test: Rewrite parser tests to a cleaner version and adding more tests

This commit is contained in:
Karim shoair
2024-12-16 13:17:09 +02:00
parent 3ff0d55c3e
commit 69e31612a5
2 changed files with 346 additions and 336 deletions
+256 -214
View File
@@ -1,288 +1,330 @@
import pickle
import unittest
import time
import pytest
from cssselect import SelectorError, SelectorSyntaxError
from scrapling import Adaptor
class TestParser(unittest.TestCase):
def setUp(self):
self.html = '''
<html>
<head>
<title>Complex Web Page</title>
<style>
.hidden { display: none; }
</style>
</head>
<body>
<header>
<nav>
<ul>
<li><a href="#home">Home</a></li>
<li><a href="#about">About</a></li>
<li><a href="#contact">Contact</a></li>
</ul>
</nav>
</header>
<main>
<section id="products" schema='{"jsonable": "data"}'>
<h2>Products</h2>
<div class="product-list">
<article class="product" data-id="1">
<h3>Product 1</h3>
<p class="description">This is product 1</p>
<span class="price">$10.99</span>
<div class="hidden stock">In stock: 5</div>
</article>
<article class="product" data-id="2">
<h3>Product 2</h3>
<p class="description">This is product 2</p>
<span class="price">$20.99</span>
<div class="hidden stock">In stock: 3</div>
</article>
<article class="product" data-id="3">
<h3>Product 3</h3>
<p class="description">This is product 3</p>
<span class="price">$15.99</span>
<div class="hidden stock">Out of stock</div>
</article>
@pytest.fixture
def html_content():
return '''
<html>
<head>
<title>Complex Web Page</title>
<style>
.hidden { display: none; }
</style>
</head>
<body>
<header>
<nav>
<ul>
<li><a href="#home">Home</a></li>
<li><a href="#about">About</a></li>
<li><a href="#contact">Contact</a></li>
</ul>
</nav>
</header>
<main>
<section id="products" schema='{"jsonable": "data"}'>
<h2>Products</h2>
<div class="product-list">
<article class="product" data-id="1">
<h3>Product 1</h3>
<p class="description">This is product 1</p>
<span class="price">$10.99</span>
<div class="hidden stock">In stock: 5</div>
</article>
<article class="product" data-id="2">
<h3>Product 2</h3>
<p class="description">This is product 2</p>
<span class="price">$20.99</span>
<div class="hidden stock">In stock: 3</div>
</article>
<article class="product" data-id="3">
<h3>Product 3</h3>
<p class="description">This is product 3</p>
<span class="price">$15.99</span>
<div class="hidden stock">Out of stock</div>
</article>
</div>
</section>
<section id="reviews">
<h2>Customer Reviews</h2>
<div class="review-list">
<div class="review" data-rating="5">
<p class="review-text">Great product!</p>
<span class="reviewer">John Doe</span>
</div>
</section>
<section id="reviews">
<h2>Customer Reviews</h2>
<div class="review-list">
<div class="review" data-rating="5">
<p class="review-text">Great product!</p>
<span class="reviewer">John Doe</span>
</div>
<div class="review" data-rating="4">
<p class="review-text">Good value for money.</p>
<span class="reviewer">Jane Smith</span>
</div>
<div class="review" data-rating="4">
<p class="review-text">Good value for money.</p>
<span class="reviewer">Jane Smith</span>
</div>
</section>
</main>
<footer>
<p>&copy; 2024 Our Company</p>
</footer>
<script id="page-data" type="application/json">
{"lastUpdated": "2024-09-22T10:30:00Z", "totalProducts": 3}
</script>
</body>
</html>
'''
self.page = Adaptor(self.html, auto_match=False)
</div>
</section>
</main>
<footer>
<p>&copy; 2024 Our Company</p>
</footer>
<script id="page-data" type="application/json">
{"lastUpdated": "2024-09-22T10:30:00Z", "totalProducts": 3}
</script>
</body>
</html>
'''
def test_css_selector(self):
"""Test Selecting elements with complex CSS selectors"""
elements = self.page.css('main #products .product-list article.product')
self.assertEqual(len(elements), 3)
in_stock_products = self.page.css(
@pytest.fixture
def page(html_content):
return Adaptor(html_content, auto_match=False)
# CSS Selector Tests
class TestCSSSelectors:
def test_basic_product_selection(self, page):
"""Test selecting all product elements"""
elements = page.css('main #products .product-list article.product')
assert len(elements) == 3
def test_in_stock_product_selection(self, page):
"""Test selecting in-stock products"""
in_stock_products = page.css(
'main #products .product-list article.product:not(:contains("Out of stock"))')
self.assertEqual(len(in_stock_products), 2)
assert len(in_stock_products) == 2
def test_xpath_selector(self):
