articles

type ArticleParser struct {
	// contains filtered or unexported fields
}

type ContentScore struct {
	Node            *html.Node
	Score           float64
	TextLength      int
	LinkDensity     float64
	ParagraphCount  int
	AncestorDepth   int
	ConfidenceLevel float64
}

type ExtractionResult struct {
	Content          string
	Title            string
	Author           string
	PublishedDate    string
	SiteName         string
	Language         string
	Confidence       float64
	ExtractionMethod string // "heuristic" or "xpath" or "dual"
}

type HeuristicExtractor struct {
	// contains filtered or unexported fields
}

type MetadataExtractor struct{}

type ParsedContent struct {
	Title            string
	Author           string
	Date             string
	Content          string
	URL              string
	Confidence       float64 // 0-1 scale, confidence in extraction quality
	ExtractionMethod string  // "xpath", "heuristic", "dual-validated", etc.
}

type Parser interface {
	// ParseURL extracts article content from a given URL
	ParseURL(url string) (*ParsedContent, error)
	// Convert HTML content directly to markdown using domain-specific rules
	Convert(htmlContent, domain, sourceURL string) (string, error)
	// GetSupportedDomains returns a list of domains that have parsing rules
	GetSupportedDomains() []string
	// SaveArticle saves the parsed content to filesystem and returns file paths
	SaveArticle(content *ParsedContent, storageDir string) (markdownPath, htmlPath string, err error)
}

type ParsingRule struct {
	Domain            string
	Title             string
	Author            string
	Date              string
	Body              string
	Strip             []string // XPath selectors for elements to remove
	StripIDsOrClasses []string
	TestURLs          []string
	Headers           map[string]string
	Prune             bool
	Tidy              bool
}

type Scorer struct {
	// contains filtered or unexported fields
}