skyscraper 0.7.0

XPath for HTML web scraping
Documentation 
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
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221

use skyscraper::html;

/// A comment token encountered in the "after head" insertion mode should be
/// inserted as a comment node (WHATWG 13.2.6.4.6).
#[test]
fn comment_in_after_head_mode_is_inserted() {
    // After </head> we're in AfterHead. A comment here should be inserted.
    let text = "<html><head></head><!-- after head comment --><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<!-- after head comment -->"),
        "Comment should be preserved: {output:?}"
    );
}

/// A DOCTYPE token encountered in the "after head" insertion mode should be
/// treated as a parse error and ignored (WHATWG 13.2.6.4.6).
#[test]
fn doctype_in_after_head_mode_is_ignored() {
    // The first DOCTYPE is consumed by Initial mode. After </head>, the parser
    // is in AfterHead mode. A second DOCTYPE here should be silently dropped.
    let text = "<!DOCTYPE html><html><head></head><!DOCTYPE html><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert_eq!(
        output.matches("<!DOCTYPE").count(),
        1,
        "Only one DOCTYPE should survive; got: {output:?}"
    );
}

/// An <html> start tag in "after head" should be processed using InBody rules,
/// which merges new attributes onto the existing <html> element (WHATWG 13.2.6.4.6).
#[test]
fn html_start_tag_in_after_head_merges_attributes() {
    let text = r#"<html><head></head><html lang="en"><body></body></html>"#;
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    // Only one <html> element should exist.
    assert_eq!(
        output.matches("<html").count(),
        1,
        "Only one <html> should exist; got: {output:?}"
    );
    // The merged attribute should be present.
    assert!(
        output.contains(r#"lang="en""#),
        "Should contain lang attribute: {output:?}"
    );
}

/// A duplicate <head> start tag in "after head" mode should be treated as a
/// parse error and ignored (WHATWG 13.2.6.4.6).
#[test]
fn duplicate_head_start_tag_in_after_head_is_ignored() {
    let text = "<html><head></head><head><title>oops</title></head><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    // Only one <head> element should exist.
    assert_eq!(
        output.matches("<head").count(),
        1,
        "Only one <head> should exist; got: {output:?}"
    );
}

/// An unexpected end tag (not body, html, or br) in "after head" mode should
/// be treated as a parse error and ignored (WHATWG 13.2.6.4.6).
#[test]
fn unexpected_end_tag_in_after_head_is_ignored() {
    let text = "<html><head></head></div><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    // The </div> should be ignored, no extra elements created.
    assert!(
        output.contains("<body></body>"),
        "Should still have <body>: {output:?}"
    );
}

/// End tags for body, html, or br in "after head" mode should trigger the
/// "anything else" path — inserting a <body> element and reprocessing
/// (WHATWG 13.2.6.4.6).
#[test]
fn body_end_tag_in_after_head_triggers_anything_else() {
    // </body> in after-head should create a <body> and reprocess.
    let text = "<html><head></head></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<body>"),
        "A <body> should be implicitly created: {output:?}"
    );
}

/// Head-level elements appearing after </head> should be processed using
/// "in head" rules with the head element pointer pushed/popped on the
/// stack of open elements (WHATWG 13.2.6.4.6).
///
/// A <meta> tag after </head> should end up inside the <head> element.
#[test]
fn meta_after_head_is_inserted_in_head() {
    let text = r#"<html><head></head><meta charset="utf-8"><body></body></html>"#;
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    // The <meta> should be inside <head>, not floating outside.
    assert!(
        output.contains("<head><meta"),
        "Meta should be inside head: {output:?}"
    );
}

/// A <title> tag appearing after </head> should be inserted into <head>
/// via the head element pointer mechanism (WHATWG 13.2.6.4.6).
#[test]
fn title_after_head_is_inserted_in_head() {
    let text = "<html><head></head><title>Late Title</title><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<head><title>"),
        "Title should be inside head: {output:?}"
    );
    assert!(
        output.contains("Late Title"),
        "Title content should be preserved: {output:?}"
    );
}

/// A <link> tag appearing after </head> should be inserted into <head>
/// via the head element pointer mechanism (WHATWG 13.2.6.4.6).
#[test]
fn link_after_head_is_inserted_in_head() {
    let text = r#"<html><head></head><link rel="stylesheet" href="a.css"><body></body></html>"#;
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<head><link"),
        "Link should be inside head: {output:?}"
    );
}

/// A <style> tag appearing after </head> should be inserted into <head>
/// via the head element pointer mechanism (WHATWG 13.2.6.4.6).
#[test]
fn style_after_head_is_inserted_in_head() {
    let text = "<html><head></head><style>body { color: red; }</style><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<head><style>"),
        "Style should be inside head: {output:?}"
    );
}

/// A <base> tag appearing after </head> should be inserted into <head>
/// via the head element pointer mechanism (WHATWG 13.2.6.4.6).
#[test]
fn base_after_head_is_inserted_in_head() {
    let text = r#"<html><head></head><base href="/"><body></body></html>"#;
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<head><base"),
        "Base should be inside head: {output:?}"
    );
}

/// A <script> tag appearing after </head> should be inserted into <head>
/// via the head element pointer mechanism (WHATWG 13.2.6.4.6).
/// This exercises the ScriptData tokenizer state path through Text mode.
#[test]
fn script_after_head_is_inserted_in_head() {
    let text = "<html><head></head><script>var x = 1;</script><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<head><script>"),
        "Script should be inside head: {output:?}"
    );
    assert!(
        output.contains("var x = 1;"),
        "Script content should be preserved: {output:?}"
    );
}

/// A <noframes> tag appearing after </head> should be inserted into <head>
/// via the head element pointer mechanism (WHATWG 13.2.6.4.6).
/// This exercises the generic raw text element parsing algorithm path.
#[test]
fn noframes_after_head_is_inserted_in_head() {
    let text = "<html><head></head><noframes>No frames</noframes><body></body></html>";
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<head><noframes>"),
        "Noframes should be inside head: {output:?}"
    );
}

/// Multiple head-level elements after </head> should all be placed inside
/// <head> via the head element pointer mechanism (WHATWG 13.2.6.4.6).
#[test]
fn multiple_head_elements_after_head_all_go_in_head() {
    let text = r#"<html><head></head><meta charset="utf-8"><title>T</title><link rel="icon" href="f"><body></body></html>"#;
    let document = html::parse(text).unwrap();
    let output = document.to_string();
    assert!(
        output.contains("<head><meta"),
        "Meta should be inside head: {output:?}"
    );
    assert!(
        output.contains("<title>"),
        "Title should be inside head: {output:?}"
    );
    assert!(
        output.contains("<link"),
        "Link should be inside head: {output:?}"
    );
}