"""Test Selecting elements with Complex XPath selectors"""
reviews = self.page.xpath(
# XPath Selector Tests
class TestXPathSelectors:
def test_high_rating_reviews(self, page):
"""Test selecting reviews with high ratings"""
reviews = page.xpath(
'//section[@id="reviews"]//div[contains(@class, "review") and @data-rating >= 4]'
)
self.assertEqual(len(reviews), 2)
assert len(reviews) == 2
high_priced_products = self.page.xpath(
def test_high_priced_products(self, page):
"""Test selecting products above a certain price"""
high_priced_products = page.xpath(
'//article[contains(@class, "product")]'
'[number(translate(substring-after(.//span[@class="price"], "$"), ",", "")) > 15]'
)
self.assertEqual(len(high_priced_products), 2)
assert len(high_priced_products) == 2
def test_find_by_text(self):
"""Test Selecting elements with Text matching"""
stock_info = self.page.find_by_regex(r'In stock: \d+', first_match=False)
self.assertEqual(len(stock_info), 2)
stock_info = self.page.find_by_regex(r'In stock: \d+', first_match=True, case_sensitive=True)
self.assertEqual(stock_info.text, 'In stock: 5')
# Text Matching Tests
class TestTextMatching:
def test_regex_multiple_matches(self, page):
"""Test finding multiple matches with regex"""
stock_info = page.find_by_regex(r'In stock: \d+', first_match=False)
assert len(stock_info) == 2
stock_info = self.page.find_by_text(r'In stock:', partial=True, first_match=False)
self.assertEqual(len(stock_info), 2)
def test_regex_first_match(self, page):
"""Test finding the first match with regex"""
stock_info = page.find_by_regex(r'In stock: \d+', first_match=True, case_sensitive=True)
assert stock_info.text == 'In stock: 5'
out_of_stock = self.page.find_by_text('Out of stock', partial=False, first_match=False)
self.assertEqual(len(out_of_stock), 1)
def test_partial_text_match(self, page):
"""Test finding elements with partial text match"""
stock_info = page.find_by_text(r'In stock:', partial=True, first_match=False)
assert len(stock_info) == 2
def test_find_similar_elements(self):
"""Test Finding similar elements of an element"""
first_product = self.page.css_first('.product')
def test_exact_text_match(self, page):
"""Test finding elements with exact text match"""
out_of_stock = page.find_by_text('Out of stock', partial=False, first_match=False)
assert len(out_of_stock) == 1
# Similar Elements Tests
class TestSimilarElements:
def test_finding_similar_products(self, page):
"""Test finding similar product elements"""
first_product = page.css_first('.product')
similar_products = first_product.find_similar()
self.assertEqual(len(similar_products), 2)
assert len(similar_products) == 2
first_review = self.page.find('div', class_='review')
def test_finding_similar_reviews(self, page):
"""Test finding similar review elements with additional filtering"""
first_review = page.find('div', class_='review')
similar_high_rated_reviews = [
review
for review in first_review.find_similar()
if int(review.attrib.get('data-rating', 0)) >= 4
]
self.assertEqual(len(similar_high_rated_reviews), 1)
assert len(similar_high_rated_reviews) == 1
def test_expected_errors(self):
"""Test errors that should raised if it does"""
with self.assertRaises(ValueError):
# Error Handling Tests
class TestErrorHandling:
def test_invalid_adaptor_initialization(self):
"""Test various invalid Adaptor initializations"""
# No arguments
with pytest.raises(ValueError):
_ = Adaptor(auto_match=False)
with self.assertRaises(TypeError):
# Invalid argument types
with pytest.raises(TypeError):
_ = Adaptor(root="ayo", auto_match=False)
with self.assertRaises(TypeError):
with pytest.raises(TypeError):
_ = Adaptor(text=1, auto_match=False)
with self.assertRaises(TypeError):
with pytest.raises(TypeError):
_ = Adaptor(body=1, auto_match=False)
with self.assertRaises(ValueError):
_ = Adaptor(self.html, storage=object, auto_match=True)
def test_invalid_storage(self, page, html_content):
"""Test invalid storage parameter"""
with pytest.raises(ValueError):
_ = Adaptor(html_content, storage=object, auto_match=True)
def test_pickleable(self):
"""Test that objects aren't pickleable"""
table = self.page.css('.product-list')[0]
with self.assertRaises(TypeError): # Adaptors
def test_bad_selectors(self, page):
"""Test handling of invalid selectors"""
with pytest.raises((SelectorError, SelectorSyntaxError)):
page.css('4 ayo')
with pytest.raises((SelectorError, SelectorSyntaxError)):
page.xpath('4 ayo')
# Pickling and Object Representation Tests
class TestPicklingAndRepresentation:
