csv

type CSVWritable interface {
	DataProvider
	WriteHeaders(w io.Writer, opts Options) error
	WriteRow(w io.Writer, index int, opts Options) error
}

type ColumnStatistics struct {
	Name            string
	Index           int
	DataType        string // "numeric", "categorical", "mixed"
	NonMissing      int
	Missing         int
	MissingPercent  float64
	Mean            float64 // For numeric columns
	StdDev          float64 // For numeric columns
	Min             float64 // For numeric columns
	Max             float64 // For numeric columns
	UniqueValues    int     // For categorical columns
	HasZeroVariance bool    // Warning flag
}

type Data struct {
	// Core numeric data (always present for PCA)
	Matrix      types.Matrix // Numeric data matrix
	Headers     []string     // Column names
	RowNames    []string     // Row names
	MissingMask [][]bool     // Track missing values (true = missing)
	Rows        int          // Number of data rows
	Columns     int          // Number of data columns

	// Additional data types (optional)
	StringData           [][]string           // Raw string data (for GoCSV)
	CategoricalColumns   map[string][]string  // Categorical columns by name
	NumericTargetColumns map[string][]float64 // Numeric target columns
}

type DataProvider interface {
	GetHeaders() []string
	GetRowNames() []string
	GetDimensions() (rows, cols int)
	HasNumericData() bool
	HasStringData() bool
}

type Options struct {
	// Parsing options
	Delimiter        rune      // Field delimiter: ',', ';', '\t'
	DecimalSeparator rune      // Decimal separator: '.', ','
	HasHeaders       bool      // First row contains column names
	HasRowNames      bool      // First column contains row names
	NullValues       []string  // Strings to treat as missing values
	ParseMode        ParseMode // How to parse the data
	TargetSuffix     string    // Suffix to identify target columns (e.g., "#target")

	// Reading options (for large files)
	SkipRows      int   // Number of rows to skip at start
	MaxRows       int   // Maximum rows to read (0 for all)
	Columns       []int // Specific columns to read (empty for all)
	StreamingMode bool  // Enable streaming for large files

	// Writing options
	FloatFormat byte // Format for float output: 'g', 'f', 'e'
	Precision   int  // Decimal precision for float output (-1 for auto)
}

type ParseMode int

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

type ValidationResult struct {
	Valid       bool
	Errors      []string
	Warnings    []string
	ColumnStats []ColumnStatistics
}

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

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