stygian-browser 0.9.2

Anti-detection browser automation library for Rust with CDP stealth features
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
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274

//! Adaptive element similarity search for [`crate::page::PageHandle`].
//!
//! Computes a structural fingerprint for a reference DOM node and finds
//! candidates on the current page that exceed a configurable similarity
//! threshold, even after class names or depth have shifted.

use crate::page::NodeHandle;

// ─── ElementFingerprint ───────────────────────────────────────────────────────

/// A structural snapshot of a DOM element for similarity comparison.
///
/// Serialisable so callers can persist fingerprints across sessions.
///
/// # Example
///
/// ```
/// use stygian_browser::similarity::ElementFingerprint;
///
/// let fp = ElementFingerprint {
///     tag: "div".to_string(),
///     classes: vec!["card".to_string(), "highlighted".to_string()],
///     attr_names: vec!["data-id".to_string()],
///     depth: 3,
/// };
/// let json = serde_json::to_string(&fp).unwrap();
/// let back: ElementFingerprint = serde_json::from_str(&json).unwrap();
/// assert_eq!(fp.tag, back.tag);
/// ```
#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
pub struct ElementFingerprint {
    /// Lower-case tag name (e.g. `"div"`, `"a"`).
    pub tag: String,
    /// Sorted CSS class list (excluding the empty string).
    pub classes: Vec<String>,
    /// Sorted attribute names present on the element, excluding `"class"` and
    /// `"id"`.
    #[serde(rename = "attrNames")]
    pub attr_names: Vec<String>,
    /// Distance from `<body>` in the DOM tree (direct children of `<body>`
    /// have depth `0`).
    pub depth: u32,
}

// ─── SimilarityConfig ─────────────────────────────────────────────────────────

/// Tuning parameters for [`crate::page::PageHandle::find_similar`].
///
/// # Example
///
/// ```
/// use stygian_browser::similarity::SimilarityConfig;
///
/// let cfg = SimilarityConfig { threshold: 0.5, max_results: 5 };
/// assert!(cfg.threshold < SimilarityConfig::DEFAULT_THRESHOLD);
/// ```
#[derive(Debug, Clone)]
pub struct SimilarityConfig {
    /// Minimum score `[0.0, 1.0]` for a candidate to be included in results.
    ///
    /// Default: [`DEFAULT_THRESHOLD`](Self::DEFAULT_THRESHOLD).
    pub threshold: f32,
    /// Maximum number of results to return.  `0` means unlimited.
    ///
    /// Default: `10`.
    pub max_results: usize,
}

impl SimilarityConfig {
    /// Default minimum similarity threshold (`0.7`).
    pub const DEFAULT_THRESHOLD: f32 = 0.7;
}

impl Default for SimilarityConfig {
    fn default() -> Self {
        Self {
            threshold: Self::DEFAULT_THRESHOLD,
            max_results: 10,
        }
    }
}

// ─── SimilarMatch ─────────────────────────────────────────────────────────────

/// A candidate element that exceeded the similarity threshold.
pub struct SimilarMatch {
    /// The matching node handle.
    pub node: NodeHandle,
    /// Similarity score in `[0.0, 1.0]`.
    pub score: f32,
}

// ─── Scoring ──────────────────────────────────────────────────────────────────

/// Compute the weighted Jaccard similarity between two element fingerprints.
///
/// Weights:
/// | Component | Weight |
/// |-----------|--------|
/// | Tag name match | 0.40 |
/// | Class list Jaccard | 0.35 |
/// | Attribute name Jaccard | 0.15 |
/// | Depth proximity | 0.10 |
///
/// Returns a score in `[0.0, 1.0]`.
///
/// # Example
///
/// ```
/// use stygian_browser::similarity::{ElementFingerprint, jaccard_weighted};
///
/// let a = ElementFingerprint { tag: "div".into(), classes: vec!["foo".into()], attr_names: vec![], depth: 2 };
/// let b = a.clone();
/// assert!((jaccard_weighted(&a, &b) - 1.0).abs() < 1e-6);
/// ```
pub fn jaccard_weighted(reference: &ElementFingerprint, candidate: &ElementFingerprint) -> f32 {
    let tag_score = if reference.tag == candidate.tag {
        1.0_f32
    } else {
        0.0_f32
    };

    let class_score = jaccard_sets(&reference.classes, &candidate.classes);
    let attr_score = jaccard_sets(&reference.attr_names, &candidate.attr_names);

    // DOM tree depth is always small (< 1000 in practice); truncate to u16
    // for a lossless f32 conversion (u16 fits exactly in f32's 23-bit mantissa).
    let ref_depth = f32::from(u16::try_from(reference.depth).unwrap_or(u16::MAX));
    let cand_depth = f32::from(u16::try_from(candidate.depth).unwrap_or(u16::MAX));
    let depth_diff = (ref_depth - cand_depth).abs();
    let depth_score = 1.0_f32 / (1.0_f32 + depth_diff);

    depth_score.mul_add(
        0.1_f32,
        attr_score.mul_add(0.15_f32, tag_score.mul_add(0.4_f32, class_score * 0.35_f32)),
    )
}

