def test_html_element_refinement():
new = HtmlElement('span', display=Display.inline, prefix=' ', suffix=' ',
limit_whitespace_affixes=True)
pre = HtmlElement('pre', display=Display.block, whitespace=WhiteSpace.pre)
code = HtmlElement('code')
# refinement with pre and whitespaces
refined = pre.get_refined_html_element(copy(new))
assert refined.prefix == ''
assert refined.suffix == ''
# refinement with code and whitespaces
refined = code.get_refined_html_element(copy(new))
assert refined.prefix == ' '
assert refined.suffix == ' '
# refinement with pre and non-whitespaces
new.prefix = ' 1. '
new.suffix = '<'
refined = pre.get_refined_html_element(copy(new))
assert refined.prefix == ' 1. '
assert refined.suffix == '<'
# refinement with code and non-whitespaces
refined = code.get_refined_html_element(copy(new))
assert refined.prefix == ' 1. '
assert refined.suffix == '<'
def test_formatting():
# standard line
h = HtmlElement()
assert _get_text(h) == 'firstEhre sei Gott!last'
h.display = Display.block
h.margin_before = 1
h.margin_after = 2
print(h)
print(_get_text(h))
assert _get_text(h) == 'first\n\nEhre sei Gott!\n\n\nlast'
# list bullet without padding_inline
h.list_bullet = "* "
assert _get_text(h) == 'first\n\n* Ehre sei Gott!\n\n\nlast'
# add a padding_inline
h.padding_inline = 3
assert _get_text(h) == 'first\n\n * Ehre sei Gott!\n\n\nlast'
# and prefixes + suffixes
h.prefix = '>>'
h.suffix = '<<'
assert _get_text(h)== 'first\n\n * >>Ehre sei Gott!<<\n\n\nlast'
def _get_text(html_element):
"""
Returns
the text formatted based on the current HTML element.
"""
c = Canvas()
html_element.canvas = c
HtmlElement().set_canvas(c).write("first")
c.open_tag(html_element)
html_element.write("Ehre sei Gott!")
c.close_tag(html_element)
HtmlElement().set_canvas(c).write("last")
c._flush_inline()
return '\n'.join(c.blocks)
def test_parse_horizontal_align():
html_element = HtmlElement()
CssParse.attr_horizontal_align('center', html_element)
assert html_element.align == HorizontalAlignment.center
# invalid value
CssParse.attr_horizontal_align('unknown', html_element)
assert html_element.align == HorizontalAlignment.center
def test_parse_vertical_align():
html_element = HtmlElement()
CssParse.attr_vertical_align('top', html_element)
assert html_element.valign == VerticalAlignment.top
# invalid value
CssParse.attr_vertical_align('unknown', html_element)
assert html_element.valign == VerticalAlignment.top
def test_html_element_str():
'''
Tests the string representation of an HtmlElement.
'''
html_element = HtmlElement('div', '', '', Display.inline, 0, 0, 0, '',
WhiteSpace.pre)
assert str(html_element) == ('<div prefix=, suffix=, '
'display=Display.inline, margin_before=0, '
'margin_after=0, padding_inline=0, '
'list_bullet=, '
'whitespace=WhiteSpace.pre, '
'align=HorizontalAlignment.left, '
'valign=VerticalAlignment.middle, '
'annotation=()>')
def test_parse():
"""
basic rule parsing.
"""
rules = {'table#border=1': ['table'], 'hr': ['horizontal-line']}
tags, attrs = AnnotationModel._parse(rules)
assert tags == {'hr': ['horizontal-line']}
apply_annotation = attrs[0]
assert apply_annotation.match_tag == 'table'
assert apply_annotation.match_value == '1'
assert apply_annotation.attr == 'border'
e = HtmlElement(tag='table')
apply_annotation.apply('1', e)
assert e.annotation == ('table', )
def test_merged_attribute():
"""
test multiple rules per attribute
"""
rules = {'#color=white': ['white'], '#color=yellow': ['yellow']}
css = deepcopy(CSS_PROFILES['strict'])
annotation_model = AnnotationModel(css, rules)
attribute_handler = Attribute()
attribute_handler.merge_attribute_map(annotation_model.css_attr)
e = HtmlElement()
attribute_handler.attribute_mapping['color']('green', e)
assert e.annotation == ()
attribute_handler.attribute_mapping['color']('yellow', e)
assert e.annotation == ('yellow', )
attribute_handler.attribute_mapping['color']('white', e)
assert e.annotation == ('yellow', 'white')
#!/usr/bin/env python3
# coding: utf-8
"""Standard CSS profiles shipped with inscriptis.
- `strict`: this profile corresponds to the defaults used by Firefox
- `relaxed`: this profile is more suited for text analytics, since it ensures
that whitespaces are inserted between span and div elements
preventing cases where two words stick together.
"""
from inscriptis.model.html_element import HtmlElement
from inscriptis.html_properties import Display, WhiteSpace
STRICT_CSS_PROFILE = {
'body':
HtmlElement(display=Display.inline, whitespace=WhiteSpace.normal),
'head':
HtmlElement(display=Display.none),
'link':
HtmlElement(display=Display.none),
'meta':
HtmlElement(display=Display.none),
'script':
HtmlElement(display=Display.none),
'title':
HtmlElement(display=Display.none),
'style':
HtmlElement(display=Display.none),
'p':
HtmlElement(display=Display.block, margin_before=1, margin_after=1),
'figure':
def test_style_unit_parsing():
html_element = HtmlElement()
CssParse.attr_style(
"margin-top:2.666666667em;margin-bottom: 2.666666667em", html_element)
assert html_element.margin_before == 3
assert html_element.margin_after == 3