Struct ExtractOptions 

pub struct ExtractOptions {
    pub debug: bool,
    pub remove_style_tags: bool,
    pub ready_for_epub: bool,
    pub strip_unlikelys: bool,
    pub weight_classes: bool,
    pub keep_classes: bool,
    pub classes_to_preserve: HashSet<String>,
    pub clean_conditionally: bool,
    pub max_elements_to_parse: u16,
    pub n_top_candidates: u16,
    pub char_threshold: u16,
    pub link_density_modifier: f64,
}

Knobs that control the behaviour of the extraction algorithm.

All fields have sensible defaults via Default; start there and only override what you need.

Examples

use readable_rs::ExtractOptions;

let mut opts = ExtractOptions::default();
opts.char_threshold = 200;   // accept shorter articles
opts.remove_style_tags = false; // keep <style> elements

Fields

debug: bool

Enable extra eprintln! tracing inside the algorithm (gated behind debug_assertions in release builds).

remove_style_tags: bool

Strip all <style> elements from the document before extraction.

ready_for_epub: bool

When true, apply additional cleanup passes that produce output suitable for embedding in an EPUB (e.g. stricter image handling).

strip_unlikelys: bool

Remove elements whose class / id / role strongly suggest they are navigation, ads, or other non-content. Disabling this is one of the retry strategies when the first pass yields too little text.

weight_classes: bool

Use class-name / id heuristics (positive/negative word lists) to adjust candidate scores. Disabling is another retry strategy.

keep_classes: bool

When true, preserve CSS class attributes on the output nodes (subject to classes_to_preserve). When false, all class attributes are stripped.

classes_to_preserve: HashSet<String>

The set of class names that are always kept even when keep_classes is false. Readability’s own marker classes (e.g. "page") are added automatically.

clean_conditionally: bool

Apply the “clean conditionally” pass, which removes elements with low content density (few commas, high link density, etc.). Disabling is the third retry strategy.

max_elements_to_parse: u16

Maximum number of scoring candidates to evaluate. 0 means no limit.

n_top_candidates: u16

How many top-scoring candidate nodes to retain before picking the winner. Higher values make the algorithm slightly more robust against mis-scored nodes.

char_threshold: u16

Minimum character count the extracted content must reach before it is accepted. If the first pass falls short, the algorithm retries with progressively relaxed options.

link_density_modifier: f64

An additive modifier applied to the link-density thresholds used in the “clean conditionally” pass. Positive values make the filter more lenient (tolerate higher link density).

Trait Implementations

impl Clone for ExtractOptions

fn clone(&self) -> ExtractOptions

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ExtractOptions

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for ExtractOptions

fn default() -> ExtractOptions

Returns the “default value” for a type.