def test_unpickleable_objects(self, page):
"""Test that Adaptor objects cannot be pickled"""
table = page.css('.product-list')[0]
with pytest.raises(TypeError):
pickle.dumps(table)
with self.assertRaises(TypeError): # Adaptor
with pytest.raises(TypeError):
pickle.dumps(table[0])
def test_overridden(self):
"""Test overridden functions"""
table = self.page.css('.product-list')[0]
self.assertTrue(issubclass(type(table.__str__()), str))
self.assertTrue(issubclass(type(table.__repr__()), str))
self.assertTrue(issubclass(type(table.attrib.__str__()), str))
self.assertTrue(issubclass(type(table.attrib.__repr__()), str))
def test_string_representations(self, page):
"""Test custom string representations of objects"""
table = page.css('.product-list')[0]
assert issubclass(type(table.__str__()), str)
assert issubclass(type(table.__repr__()), str)
assert issubclass(type(table.attrib.__str__()), str)
assert issubclass(type(table.attrib.__repr__()), str)
def test_bad_selector(self):
"""Test object can handle bad selector"""
with self.assertRaises((SelectorError, SelectorSyntaxError,)):
self.page.css('4 ayo')
with self.assertRaises((SelectorError, SelectorSyntaxError,)):
self.page.xpath('4 ayo')
def test_selectors_generation(self):
"""Try to create selectors for all elements in the page"""
def _traverse(element: Adaptor):
self.assertTrue(type(element.generate_css_selector) is str)
self.assertTrue(type(element.generate_xpath_selector) is str)
for branch in element.children:
_traverse(branch)
_traverse(self.page)
def test_getting_all_text(self):
"""Test getting all text"""
self.assertNotEqual(self.page.get_all_text(), '')
def test_element_navigation(self):
"""Test moving in the page from selected element"""
table = self.page.css('.product-list')[0]
self.assertIsNot(table.path, [])
self.assertNotEqual(table.html_content, '')
self.assertNotEqual(table.prettify(), '')
# Navigation and Traversal Tests
class TestElementNavigation:
def test_basic_navigation_properties(self, page):
"""Test basic navigation properties of elements"""
table = page.css('.product-list')[0]
assert table.path is not None
assert table.html_content != ''
assert table.prettify() != ''
def test_parent_and_sibling_navigation(self, page):
"""Test parent and sibling navigation"""
table = page.css('.product-list')[0]
parent = table.parent
self.assertEqual(parent.attrib['id'], 'products')
children = table.children
self.assertEqual(len(children), 3)
assert parent.attrib['id'] == 'products'
parent_siblings = parent.siblings
self.assertEqual(len(parent_siblings), 1)
assert len(parent_siblings) == 1
child = table.find({'data-id': "1"})
def test_child_navigation(self, page):
"""Test child navigation"""
table = page.css('.product-list')[0]
children = table.children
assert len(children) == 3
def test_next_and_previous_navigation(self, page):
"""Test next and previous element navigation"""
child = page.css('.product-list')[0].find({'data-id': "1"})
next_element = child.next
self.assertEqual(next_element.attrib['data-id'], '2')
assert next_element.attrib['data-id'] == '2'
prev_element = next_element.previous
self.assertEqual(prev_element.tag, child.tag)
assert prev_element.tag == child.tag
all_prices = self.page.css('.price')
def test_ancestor_finding(self, page):
"""Test finding ancestors of elements"""
all_prices = page.css('.price')
products_with_prices = [
price.find_ancestor(lambda p: p.has_class('product'))
for price in all_prices
]
self.assertEqual(len(products_with_prices), 3)
assert len(products_with_prices) == 3
def test_empty_return(self):
"""Test cases where functions shouldn't have results"""
test_html = """
<html>
<span id="a"><a></a><!--comment--></span>
<span id="b"><!--comment--><a></a></span>
</html>"""
soup = Adaptor(test_html, auto_match=False, keep_comments=False)
html_tag = soup.css('html')[0]
self.assertEqual(html_tag.path, [])
self.assertEqual(html_tag.siblings, [])
self.assertEqual(html_tag.parent, None)
self.assertEqual(html_tag.find_ancestor(lambda e: e), None)
self.assertEqual(soup.css('#a a')[0].next, None)
self.assertEqual(soup.css('#b a')[0].previous, None)
def test_text_to_json(self):
"""Test converting text to json"""
script_content = self.page.css('#page-data::text')[0]
self.assertTrue(issubclass(type(script_content.sort()), str))
# JSON and Attribute Tests
class TestJSONAndAttributes:
def test_json_conversion(self, page):
"""Test converting content to JSON"""
