not shorten them. + """ + old_html = ( + '' + '' # noqa E501 + '' + '' + '
    Alphashared setup textRate: checkLong notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkRinse.
    Gammashared setup textRate: checkRinse.
    ' + ) + new_html = ( + '' + '' # noqa E501 + '' # noqa E501 + '' # noqa E501 + '
    Alphashared setup textRate: changed1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: changed2Rinse.
    Gammashared setup textRate: changed3Rinse.
    ' + ) + expected = ( + '' + '' # noqa E501 + '' # noqa E501 + '' # noqa E501 + '
    Alphashared setup textRate: checkchanged1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkchanged2Rinse.
    Gammashared setup textRate: checkchanged3Rinse.
    ' + ) + assert diff(old_html, new_html) == expected + + +def test_similar_rows_not_misaligned_without_colgroup(): + """ + Same non-transitive fuzzy equality problem as above, but triggered + without a colgroup -- every row has a small change so no exact matches + exist and all rows go through fuzzy matching. + + The text lengths here are tuned to trigger SequenceMatcher misalignment + in the old code path; do not shorten them. + """ + old_html = ( + '' + '' # noqa E501 + '' + '' + '
    Alphashared setup textRate: checkLong notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkRinse.
    Gammashared setup textRate: checkRinse.
    ' + ) + new_html = ( + '' + '' # noqa E501 + '' # noqa E501 + '' # noqa E501 + '
    Alphashared setup textRate: changed1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: changed2Rinse.
    Gammashared setup textRate: changed3Rinse.
    ' + ) + expected = ( + '' + '' # noqa E501 + '' # noqa E501 + '' # noqa E501 + '
    Alphashared setup textRate: checkchanged1Long notes requiring careful monitoring and administration throughout procedure.
    Betashared setup textRate: checkchanged2Rinse.
    Gammashared setup textRate: checkchanged3Rinse.
    ' + ) + assert diff(old_html, new_html) == expected + + +cases = [ + ( + 'empty del tag', + '', + '', + ), + ( + 'del tag with space', + ' ', + ' ', + ), + ( + 'del tag with child', + '

    ', + '

    ', + ), + ( + 'del tag with spaces and characters', + ' abc ', + ' abc ', + ), + +] + + +@pytest.mark.parametrize( + 'test_name,test_input,expected_result', + cases, + ids=[case[0] for case in cases], +) +def test_add_class_to_empty_del_tags(test_name, test_input, expected_result): + dom = parse_minidom(test_input, strict_xml=True) + add_class_to_empty_del_tags(dom) + assert_html_equal(minidom_tostring(dom), expected_result) diff --git a/htmltreediff/tests/test_test_util.py b/htmltreediff/tests/test_test_util.py new file mode 100644 index 0000000..7f6cd85 --- /dev/null +++ b/htmltreediff/tests/test_test_util.py @@ -0,0 +1,40 @@ +from unittest.mock import patch + +from htmltreediff.text import WordMatcher +from htmltreediff.util import ( + check_text_similarity, + node_compare, + parse_minidom, + walk_dom, +) + + +def test_node_compare(): + del_node = list(walk_dom(parse_minidom('')))[-1] + ins_node = list(walk_dom(parse_minidom('')))[-1] + assert -1 == node_compare(del_node, ins_node) + assert 1 == node_compare(ins_node, del_node) + + +def _uses_autojunk(html): + dom = parse_minidom(html) + node = dom.documentElement.firstChild + captured = [] + original_init = WordMatcher.__init__ + + def spy_init(self, **kwargs): + captured.append(kwargs.get('autojunk', True)) + original_init(self, **kwargs) + + with patch.object(WordMatcher, '__init__', spy_init): + check_text_similarity(node, node, cutoff=0.4) + + return any(captured) + + +def test_check_text_similarity_autojunk_disabled_for_table_element(): + assert _uses_autojunk('Some cell text') is False + + +def test_check_text_similarity_autojunk_enabled_for_non_table_element(): + assert _uses_autojunk('

    Some text here

