harser
harser copied to clipboard
Published
20 hours ago •
sihaelov
sihaelov
Easy way for HTML parsing and building XPath
Harser
Harser is a library for easy extracting data from HTML and building XPath.
Installation
pip install harser
Examples
>>> from harser import Harser
>>> HTML = '''
<html><body>
<div class="header" id="id-header">
<li class="nav-item" data-nav="first-item" href="/nav1">First item</li>
<li class="nav-item" data-nav="second-item" href="/nav2">Second item</li>
<li class="nav-item" data-nav="third-item" href="/nav3">Third item</li>
</div>
<div>First layer
<h3>Lorem Ipsum</h3>
<span>Dolor sit amet</span>
</div>
<div>Second layer</div>
<div>Third layer
<span class="text">first block</span>
<span class="text">second block</span>
<span>third block</span>
</div>
<span>fourth layer</span>
<img />
<div class="footer" id="id-foobar" foobar="ab bc cde">
<h3 some-attr="hey">
<span id="foobar-span">foo ter</span>
</h3>
</div>
</body></html>
'''
>>> harser = Harser(HTML)
>>> harser.find('div', class_='header').children(class_='nav-item').find('text').extract()
# Or just
# harser.find(class_='nav-item').find('text').extract()
['First item', 'Second item', 'Third item']
>>> harser.find(class_='nav-item').get_attr('href').extract()
['/nav1', '/nav2', '/nav3']
# It is equally
>>> harser.find('div', class_='header', id='id-header')
>>> harser.find('div', attrs={'class': 'header', 'id': 'id-header'})
>>> harser.find(id__contains='bar').get_attr('class').extract()
['footer']
>>> harser.find(href__not_contains='2').find('text').extract()
['First item', 'Third item']
>>> harser.find(attrs={'data-nav__contains': 'second'}).next_siblings().find('text').extract()
['Third item']
>>> harser.find('li').parent().next_siblings(filters={'text__contains': 'Second'}).clean_extract()
['<div>Second layer</div>']
>>> harser.find('h3', filters={'span.@id__starts_with': 'foo'}).get_attr('some-attr').extract()
['hey']
>>> harser.find('div').children('h3').xpath
'//descendant::div/h3'
Support the project
Please contact Michael Sinov if you want to support the Harser project.
sihaelov