package index

import (
	"encoding/gob"
	"os"
	"searchEngine/core/analyzer"
	"sync"
)

type Posting struct {
	DocId          int
	BodyPositions  []int
	TitlePositions []int
}

type InvertedIndex struct {
	Mu       sync.RWMutex `gob:"-"`
	Data     map[string][]Posting
	DocNames map[int]string
}

func New() *InvertedIndex {
	return &InvertedIndex{
		Data:     make(map[string][]Posting),
		DocNames: make(map[int]string),
	}
}

func (idx *InvertedIndex) Add(docID int, title string, bodyTokens []string) {
	titleTokens := analyzer.ProcessText(title)

	type fieldStats struct {
		bodyPositions  []int
		titlePositions []int
	}
	docStats := make(map[string]fieldStats)

	for position, token := range titleTokens {
		stats := docStats[token]
		stats.titlePositions = append(stats.titlePositions, position)
		docStats[token] = stats
	}

	for position, token := range bodyTokens {
		stats := docStats[token]
		stats.bodyPositions = append(stats.bodyPositions, position)
		docStats[token] = stats
	}

	idx.Mu.Lock()
	defer idx.Mu.Unlock()

	idx.DocNames[docID] = title

	for token, stats := range docStats {
		idx.Data[token] = append(idx.Data[token], Posting{
			DocId:          docID,
			BodyPositions:  stats.bodyPositions,
			TitlePositions: stats.titlePositions,
		})
	}
}

func (idx *InvertedIndex) Save(filepath string) error {
	idx.Mu.RLock()
	defer idx.Mu.RUnlock()

	file, err := os.Create(filepath)
	if err != nil {
		return err
	}
	defer file.Close()

	encoder := gob.NewEncoder(file)
	return encoder.Encode(idx)
}

func Load(filepath string) (*InvertedIndex, error) {
	file, err := os.Open(filepath)
	if err != nil {
		return nil, err
	}
	defer file.Close()

	idx := &InvertedIndex{}
	decoder := gob.NewDecoder(file)
	err = decoder.Decode(idx)
	if err != nil {
		return nil, err
	}
	return idx, nil
}