//                           _       _
// __      _____  __ ___   ___  __ _| |_ ___
// \ \ /\ / / _ \/ _` \ \ / / |/ _` | __/ _ \
//  \ V  V /  __/ (_| |\ V /| | (_| | ||  __/
//   \_/\_/ \___|\__,_| \_/ |_|\__,_|\__\___|
//
//  Copyright © 2016 - 2024 Weaviate B.V. All rights reserved.
//
//  CONTACT: hello@weaviate.io
//

package lsmkv

import (
	"bufio"
	"bytes"
	"errors"
	"fmt"
	"io"

	"github.com/weaviate/weaviate/adapters/repos/db/lsmkv/segmentindex"
	"github.com/weaviate/weaviate/entities/lsmkv"
)

type compactorReplace struct {
	// c1 is always the older segment, so when there is a conflict c2 wins
	// (because of the replace strategy)
	c1 *segmentCursorReplace
	c2 *segmentCursorReplace

	// the level matching those of the cursors
	currentLevel uint16
	// Tells if tombstones or keys without corresponding values
	// can be removed from merged segment.
	// (left segment is root (1st) one, keepTombstones is off for bucket)
	cleanupTombstones   bool
	secondaryIndexCount uint16

	w                io.WriteSeeker
	bufw             *bufio.Writer
	scratchSpacePath string
}

func newCompactorReplace(w io.WriteSeeker,
	c1, c2 *segmentCursorReplace, level, secondaryIndexCount uint16,
	scratchSpacePath string, cleanupTombstones bool,
) *compactorReplace {
	return &compactorReplace{
		c1:                  c1,
		c2:                  c2,
		w:                   w,
		bufw:                bufio.NewWriterSize(w, 256*1024),
		currentLevel:        level,
		cleanupTombstones:   cleanupTombstones,
		secondaryIndexCount: secondaryIndexCount,
		scratchSpacePath:    scratchSpacePath,
	}
}

func (c *compactorReplace) do() error {
	if err := c.init(); err != nil {
		return fmt.Errorf("init: %w", err)
	}

	kis, err := c.writeKeys()
	if err != nil {
		return fmt.Errorf("write keys: %w", err)
	}

	if err := c.writeIndices(kis); err != nil {
		return fmt.Errorf("write indices: %w", err)
	}

	// flush buffered, so we can safely seek on underlying writer
	if err := c.bufw.Flush(); err != nil {
		return fmt.Errorf("flush buffered: %w", err)
	}

	var dataEnd uint64 = segmentindex.HeaderSize
	if len(kis) > 0 {
		dataEnd = uint64(kis[len(kis)-1].ValueEnd)
	}

	if err := c.writeHeader(c.currentLevel, 0, c.secondaryIndexCount, dataEnd); err != nil {
		return fmt.Errorf("write header: %w", err)
	}

	return nil
}

func (c *compactorReplace) init() error {
	// write a dummy header, we don't know the contents of the actual header yet,
	// we will seek to the beginning and overwrite the actual header at the very
	// end

	if _, err := c.bufw.Write(make([]byte, segmentindex.HeaderSize)); err != nil {
		return fmt.Errorf("write empty header: %w", err)
	}

	return nil
}

func (c *compactorReplace) writeKeys() ([]segmentindex.Key, error) {
	res1, err1 := c.c1.firstWithAllKeys()
	res2, err2 := c.c2.firstWithAllKeys()

	// the (dummy) header was already written, this is our initial offset
	offset := segmentindex.HeaderSize

	var kis []segmentindex.Key

	for {
		if res1.primaryKey == nil && res2.primaryKey == nil {
			break
		}
		if bytes.Equal(res1.primaryKey, res2.primaryKey) {
			if !(c.cleanupTombstones && errors.Is(err2, lsmkv.Deleted)) {
				ki, err := c.writeIndividualNode(offset, res2.primaryKey, res2.value,
					res2.secondaryKeys, errors.Is(err2, lsmkv.Deleted))
				if err != nil {
					return nil, fmt.Errorf("write individual node (equal keys): %w", err)
				}

				offset = ki.ValueEnd
				kis = append(kis, ki)
			}
			// advance both!
			res1, err1 = c.c1.nextWithAllKeys()
			res2, err2 = c.c2.nextWithAllKeys()
			continue
		}

		if (res1.primaryKey != nil && bytes.Compare(res1.primaryKey, res2.primaryKey) == -1) || res2.primaryKey == nil {
			// key 1 is smaller
			if !(c.cleanupTombstones && errors.Is(err1, lsmkv.Deleted)) {
				ki, err := c.writeIndividualNode(offset, res1.primaryKey, res1.value,
					res1.secondaryKeys, errors.Is(err1, lsmkv.Deleted))
				if err != nil {
					return nil, fmt.Errorf("write individual node (res1.primaryKey smaller)")
				}

				offset = ki.ValueEnd
				kis = append(kis, ki)
			}
			res1, err1 = c.c1.nextWithAllKeys()
		} else {
			// key 2 is smaller
			if !(c.cleanupTombstones && errors.Is(err2, lsmkv.Deleted)) {
				ki, err := c.writeIndividualNode(offset, res2.primaryKey, res2.value,
					res2.secondaryKeys, errors.Is(err2, lsmkv.Deleted))
				if err != nil {
					return nil, fmt.Errorf("write individual node (res2.primaryKey smaller): %w", err)
				}

				offset = ki.ValueEnd
				kis = append(kis, ki)
			}
			res2, err2 = c.c2.nextWithAllKeys()
		}
	}

	return kis, nil
}

func (c *compactorReplace) writeIndividualNode(offset int, key, value []byte,
	secondaryKeys [][]byte, tombstone bool,
) (segmentindex.Key, error) {
	segNode := segmentReplaceNode{
		offset:              offset,
		tombstone:           tombstone,
		value:               value,
		primaryKey:          key,
		secondaryIndexCount: c.secondaryIndexCount,
		secondaryKeys:       secondaryKeys,
	}

	return segNode.KeyIndexAndWriteTo(c.bufw)
}

func (c *compactorReplace) writeIndices(keys []segmentindex.Key) error {
	indices := &segmentindex.Indexes{
		Keys:                keys,
		SecondaryIndexCount: c.secondaryIndexCount,
		ScratchSpacePath:    c.scratchSpacePath,
	}

	_, err := indices.WriteTo(c.bufw)
	return err
}

// writeHeader assumes that everything has been written to the underlying
// writer and it is now safe to seek to the beginning and override the initial
// header
func (c *compactorReplace) writeHeader(level, version, secondaryIndices uint16,
	startOfIndex uint64,
) error {
	if _, err := c.w.Seek(0, io.SeekStart); err != nil {
		return fmt.Errorf("seek to beginning to write header: %w", err)
	}

	h := &segmentindex.Header{
		Level:            level,
		Version:          version,
		SecondaryIndices: secondaryIndices,
		Strategy:         segmentindex.StrategyReplace,
		IndexStart:       startOfIndex,
	}

	if _, err := h.WriteTo(c.w); err != nil {
		return err
	}

	return nil
}