script_content = page.css('#page-data::text')[0]
assert issubclass(type(script_content.sort()), str)
page_data = script_content.json()
self.assertEqual(page_data['totalProducts'], 3)
self.assertTrue('lastUpdated' in page_data)
assert page_data['totalProducts'] == 3
assert 'lastUpdated' in page_data
def test_regex_on_text(self):
"""Test doing regex on a selected text"""
element = self.page.css('[data-id="1"] .price')[0]
match = element.re_first(r'[\.\d]+')
self.assertEqual(match, '10.99')
match = element.text.re(r'(\d+)', replace_entities=False)
self.assertEqual(len(match), 2)
def test_attribute_operations(self):
"""Test operations on elements attributes"""
products = self.page.css('.product')
def test_attribute_operations(self, page):
"""Test various attribute-related operations"""
# Product ID extraction
products = page.css('.product')
product_ids = [product.attrib['data-id'] for product in products]
self.assertEqual(product_ids, ['1', '2', '3'])
self.assertTrue('data-id' in products[0].attrib)
assert product_ids == ['1', '2', '3']
assert 'data-id' in products[0].attrib
reviews = self.page.css('.review')
# Review rating calculations
reviews = page.css('.review')
review_ratings = [int(review.attrib['data-rating']) for review in reviews]
self.assertEqual(sum(review_ratings) / len(review_ratings), 4.5)
assert sum(review_ratings) / len(review_ratings) == 4.5
# Attribute searching
key_value = list(products[0].attrib.search_values('1', partial=False))
self.assertEqual(list(key_value[0].keys()), ['data-id'])
assert list(key_value[0].keys()) == ['data-id']
key_value = list(products[0].attrib.search_values('1', partial=True))
self.assertEqual(list(key_value[0].keys()), ['data-id'])
assert list(key_value[0].keys()) == ['data-id']
attr_json = self.page.css_first('#products').attrib['schema'].json()
self.assertEqual(attr_json, {'jsonable': 'data'})
self.assertEqual(type(self.page.css('#products')[0].attrib.json_string), bytes)
def test_performance(self):
"""Test parsing and selecting speed"""
import time
large_html = '<html><body>' + '<div class="item">' * 5000 + '</div>' * 5000 + '</body></html>'
start_time = time.time()
parsed = Adaptor(large_html, auto_match=False)
elements = parsed.css('.item')
end_time = time.time()
self.assertEqual(len(elements), 5000)
# Converting 5000 elements to a class and doing operations on them will take time
# Based on my tests with 100 runs, 1 loop each Scrapling (given the extra work/features) takes 10.4ms on average
self.assertLess(end_time - start_time, 0.5) # Locally I test on 0.1 but on GitHub actions with browsers and threading sometimes closing adds fractions of seconds
# JSON attribute conversion
attr_json = page.css_first('#products').attrib['schema'].json()
assert attr_json == {'jsonable': 'data'}
assert isinstance(page.css('#products')[0].attrib.json_string, bytes)
# Use `coverage run -m unittest --verbose tests/test_parser_functions.py` instead for the coverage report
# if __name__ == '__main__':
# unittest.main(verbosity=2)
# Performance Test
def test_large_html_parsing_performance():
"""Test parsing and selecting performance on large HTML"""
large_html = '<html><body>' + '<div class="item">' * 5000 + '</div>' * 5000 + '</body></html>'
start_time = time.time()
parsed = Adaptor(large_html, auto_match=False)
elements = parsed.css('.item')
end_time = time.time()
assert len(elements) == 5000
# Converting 5000 elements to a class and doing operations on them will take time
# Based on my tests with 100 runs, 1 loop each Scrapling (given the extra work/features) takes 10.4ms on average
assert end_time - start_time < 0.5 # Locally I test on 0.1 but on GitHub actions with browsers and threading sometimes closing adds fractions of seconds
# Selector Generation Test
def test_selectors_generation(page):
"""Try to create selectors for all elements in the page"""
def _traverse(element: Adaptor):
assert isinstance(element.generate_css_selector, str)
assert isinstance(element.generate_xpath_selector, str)
for branch in element.children:
_traverse(branch)
_traverse(page)
# Miscellaneous Tests
def test_getting_all_text(page):
"""Test getting all text from the page"""
assert page.get_all_text() != ''
def test_regex_on_text(page):
"""Test regex operations on text"""
element = page.css('[data-id="1"] .price')[0]
match = element.re_first(r'[\.\d]+')
assert match == '10.99'
match = element.text.re(r'(\d+)', replace_entities=False)
assert len(match) == 2