ebook-rs 0.16.4

use crate::book::Book;
use serde::{Deserialize, Serialize};

/// Stable fingerprint representation of an eBook for content deduplication.
#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
pub struct BookFingerprint {
    /// SHA-256 hash of normalized text across all spine sections (metadata independent).
    pub content_hash: String,
    /// SHA-256 hash of core metadata (title + creator + identifier).
    pub metadata_hash: String,
    /// Total number of spine reading sections.
    pub total_sections: usize,
    /// Total character count across all text sections.
    pub total_characters: usize,
}

impl BookFingerprint {
    /// Compute similarity match score (0.0 to 1.0) against another `BookFingerprint`.
    pub fn match_score(&self, other: &BookFingerprint) -> f64 {
        if self.content_hash == other.content_hash {
            return 1.0;
        }

        if self.total_sections == other.total_sections && self.total_characters > 0 {
            let char_diff = (self.total_characters as f64 - other.total_characters as f64).abs();
            let len_similarity =
                1.0 - (char_diff / (self.total_characters.max(other.total_characters) as f64));

            if self.metadata_hash == other.metadata_hash {
                return (0.6 + 0.4 * len_similarity).clamp(0.0, 1.0);
            } else {
                return (0.8 * len_similarity).clamp(0.0, 1.0);
            }
        }

        if self.metadata_hash == other.metadata_hash {
            return 0.5;
        }

        0.0
    }

    /// Check if two book fingerprints represent the same book content.
    pub fn is_duplicate_of(&self, other: &BookFingerprint) -> bool {
        self.match_score(other) >= 0.85
    }
}

/// Book Fingerprint generator engine.
pub struct FingerprintGenerator;

impl FingerprintGenerator {
    /// Generate stable `BookFingerprint` for a `Book` instance.
    pub fn generate(book: &Book) -> BookFingerprint {
        let meta = book.metadata();

        // 1. Compute Metadata Hash
        let id_str = meta.identifier.as_deref().unwrap_or("");
        let meta_raw = format!(
            "{}:{}:{}",
            meta.title.trim().to_lowercase(),
            meta.creator().trim().to_lowercase(),
            id_str.trim().to_lowercase()
        );
        let metadata_hash = hex_sha256(meta_raw.as_bytes());

        // 2. Compute Content Hash over all section plain texts
        let mut text_buf = String::new();
        let mut total_characters = 0;

        let hydrated = book.get_all_sections_hydrated();
        for section in &hydrated {
            let norm_text: String = section
                .plain_text
                .chars()
                .filter(|c| !c.is_whitespace())
                .collect();
            total_characters += section.char_count;
            text_buf.push_str(&norm_text);
        }

        let content_hash = hex_sha256(text_buf.as_bytes());

        BookFingerprint {
            content_hash,
            metadata_hash,
            total_sections: book.sections.len(),
            total_characters,
        }
    }
}

const K256: [u32; 64] = [
    0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5, 0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,
    0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3, 0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,
    0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc, 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,
    0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7, 0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,
    0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13, 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
    0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3, 0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,
    0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5, 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
    0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208, 0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2,
];

pub fn sha256_bytes(input: &[u8]) -> [u8; 32] {
    let mut h: [u32; 8] = [
        0x6a09e667, 0xbb67ae85, 0x3c6ef372, 0xa54ff53a, 0x510e527f, 0x9b05688c, 0x1f83d9ab,
        0x5be0cd19,
    ];

    let mut padded = input.to_vec();
    let bit_len = (input.len() as u64) * 8;
    padded.push(0x80);
    while (padded.len() % 64) != 56 {
        padded.push(0x00);
    }
    padded.extend_from_slice(&bit_len.to_be_bytes());

    for chunk in padded.as_chunks::<64>().0 {
        let mut w = [0u32; 64];
        for i in 0..16 {
            w[i] = u32::from_be_bytes([
                chunk[i * 4],
                chunk[i * 4 + 1],
                chunk[i * 4 + 2],
                chunk[i * 4 + 3],
            ]);
        }
        for i in 16..64 {
            let s0 = w[i - 15].rotate_right(7) ^ w[i - 15].rotate_right(18) ^ (w[i - 15] >> 3);
            let s1 = w[i - 2].rotate_right(17) ^ w[i - 2].rotate_right(19) ^ (w[i - 2] >> 10);
            w[i] = w[i - 16]
                .wrapping_add(s0)
                .wrapping_add(w[i - 7])
                .wrapping_add(s1);
        }

        let [mut a, mut b, mut c, mut d, mut e, mut f, mut g, mut h_val] = h;

        for i in 0..64 {
            let s1 = e.rotate_right(6) ^ e.rotate_right(11) ^ e.rotate_right(25);
            let ch = (e & f) ^ (!e & g);
            let temp1 = h_val
                .wrapping_add(s1)
                .wrapping_add(ch)
                .wrapping_add(K256[i])
                .wrapping_add(w[i]);
            let s0 = a.rotate_right(2) ^ a.rotate_right(13) ^ a.rotate_right(22);
            let maj = (a & b) ^ (a & c) ^ (b & c);
            let temp2 = s0.wrapping_add(maj);

            h_val = g;
            g = f;
            f = e;
            e = d.wrapping_add(temp1);
            d = c;
            c = b;
            b = a;
            a = temp1.wrapping_add(temp2);
        }

        h[0] = h[0].wrapping_add(a);
        h[1] = h[1].wrapping_add(b);
        h[2] = h[2].wrapping_add(c);
        h[3] = h[3].wrapping_add(d);
        h[4] = h[4].wrapping_add(e);
        h[5] = h[5].wrapping_add(f);
        h[6] = h[6].wrapping_add(g);
        h[7] = h[7].wrapping_add(h_val);
    }

    let mut out = [0u8; 32];
    for i in 0..8 {
        let bytes = h[i].to_be_bytes();
        out[i * 4..i * 4 + 4].copy_from_slice(&bytes);
    }
    out
}

fn hex_sha256(input: &[u8]) -> String {
    sha256_bytes(input)
        .iter()
        .map(|b| format!("{:02x}", b))
        .collect()
}