    ') is True diff --git a/htmltreediff/tests/test_text.py b/htmltreediff/tests/test_text.py new file mode 100644 index 0000000..98ec83a --- /dev/null +++ b/htmltreediff/tests/test_text.py @@ -0,0 +1,95 @@ +import pytest + +from htmltreediff.html import diff +from htmltreediff.text import split_text + + +cases = [ + ('word', ['word']), + ('two words', ['two', ' ', 'words']), + ('abcdef12', ['abcdef', '12']), + ('entity"s', ['entity', '"', 's']), + ('stuff stuff', ['stuff', ' ', 'stuff']), + ( + 'Stuff with an ampersand A&B stuff. Stuff with a semicolon; more stuff.', # noqa + [ + 'Stuff', ' ', 'with', ' ', 'an', ' ', 'ampersand', ' ', 'A', + '&', 'B', ' ', 'stuff', '.', ' ', 'Stuff', ' ', 'with', ' ', + 'a', ' ', 'semicolon', ';', ' ', 'more', ' ', 'stuff', '.', + ], + ), + ("we're excited", ["we're", " ", "excited"]), + ('dial 1-800-555-1234', ['dial', ' ', '1-800-555-1234']), + ('Effective 1/2/2003', ['Effective', ' ', '1/2/2003']), + (u'über français', [u'über', u' ', u'français']), + (u'em dash \u2013 \u2013', [u'em', u' ', u'dash', u' ', u'\u2013', u' ', u'\u2013']), +] + + +@pytest.mark.parametrize('text,target', cases) +def test_text_split(text, target): + assert split_text(text) == target + + +cases = [ + ( + 'sub-word changes', + 'The quick brown fox jumps over the lazy dog.', + 'The very quick brown foxes jump over the dog.', + 'The very quick brown fox jumpsfoxes jump over the lazy dog.', # noqa + ), + ( + 'special characters', + 'Assume that A < B, and A & B = {}', + 'If we assume that A < B, and A & B = {}', + 'AssumeIf we assume that A < B, and A & B = {}', # noqa + ), + ( + 'contractions', + "we were excited", + "we're excited", + "we werewe're excited", + ), + ( + 'dates', + 'Effective 1/2/2003', + 'Effective 3/4/2005', + 'Effective 1/2/20033/4/2005', + ), + ( + 'text diff with <', + 'x', + '<', + 'x<', + ), + ( + 'text diff with >', + 'x', + '>', + 'x>', + ), + ( + 'text diff with &', + 'x', + '&', + 'x&', + ), + ( + 'do not remove newlines unless necessary', + 'one two three\nfour six', + 'one three\nfour five six', + 'one two three\nfour five six', + ), + # long text diff is broken + # ( + # 'long text diff', + # open('htmltreediff/fixtures/long_diff/before.txt').read(), + # open('htmltreediff/fixtures/long_diff/after.txt').read(), + # open('htmltreediff/fixtures/long_diff/diff.html').read(), + # ), +] + + +@pytest.mark.parametrize('description,old,new,changes', cases, ids=[c[0] for c in cases]) +def test_text_diff(description, old, new, changes): + assert diff(old, new, plaintext=True) == changes diff --git a/htmltreediff/test_xml.py b/htmltreediff/tests/test_xml.py similarity index 94% rename from htmltreediff/test_xml.py rename to htmltreediff/tests/test_xml.py index 7fa2ac4..2df11fb 100644 --- a/htmltreediff/test_xml.py +++ b/htmltreediff/tests/test_xml.py @@ -1,8 +1,7 @@ -from nose.tools import assert_equal - -from htmltreediff.test_util import collapse -from htmltreediff.util import parse_minidom, minidom_tostring from htmltreediff.changes import dom_diff +from htmltreediff.util import parse_minidom, minidom_tostring + +from htmltreediff.tests.util import collapse test_cases = [ ( @@ -108,4 +107,4 @@ def test_xml_diff(): old_dom = parse_minidom(old_html, strict_xml=True) new_dom = parse_minidom(new_html, strict_xml=True) changes_xml = minidom_tostring(dom_diff(old_dom, new_dom)) - assert_equal(changes_xml, target) + assert changes_xml == target diff --git a/htmltreediff/test_util.py b/htmltreediff/tests/util.py similarity index 75% rename from htmltreediff/test_util.py rename to htmltreediff/tests/util.py index c25c012..a553ef2 100644 --- a/htmltreediff/test_util.py +++ b/htmltreediff/tests/util.py @@ -1,17 +1,13 @@ -from unittest.mock import patch - from htmltreediff.diff_core import Differ from htmltreediff.edit_script_runner import EditScriptRunner from htmltreediff.changes import ( split_text_nodes, sort_nodes, ) -from htmltreediff.text import WordMatcher from htmltreediff.util import ( attribute_dict, - check_text_similarity, + html_equal, minidom_tostring, - node_compare, parse_minidom, remove_node, unwrap, @@ -19,6 +15,23 @@ ) +def assert_html_equal(a_html, b_html): + assert html_equal(a_html, b_html), ( + f'These html documents are not equal:\n{a_html!r}\n====\n{b_html!r}' + ) + + +def assert_html_not_equal(a_html, b_html): + assert not html_equal(a_html, b_html), ( + f'These html documents should not be equal:\n{a_html!r}\n====\n{b_html!r}' + ) + + +def assert_strip_changes(old_html, new_html, changes): + assert_html_equal(old_html, strip_changes_old(changes)) + assert_html_equal(new_html, strip_changes_new(changes)) + + def reverse_edit_script(edit_script): if edit_script is None: return None @@ -130,34 +143,3 @@ def parse_cases(cases): else: raise ValueError('Invalid test spec: %r' % (args,)) yield case - - -def test_node_compare(): - del_node = list(walk_dom(parse_minidom('')))[-1] - ins_node = list(walk_dom(parse_minidom('')))[-1] - assert -1 == node_compare(del_node, ins_node) - assert 1 == node_compare(ins_node, del_node) - - -def _uses_autojunk(html): - dom = parse_minidom(html) - node = dom.documentElement.firstChild - captured = [] - original_init = WordMatcher.__init__ - - def spy_init(self, **kwargs): - captured.append(kwargs.get('autojunk', True)) - original_init(self, **kwargs) - - with patch.object(WordMatcher, '__init__', spy_init): - check_text_similarity(node, node, cutoff=0.4) - - return any(captured) - - -def test_check_text_similarity_autojunk_disabled_for_table_element(): - assert _uses_autojunk('Some cell text') is False - - -def test_check_text_similarity_autojunk_enabled_for_non_table_element(): - assert _uses_autojunk('

    Some text here

    ') is True diff --git a/htmltreediff/text.py b/htmltreediff/text.py index 3d2d15c..983e9ad 100644 --- a/htmltreediff/text.py +++ b/htmltreediff/text.py @@ -1,7 +1,6 @@ import re import string -import six from difflib import SequenceMatcher, _calculate_ratio @@ -29,7 +28,7 @@ def full_split(text, regex): def multi_split(text, regexes): - """ + r""" Split the text by the given regexes, in priority order. Make sure that the regex is parenthesized so that matches are returned in @@ -48,7 +47,7 @@ def multi_split(text, regexes): 'one234five| |678' """ def make_regex(s): - return re.compile(s) if isinstance(s, six.string_types) else s + return re.compile(s) if isinstance(s, str) else s regexes = [make_regex(r) for r in regexes] # Run the list of pieces through the regex split, splitting it into more diff --git a/htmltreediff/util.py b/htmltreediff/util.py index a0cc0f3..e128e37 100644 --- a/htmltreediff/util.py +++ b/htmltreediff/util.py @@ -2,8 +2,6 @@ from textwrap import dedent from xml.dom import minidom, Node -import six - from htmltreediff.text import WordMatcher, split_text # DOM utilities ## @@ -105,7 +103,7 @@ def remove_non_printing_characters(xml, replace_char=' '): non_printing_chars = range(32) replace_chars = len(non_printing_chars) * replace_char translation_map = dict(zip(non_printing_chars, replace_chars)) - return six.text_type(xml).translate(translation_map) + return str(xml).translate(translation_map) def remove_newlines(xml): diff --git a/pylint.sh b/pylint.sh deleted file mode 100755 index 398e8c2..0000000 --- a/pylint.sh +++ /dev/null @@ -1,7 +0,0 @@ -#! /bin/sh - -files=$@ -if [ -z "$files" ]; then - files="htmltreediff" -fi -pylint --rcfile $(dirname $0)/pylintrc --reports=n $files diff --git a/pylintrc b/pylintrc deleted file mode 100644 index 80cf952..0000000 --- a/pylintrc +++ /dev/null @@ -1,269 +0,0 @@ -# lint Python modules using external checkers. -# -# This is the main checker controling the other ones and the reports -# generation. It is itself both a raw checker and an astng checker in order -# to: -# * handle message activation / deactivation at the module level -# * handle some basic but necessary stats'data (number of classes, methods...) -# -[MASTER] - -# Add