trafilatura 0.2.0

Extract readable content, comments, and metadata from web pages
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
222
223
224
225
226
227
228
229
230
231

// Port of go-trafilatura/internal/selector/content.go

use crate::dom::{Document, NodeId};
use crate::selector::utils::{contains, lower, starts_with};

/// Ordered list of content extraction rules.
/// Applied in sequence; the first match wins.
pub(crate) const CONTENT: &[super::Rule] = &[
    content_rule1,
    content_rule2,
    content_rule3,
    content_rule4,
    content_rule5,
];

/// Matches common article/post content containers by class/id patterns.
///
/// Port of `contentRule1`.
fn content_rule1(doc: &Document, id: NodeId) -> bool {
    let tag = doc.tag_name(id);
    match tag {
        "article" | "div" | "main" | "section" => {}
        _ => return false,
    }

    let class = doc.class_name(id);
    let elem_id = doc.id_attr(id);
    let item_prop = doc.get_attribute(id, "itemprop").unwrap_or_default();

    matches!(class.as_str(), "post" | "entry")
        || contains(&class, "post-text")
        || contains(&class, "post_text")
        || contains(&class, "post-body")
        || contains(&class, "post-entry")
        || contains(&class, "postentry")
        || contains(&class, "post-content")
        || contains(&class, "post_content")
        || contains(&lower(&class), "postcontent")
        || contains(&class, "post_inner_wrapper")
        || contains(&class, "article-text")
        || contains(&lower(&class), "articletext")
        || contains(&elem_id, "entry-content")
        || contains(&class, "entry-content")
        || contains(&elem_id, "article-content")
        || contains(&class, "article-content")
        || contains(&elem_id, "article__content")
        || contains(&class, "article__content")
        || contains(&elem_id, "article-body")
        || contains(&class, "article-body")
        || contains(&elem_id, "article__body")
        || contains(&class, "article__body")
        || item_prop == "articleBody"
        || contains(&lower(&elem_id), "articlebody")
        || contains(&lower(&class), "articlebody")
        || elem_id == "articleContent"
        || contains(&class, "ArticleContent")
        || contains(&class, "page-content")
        || contains(&class, "text-content")
        || contains(&elem_id, "body-text")
        || contains(&class, "body-text")
        || contains(&class, "article__container")
        || contains(&elem_id, "art-content")
        || contains(&class, "art-content")
}

/// Matches any `<article>` element.
///
/// Port of `contentRule2`.
fn content_rule2(doc: &Document, id: NodeId) -> bool {
    doc.tag_name(id) == "article"
}

/// Matches content containers by story/post/blog class patterns, plus `role="article"`.
///
/// Port of `contentRule3`.
fn content_rule3(doc: &Document, id: NodeId) -> bool {
    let tag = doc.tag_name(id);
    match tag {
        "article" | "div" | "main" | "section" => {}
        _ => return false,
    }

    let class = doc.class_name(id);
    let elem_id = doc.id_attr(id);
    let role = doc.get_attribute(id, "role").unwrap_or_default();

    contains(&class, "post-bodycopy")
        || contains(&class, "storycontent")
        || contains(&class, "story-content")
        || class == "postarea"
        || class == "art-postcontent"
        || contains(&class, "theme-content")
        || contains(&class, "blog-content")
        || contains(&class, "section-content")
        || contains(&class, "single-content")
        || contains(&class, "single-post")
        || contains(&class, "main-column")
        || contains(&class, "wpb_text_column")
        || starts_with(&elem_id, "primary")
        || starts_with(&class, "article")
        || class == "text"
        || elem_id == "article"
        || class == "cell"
        || elem_id == "story"
        || class == "story"
        || contains(&class, "story-body")
        || contains(&elem_id, "story-body")
        || contains(&class, "field-body")
        || contains(&lower(&class), "fulltext")
        || role == "article"
}

/// Matches content containers by content-main/content-body/main-content patterns.
///
/// Port of `contentRule4`.
fn content_rule4(doc: &Document, id: NodeId) -> bool {
    let tag = doc.tag_name(id);
    match tag {
        "article" | "div" | "main" | "section" => {}
        _ => return false,
    }

    let class = doc.class_name(id);
    let elem_id = doc.id_attr(id);

    contains(&elem_id, "content-main")
        || contains(&class, "content-main")
        || contains(&class, "content_main")
        || contains(&elem_id, "content-body")
        || contains(&class, "content-body")
        || contains(&elem_id, "contentBody")
        || contains(&class, "content__body")
        || contains(&lower(&elem_id), "main-content")
        || contains(&lower(&class), "main-content")
        || contains(&lower(&class), "page-content")
        || elem_id == "content"
        || class == "content"
}

/// Matches containers starting with "main" in class/id/role, plus `<main>` itself.
///
/// Port of `contentRule5`.
fn content_rule5(doc: &Document, id: NodeId) -> bool {
    let tag = doc.tag_name(id);
    match tag {
        "article" | "div" | "section" => {}
        "main" => return true,
        _ => return false,
    }

    let class = doc.class_name(id);
    let elem_id = doc.id_attr(id);
    let role = doc.get_attribute(id, "role").unwrap_or_default();

    starts_with(&class, "main") || starts_with(&elem_id, "main") || starts_with(&role, "main")
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::dom::Document;
    use crate::selector::{query, query_all};

    fn parse(html: &str) -> Document {
        Document::parse(html)
    }

    #[test]
    fn test_content_rule1_post_class() {
        let doc = parse(r#"<html><body><div class="post"><p>text</p></div></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_some());
    }

    #[test]
    fn test_content_rule1_article_body() {
        let doc = parse(r#"<html><body><div class="article-body">text</div></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_some());
    }

    #[test]
    fn test_content_rule2_article_tag() {
        let doc = parse(r#"<html><body><article>text</article></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_some());
    }

    #[test]
    fn test_content_rule3_story_content() {
        let doc = parse(r#"<html><body><div class="story-content">text</div></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_some());
    }

    #[test]
    fn test_content_rule4_content_main() {
        let doc = parse(r#"<html><body><div id="content-main">text</div></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_some());
    }

    #[test]
    fn test_content_rule5_main_tag() {
        let doc = parse(r#"<html><body><main>text</main></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_some());
    }

    #[test]
    fn test_content_rule5_main_class() {
        let doc = parse(r#"<html><body><div class="main-container">text</div></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_some());
    }

    #[test]
    fn test_content_no_match() {
        let doc = parse(r#"<html><body><div class="footer">text</div></body></html>"#);
        let body = doc.body().unwrap();
        assert!(query(&doc, body, CONTENT).is_none());
    }

    #[test]
    fn test_query_all_returns_multiple() {
        let doc =
            parse(r#"<html><body><article>one</article><article>two</article></body></html>"#);
        let body = doc.body().unwrap();
        assert_eq!(query_all(&doc, body, CONTENT).len(), 2);
    }
}