shotext 0.1.6

High-performance OCR & Full-Text Search Indexer for macOS Screenshots
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

// OCR module — extracts text from screenshot images using Tesseract.
use tesseract::Tesseract;

/// Run Tesseract OCR on the image at `path` using the given language code
/// (e.g. "eng", "swe", "deu").
pub fn extract_text(path: &str, language: &str) -> anyhow::Result<String> {
    let mut tes = Tesseract::new(None, Some(language))?.set_image(path)?;
    let text = tes.get_text()?;
    // Trim trailing whitespace/newlines that Tesseract often appends
    Ok(text.trim().to_string())
}

/// Return at most `max_chars` characters, appending "…" if truncated.
pub fn truncate(text: &str, max_chars: usize) -> String {
    let char_count = text.chars().count();
    if char_count <= max_chars {
        text.to_string()
    } else {
        let end = text
            .char_indices()
            .nth(max_chars)
            .map(|(i, _)| i)
            .unwrap_or(text.len());
        format!("{}…", &text[..end])
    }
}