to the black list. It should be a base name, not a -# path. You may set this option multiple times. -## Ignore migrations because they are auto-generated. -ignore=.svn,migrations - -# Pickle collected data for later comparisons. -persistent=yes - -# List of plugins (as comma separated values of python modules names) to load, -# usually to register additional checkers. -load-plugins= - - -[MESSAGES CONTROL] - -# Enable the message(s) with the given id(s). -#enable=W0611 - -# Disable the message(s) with the given id(s). -disable=C0111,W0142,R0904,E1101,W0613,W0221,W0232,W0212 -#,C0323,C0301,C0103,E0213,C0302,C0203,W0703 - - -[REPORTS] - -# set the output format. Available formats are text, parseable, colorized and -# html -output-format=colorized - -# Include message's id in output -include-ids=yes - -# Put messages in a separate file for each module / package specified on the -# command line instead of printing them on stdout. Reports (if any) will be -# written in a file name "pylint_global.[txt|html]". -files-output=no - -# Tells wether to display a full report or only the messages -reports=yes - -# Python expression which should return a note less than 10 (10 is the highest -# note).You have access to the variables errors warning, statement which -# respectivly contain the number of errors / warnings messages and the total -# number of statements analyzed. This is used by the global evaluation report -# (R0004). -evaluation=10.0 - ((float(5 * error + warning + refactor + convention) / statement) * 10) - -# Add a comment according to your evaluation note. This is used by the global -# evaluation report (R0004). -comment=no - -# checks for -# * unused variables / imports -# * undefined variables -# * redefinition of variable from builtins or from an outer scope -# * use of variable before assigment -# -[VARIABLES] - -# Tells wether we should check for unused import in __init__ files. -init-import=no - -# A regular expression matching names used for dummy variables (i.e. not used). -## Any variable name ending in an underscore, or a single underscore, is a dummy. -dummy-variables-rgx=[a-z_][a-z0-9_]{0,30}_|_$ - -# List of additional names supposed to be defined in builtins. Remember that -# you should avoid to define new builtins when possible. -additional-builtins= - - -# try to find bugs in the code using type inference -# -[TYPECHECK] - -# Tells wether missing members accessed in mixin class should be ignored. A -# mixin class is detected if its name ends with "mixin" (case insensitive). -ignore-mixin-members=yes - -# When zope mode is activated, consider the acquired-members option to ignore -# access to some undefined attributes. -zope=yes - -generated-members=objects,is_valid,cleaned_data,media,DoesNotExist - -# checks for : -# * doc strings -# * modules / classes / functions / methods / arguments / variables name -# * number of arguments, local variables, branchs, returns and statements in -# functions, methods -# * required module attributes -# * dangerous default values as arguments -# * redefinition of function / method / class -# * uses of the global statement -# -[BASIC] - -# Required attributes for module, separated by a comma -required-attributes= - -# Regular expression which should only match functions or classes name which do -# not require a docstring -no-docstring-rgx=__.*__ - -# Regular expression which should only match correct module names -module-rgx=(([a-z_][a-z0-9_]*)|([A-Z][a-zA-Z0-9]+))$ - -# Regular expression which should only match correct module level names -const-rgx=(([a-zA-Z_][a-zA-Z1-9_]*)|(__.