1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
|
"""
Encapsulation of html-related functionality.
BeautifulSoup should only get used here.
"""
import html
import logging
import re
from abc import ABC
from dataclasses import dataclass
from functools import cached_property
from bs4 import BeautifulSoup
from bs4 import NavigableString
@dataclass
class HTMLMatch(ABC):
node: "Node" = None
@dataclass
class HTMLExactTextMatch(HTMLMatch):
pass
@dataclass
class HTMLAttributeMatch(HTMLMatch):
attr: str = None
class Node:
soup = None
_page = None
def __init__(self, soup, page: "Page"):
self.soup = soup
self._hash = soup.__hash__()
self._page = page
@property
def root(self):
return self._page
@cached_property
def depth(self):
return self.parent.depth
@cached_property
def parent(self):
return self._page._get_node_for_soup(self.soup.parent)
@cached_property
def text(self):
return self.soup.text
def find_all(self, item) -> list[HTMLMatch]:
return list(self._generate_find_all(item))
def _generate_find_all(self, item):
assert isinstance(item, str), "can only search for str at the moment"
# text
# - since text matches including whitespace, a regex is used
for soup_node in self.soup.find_all(
string=re.compile(r"^\s*%s\s*$" % html.escape(item))
):
# use parent node as found text is NaviableString and not Tag
node = self._page._get_node_for_soup(soup_node.parent)
yield HTMLExactTextMatch(node)
# attributes
for soup_node in self.soup.find_all():
for attr in soup_node.attrs:
if soup_node[attr] == item:
node = self._page._get_node_for_soup(soup_node)
yield HTMLAttributeMatch(node, attr)
# todo implement other find methods
def has_parent(self, node: "Node"):
for p in self.soup.parents:
if p == node.soup:
return True
return False
@cached_property
def parents(self):
return [self._page._get_node_for_soup(p) for p in self.soup.parents]
@property
def classes(self):
return self.soup.attrs.get("class", [])
@property
def id(self):
return self.soup.attrs.get("id", None)
@property
def tag_name(self):
return self.soup.name
@property
def html_attributes(self):
return self.soup.attrs
def select(self, css_selector):
return [
self._page._get_node_for_soup(n) for n in self.soup.select(css_selector)
]
def __repr__(self):
if isinstance(self.soup, NavigableString):
return f"<{self.__class__.__name__} {self.soup.strip()[:10]=}>"
return (
f"<{self.__class__.__name__} {self.soup.name=}"
f" classes={self.soup.get('class', None)},"
f" text={''.join(self.soup.stripped_strings)[:10]}...>"
)
def __hash__(self):
return self._hash
# return self.soup.__hash__()
# return super().__hash__()
def __eq__(self, other):
return isinstance(other, Node) and self.soup == other.soup
class Page(Node):
"""
One page, i.e. one HTML document.
"""
_node_registry = None
def __init__(self, html):
self.html = html
soup = BeautifulSoup(self.html, "lxml")
# register node for each soup
self._node_registry = {soup: self}
super().__init__(soup, self)
@property
def depth(self):
return 0
def _get_node_for_soup(self, soup) -> Node:
if soup not in self._node_registry:
self._node_registry[soup] = Node(soup, self)
return self._node_registry[soup]
def get_root_node(nodes: list[Node]) -> Node:
pages = [n._page for n in nodes]
assert len(set(pages)) == 1, "different pages found, cannot get a root"
# generate parent paths from top to bottom
# [elem, parent, ancestor, root]
parent_paths = [reversed([n] + n.parents) for n in nodes]
# start looping from bottom to top
# zip automatically uses common length
# -> last element is the first one, where len(nodes) roots to compare exist
for layer_nodes in reversed(list(zip(*parent_paths))):
if len(set(layer_nodes)) == 1:
return layer_nodes[0]
raise RuntimeError("no root found")
def get_relative_depth(node: Node, root: Node):
node_parents = list(node.soup.parents)
# depth of root
i = node_parents.index(root.soup)
# depth of element
j = len(node_parents)
return j - i
def selector_matches_nodes(root: Node, selector: str, expected: list[Node]):
"""
Check whether the given selector matches the expected nodes.
"""
logging.info(
f"checking if selector matches nodes ({root=}, {selector=}, {expected=})"
)
# we care for equality here
# as selector should match the expected nodes in the exact given order
# we do this here, as wrapping Nodes can have side effects regarding equality
return root.soup.select(selector) == [n.soup for n in expected]
|
