Struct FileSourceConfig 

pub struct FileSourceConfig {

    pub source_id: SourceId,
    pub root: PathBuf,
    pub trust: f32,
    pub category_trust: HashMap<CategoryId, f32>,
    pub follow_links: bool,
    pub text_files_only: bool,
    pub group_by_directory: bool,
    pub title_replace_underscores: bool,
    pub include_date_aware_default_recipe: bool,
    pub index_dir: Option<PathBuf>,
    pub default_triplet_recipes: Vec<TripletRecipe>,
    pub taxonomy_builder: TaxonomyBuilder,
    pub section_builder: SectionBuilder,
}

Configuration for a generic filesystem-backed data source.

Fields

source_id: SourceId

Stable source identifier used in records and persistence keys.

root: PathBuf

Root directory containing source files.

trust: f32

Default quality trust score applied to generated records.

category_trust: HashMap<CategoryId, f32>

Optional trust overrides keyed by taxonomy segment.

follow_links: bool

Whether to follow symlinks during index walking.

text_files_only: bool

Whether indexing should include only text files.

group_by_directory: bool

Whether deterministic directory grouping is enabled.

title_replace_underscores: bool

Whether title extraction should replace underscores with spaces.

include_date_aware_default_recipe: bool

Whether default recipe set includes the date-aware negative lane.

index_dir: Option<PathBuf>

Optional directory used for persisted file-corpus index stores.

When None, file-corpus indexing uses the managed cache discovery root. Set this in tests to keep index writes inside temporary directories.

default_triplet_recipes: Vec<TripletRecipe>

Optional default recipes returned by this source.

taxonomy_builder: TaxonomyBuilder

Taxonomy builder invoked per file.

section_builder: SectionBuilder

Section builder invoked per file.

Implementations

impl FileSourceConfig

pub fn new(source_id: impl Into<SourceId>, root: impl Into<PathBuf>) -> Self

Create a config for a filesystem source with explicit id and root.

pub fn with_trust(self, trust: f32) -> Self

Override default trust score.

pub fn with_category_trust( self, category: impl Into<String>, trust: f32, ) -> Self

Add a taxonomy-segment trust override.

pub fn with_follow_links(self, follow_links: bool) -> Self

Override whether symlinks are followed during index walk.

pub fn with_text_files_only(self, text_files_only: bool) -> Self

Override whether index walk includes only text files.

pub fn with_directory_grouping(self, group_by_directory: bool) -> Self

Enable or disable deterministic directory grouping.

pub fn with_title_replace_underscores(self, replace_underscores: bool) -> Self

Set whether title extraction replaces underscores with spaces.

pub fn with_date_aware_default_recipe(self, include: bool) -> Self

Enable/disable the date-aware default recipe lane (WrongPublicationDate).

“Date-aware” here uses publication-date metadata on records (for example taxonomy/meta date fields), not filesystem timestamps from source files.

pub fn with_index_dir(self, index_dir: impl Into<PathBuf>) -> Self

Override the directory used to persist file-corpus index stores.

pub fn with_default_triplet_recipes(self, recipes: Vec<TripletRecipe>) -> Self

Set source-provided default triplet recipes.

pub fn with_taxonomy_builder(self, taxonomy_builder: TaxonomyBuilder) -> Self

Set a custom taxonomy builder.

pub fn with_section_builder(self, section_builder: SectionBuilder) -> Self

Set a custom section builder.

Trait Implementations

impl Clone for FileSourceConfig

fn clone(&self) -> FileSourceConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.