*__))$ - -# Regular expression which should only match correct class names -class-rgx=[A-Z_][a-zA-Z0-9]+$ - -# Regular expression which should only match correct function names -function-rgx=[a-z_][a-z0-9_]{2,35}$ - -# Regular expression which should only match correct method names -# Limit of 30 characters unless they're test_ methods -method-rgx=([a-z_][a-z0-9_]{2,30}$|(test_[a-z0-9_]+$)) - -# Regular expression which should only match correct instance attribute names -## At least 2 characters long. -attr-rgx=[a-z_][a-z0-9_]{1,30}$ - -# Regular expression which should only match correct argument names -## At least 2 characters long. -argument-rgx=[a-z_][a-z0-9_]{1,30}$ - -# Regular expression which should only match correct variable names -## At least 1 character long. -variable-rgx=[a-z_][a-z0-9_]{0,30}$ - -# Regular expression which should only match correct list comprehension / -# generator expression variable names -## Lowercase only, minimum length 1 -inlinevar-rgx=[a-z_][a-z0-9_]*$ - -# Good variable names which should always be accepted, separated by a comma -good-names=i,j,k,_ - -# Bad variable names which should always be refused, separated by a comma -bad-names=foo,bar,baz,toto,tutu,tata - -# List of builtins function names that should not be used, separated by a comma -bad-functions=apply,input - - -# checks for sign of poor/misdesign: -# * number of methods, attributes, local variables... -# * size, complexity of functions, methods -# -[DESIGN] - -# Maximum number of arguments for function / method -max-args=12 - -# Maximum number of locals for function / method body -max-locals=30 - -# Maximum number of return / yield for function / method body -max-returns=12 - -# Maximum number of branch for function / method body -max-branchs=30 - -# Maximum number of statements in function / method body -max-statements=60 - -# Maximum number of parents for a class (see R0901). -max-parents=7 - -# Maximum number of attributes for a class (see R0902). -max-attributes=20 - -# Minimum number of public methods for a class (see R0903). -min-public-methods=0 - -# Maximum number of public methods for a class (see R0904). -max-public-methods=20 - -[IMPORTS] - -# Deprecated modules which should not be used, separated by a comma -deprecated-modules=regsub,TERMIOS,Bastion,rexec - -# Create a graph of every (i.e. internal and external) dependencies in the -# given file (report R0402 must not be disabled) -import-graph= - -# Create a graph of external dependencies in the given file (report R0402 must -# not be disabled) -ext-import-graph= - -# Create a graph of internal dependencies in the given file (report R0402 must -# not be disabled) -int-import-graph= - - -# checks for : -# * methods without self as first argument -# * overridden methods signature -# * access only to existant members via self -# * attributes not defined in the __init__ method -# * supported interfaces implementation -# * unreachable code -# -[CLASSES] - -# List of interface methods to ignore, separated by a comma. This is used for -# instance to not check methods defines in Zope's Interface base class. -ignore-iface-methods=isImplementedBy,deferred,extends,names,namesAndDescriptions,queryDescriptionFor,getBases,getDescriptionFor,getDoc,getName,getTaggedValue,getTaggedValueTags,isEqualOrExtendedBy,setTaggedValue,isImplementedByInstancesOf,adaptWith,is_implemented_by - -# List of method names used to declare (i.e. assign) instance attributes. -defining-attr-methods=__init__,__new__,setUp - - -# checks for similarities and duplicated code. This computation may be -# memory / CPU intensive, so you should disable it if you experiments some -# problems. -# -[SIMILARITIES] - -# Minimum lines number of a similarity. -min-similarity-lines=10 - -# Ignore comments when computing similarities. -ignore-comments=yes - -# Ignore docstrings when computing similarities. -ignore-docstrings=yes - - -# checks for: -# * warning notes in the code like FIXME, XXX -# * PEP 263: source code with non ascii character but no encoding declaration -# -[MISCELLANEOUS] - -# List of note tags to take in consideration, separated by a comma. -notes=FIXME,XXX,TODO - - -# checks for : -# * unauthorized constructions -# * strict indentation -# * line length -# * use of <> instead of != -# -[FORMAT] - -# Maximum number of characters on a single line. -max-line-length=90 - -# Maximum number of lines in a module -max-module-lines=1000 - -# String used as indentation unit. This is usually " " (4 spaces) or "\t" (1 -# tab). -indent-string=' ' diff --git a/pyproject.toml b/pyproject.toml new file mode 100644 index 0000000..c640904 --- /dev/null +++ b/pyproject.toml @@ -0,0 +1,5 @@ +[tool.pytest.ini_options] +# Previously, coverage was at 100 but that is failing, so it's set to 99 for +# now. We can increase it back to 100 once we have more tests. +# TODO (GH#26): Increase coverage back to 100 +addopts = "--doctest-modules --cov=htmltreediff --cov-report=term-missing --cov-fail-under=99" diff --git a/requirements/default.txt b/requirements/default.txt index d6c2d19..bba8053 100644 --- a/requirements/default.txt +++ b/requirements/default.txt @@ -1,3 +1 @@ -html5lib==0.90 -lxml==4.2.5 -six==1.15.0 +lxml>=4.2.5 diff --git a/requirements/testing.txt b/requirements/testing.txt index abb1803..9955dec 100644 --- a/requirements/testing.txt +++ b/requirements/testing.txt @@ -1,3 +1,2 @@ -coverage -nose -flake8 +pytest +pytest-cov diff --git a/run_tests.sh b/run_tests.sh deleted file mode 100755 index f48ae3d..0000000 --- a/run_tests.sh +++ /dev/null @@ -1,3 +0,0 @@ -#! /bin/sh - -nosetests --verbose --with-doctest --with-coverage --cover-erase --cover-min-percentage=100 --cover-package htmltreediff $@ && find htmltreediff/ -name '*.py' | xargs flake8 diff --git a/setup.py b/setup.py index ba9d89c..d279727 100644 --- a/setup.py +++ b/setup.py @@ -1,17 +1,11 @@ #! /usr/bin/env python -# coding: utf-8 import codecs import os -try: - from setuptools import setup, find_packages -except ImportError: - from ez_setup import use_setuptools - use_setuptools() - from setuptools import setup, find_packages +from setuptools import setup, find_packages -long_description = codecs.open("README.rst", "r", "utf-8").read() +long_description = codecs.open("README.md", "r", "utf-8").read() def strip_comments(line): @@ -38,6 +32,7 @@ def get_requirements(path): scripts=[], zip_safe=False, install_requires=list(get_requirements('requirements/default.txt')), + python_requires=">=3.8", tests_require=list(get_requirements('requirements/testing.txt')), cmdclass={}, classifiers=[ @@ -46,6 +41,12 @@ def get_requirements(path): "Intended Audience :: Developers", "License :: OSI Approved :: BSD License", "Operating System :: OS Independent", + "Programming Language :: Python :: 3", + "Programming Language :: Python :: 3.8", + "Programming Language :: Python :: 3.9", + "Programming Language :: Python :: 3.10", + "Programming Language :: Python :: 3.11", + "Programming Language :: Python :: 3.12", "Topic :: Text Processing :: Markup :: HTML", "Topic :: Text Processing :: Markup :: XML", ], diff --git a/tox.ini b/tox.ini deleted file mode 100644 index 62a210c..0000000 --- a/tox.ini +++ /dev/null @@ -1,27 +0,0 @@ -# Tox (http://tox.testrun.org/) is a tool for running tests -# in multiple virtualenvs. This configuration file will run the -# test suite on all supported python versions. To use it, "pip install tox" -# and then run "tox" from this directory. - -[tox] -envlist = py37pep8, py37, py37-coverage - -[testenv] -commands = - nosetests --with-doctest [] -deps = - -r{toxinidir}/requirements/testing.txt - -# Coverage for python 3.7 -[testenv:py37-coverage] -commands = - nosetests --with-doctest --with-coverage --cover-package htmltreediff [] - -[testenv:py37pep8] -basepython = python3.7 -deps = flake8 -commands = flake8 htmltreediff - -[flake8] -select = E,W,F -max-line-length = 95 diff --git a/vendor/lxml-4.2.5-cp38-cp38-linux_x86_64.whl b/vendor/lxml-4.2.5-cp38-cp38-linux_x86_64.whl new file mode 100644 index 0000000..f9d61b0 Binary files /dev/null and b/vendor/lxml-4.2.5-cp38-cp38-linux_x86_64.whl differ