halldyll-core 0.1.0

Core scraping engine for Halldyll - high-performance async web scraper for AI agents
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

//! Audios - Audio extraction

use scraper::{Html, Selector};
use url::Url;

use crate::types::assets::{AudioAsset, AudioType};

/// Audio extractor
pub struct AudioExtractor;

impl Default for AudioExtractor {
    fn default() -> Self {
        Self
    }
}

impl AudioExtractor {
    /// New extractor
    pub fn new() -> Self {
        Self
    }

    /// Extract all audios
    pub fn extract(&self, html: &str, base_url: &Url) -> Vec<AudioAsset> {
        let document = Html::parse_document(html);
        let mut audios = Vec::new();

        // <audio><source> tags
        audios.extend(self.extract_audio_tags(&document, base_url));

        // Deduplicate
        audios.dedup_by(|a, b| a.url == b.url);

        audios
    }

    /// Extract <audio> tags
    fn extract_audio_tags(&self, document: &Html, base_url: &Url) -> Vec<AudioAsset> {
        let audio_selector = Selector::parse("audio").unwrap();
        let source_selector = Selector::parse("source").unwrap();
        let mut audios = Vec::new();

        for audio in document.select(&audio_selector) {
            let attrs = audio.value();

            // Direct src attribute
            if let Some(src) = attrs.attr("src") {
                if let Ok(url) = base_url.join(src) {
                    let audio_type = self.detect_audio_type(&url);
                    audios.push(AudioAsset {
                        url,
                        audio_type,
                        duration: None,
                        title: None,
                    });
                }
            }

            // Child sources
            for source in audio.select(&source_selector) {
                if let Some(src) = source.value().attr("src") {
                    if let Ok(url) = base_url.join(src) {
                        let audio_type = source.value().attr("type")
                            .map(|t| self.audio_type_from_mime(t))
                            .unwrap_or_else(|| self.detect_audio_type(&url));
                        
                        audios.push(AudioAsset {
                            url,
                            audio_type,
                            duration: None,
                            title: None,
                        });
                    }
                }
            }
        }

        audios
    }

    /// Detect audio type from URL
    fn detect_audio_type(&self, url: &Url) -> AudioType {
        let path = url.path().to_lowercase();
        
        if path.ends_with(".mp3") {
            AudioType::Mp3
        } else if path.ends_with(".wav") {
            AudioType::Wav
        } else if path.ends_with(".ogg") || path.ends_with(".oga") {
            AudioType::Ogg
        } else if path.ends_with(".aac") || path.ends_with(".m4a") {
            AudioType::Aac
        } else if path.ends_with(".flac") {
            AudioType::Flac
        } else {
            AudioType::Unknown
        }
    }

    /// Type d'audio depuis le MIME type
    fn audio_type_from_mime(&self, mime: &str) -> AudioType {
        let mime = mime.to_lowercase();
        
        if mime.contains("mp3") || mime.contains("mpeg") {
            AudioType::Mp3
        } else if mime.contains("wav") {
            AudioType::Wav
        } else if mime.contains("ogg") {
            AudioType::Ogg
        } else if mime.contains("aac") || mime.contains("mp4") {
            AudioType::Aac
        } else if mime.contains("flac") {
            AudioType::Flac
        } else {
            AudioType::Unknown
        }
    }
}

/// Extract podcast RSS links from a page
pub fn extract_podcast_links(html: &str, base_url: &Url) -> Vec<Url> {
    let document = Html::parse_document(html);
    let selector = Selector::parse(r#"link[type="application/rss+xml"], link[type="application/atom+xml"]"#).unwrap();
    
    document
        .select(&selector)
        .filter_map(|el| {
            let href = el.value().attr("href")?;
            let title = el.value().attr("title").unwrap_or("");
            
            // Check if it's probably a podcast
            if title.to_lowercase().contains("podcast") 
                || title.to_lowercase().contains("audio")
                || href.to_lowercase().contains("podcast")
            {
                base_url.join(href).ok()
            } else {
                None
            }
        })
        .collect()
}