seams 0.1.14

High-throughput sentence extractor for Project Gutenberg texts with dialog-aware detection
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

// WHY: Main detector interface with dual API for borrowed vs owned data experiments
// Enables performance comparison between mmap-friendly and async I/O-friendly approaches


pub mod normalization;
pub mod dialog_detector;
// pub mod dialog_detector2;  // Disabled
pub mod abbreviations;

// Re-export core types
pub use normalization::normalize_sentence;
pub use abbreviations::AbbreviationChecker;

/// Position in a text file using 1-based indexing as specified in PRD section 2
#[derive(Debug, Clone, PartialEq, Eq)]
pub struct Span {
    pub start_line: usize,
    pub start_col: usize,
    pub end_line: usize,
    pub end_col: usize,
}


/// Borrowed variant - zero allocation detection (mmap-optimized)
#[derive(Debug, Clone)]
pub struct DetectedSentenceBorrowed<'a> {
    pub index: usize,
    pub raw_content: &'a str,  // Borrowed from source text
    pub span: Span,
}

impl<'a> DetectedSentenceBorrowed<'a> {
    /// Normalize content with new allocation
    pub fn normalize(&self) -> String {
        normalize_sentence(self.raw_content)
    }
}