/// Compute Jaccard similarity (|intersection| / |union|) for two **sorted**
/// string slices.
///
/// Returns `1.0` when both slices are empty (they are identical).
fn jaccard_sets(a: &[String], b: &[String]) -> f32 {
    if a.is_empty() && b.is_empty() {
        return 1.0_f32;
    }
    let mut intersection: usize = 0;
    let mut i = 0_usize;
    let mut j = 0_usize;
    while i < a.len() && j < b.len() {
        let (Some(ai), Some(bj)) = (a.get(i), b.get(j)) else {
            break;
        };
        match ai.cmp(bj) {
            std::cmp::Ordering::Equal => {
                intersection += 1;
                i += 1;
                j += 1;
            }
            std::cmp::Ordering::Less => i += 1,
            std::cmp::Ordering::Greater => j += 1,
        }
    }
    let union = a.len() + b.len() - intersection;
    // union > 0 because at least one slice is non-empty (guarded above)
    // Class/attribute counts are tiny (< 100); u16 fits losslessly in f32.
    let i_f = f32::from(u16::try_from(intersection).unwrap_or(u16::MAX));
    let u_f = f32::from(u16::try_from(union).unwrap_or(u16::MAX));
    i_f / u_f
}

// ─── Tests ────────────────────────────────────────────────────────────────────

#[cfg(test)]
#[allow(clippy::expect_used)] // serde failures in deterministic tests are programmer errors
mod tests {
    use super::*;

    fn fp(tag: &str, classes: &[&str], attrs: &[&str], depth: u32) -> ElementFingerprint {
        ElementFingerprint {
            tag: tag.to_string(),
            classes: classes.iter().map(|s| (*s).to_string()).collect(),
            attr_names: attrs.iter().map(|s| (*s).to_string()).collect(),
            depth,
        }
    }

    #[test]
    fn jaccard_identical() {
        let a = fp("div", &["card", "highlighted"], &["data-id"], 3);
        let b = a.clone();
        let score = jaccard_weighted(&a, &b);
        assert!(
            (score - 1.0_f32).abs() < 1e-5_f32,
            "identical fingerprints should score 1.0, got {score}"
        );
    }

    #[test]
    fn jaccard_disjoint() {
        let a = fp("div", &["foo", "bar"], &["data-x"], 0);
        let b = fp("span", &["baz", "qux"], &["data-y"], 20);
        let score = jaccard_weighted(&a, &b);
        // tag=0, classes=0, attrs=0; depth_prox = 1/(1+20) ≈ 0.0476; weight=0.1
        // expected ≈ 0.00476
        assert!(
            score < 0.05_f32,
            "disjoint fingerprints should score near 0, got {score}"
        );
        assert!(score >= 0.0_f32, "score must be non-negative, got {score}");
    }

    #[test]
    fn jaccard_partial() {
        // Same tag, half classes in common, no attrs, same depth
        let a = fp("div", &["a", "b"], &[], 2);
        let b = fp("div", &["a", "c"], &[], 2);
        let score = jaccard_weighted(&a, &b);
        // tag=1*0.4=0.4, classes=jaccard({a,b},{a,c})=1/3≈0.333*0.35≈0.1167
        // attrs=both empty→1.0*0.15=0.15, depth=1/(1+0)=1.0*0.1=0.1
        // total ≈ 0.4 + 0.1167 + 0.15 + 0.1 = 0.7667
        assert!(
            score > 0.5_f32,
            "partial-match fingerprint should score > 0.5, got {score}"
        );
        assert!(
            score < 0.9_f32,
            "partial-match fingerprint should score < 0.9, got {score}"
        );
    }

    #[test]
    fn similarity_config_default_threshold() {
        assert!(
            (SimilarityConfig::DEFAULT_THRESHOLD - 0.7_f32).abs() < f32::EPSILON,
            "DEFAULT_THRESHOLD should be 0.7"
        );
        let cfg = SimilarityConfig::default();
        assert!(
            (cfg.threshold - SimilarityConfig::DEFAULT_THRESHOLD).abs() < f32::EPSILON,
            "default threshold should equal DEFAULT_THRESHOLD"
        );
        assert_eq!(cfg.max_results, 10);
    }

    #[test]
    fn fingerprint_serde_roundtrip() {
        let original = ElementFingerprint {
            tag: "section".to_string(),
            classes: vec!["main".to_string(), "wrapper".to_string()],
            attr_names: vec!["aria-label".to_string(), "data-section".to_string()],
            depth: 5,
        };
        let json = serde_json::to_string(&original).expect("serialize");
        let decoded: ElementFingerprint = serde_json::from_str(&json).expect("deserialize");
        assert_eq!(original, decoded);
    }

    #[test]
    fn fingerprint_serde_attr_names_key() {
        // The JSON key must be "attrNames" (camelCase) to match the JS output.
        let fp_val = ElementFingerprint {
            tag: "a".to_string(),
            classes: vec![],
            attr_names: vec!["href".to_string()],
            depth: 1,
        };
        let json = serde_json::to_string(&fp_val).expect("serialize");
        assert!(
            json.contains("\"attrNames\""),
            "JSON key must be 'attrNames', got: {json}"
        );
    }
}