package adapters

import (
	"bytes"
	"context"
	"encoding/json"
	"fmt"
	"io"
	"net/http"
	"os"
	"strings"
	"time"

	"github.com/xuri/excelize/v2"

	"github.com/missionworkforce/missionmatch/internal/modules/ontology"
)

// PSCSource ingests the federal Product and Service Code manual, the cleanest fully enumerable procurement
// taxonomy. It reads the published XLSX (a URL or a local file) and yields one KindPSC record per code.
type PSCSource struct {
	url    string
	client *http.Client
}

// NewPSCSource builds the adapter. url is the PSC manual XLSX location: an http or https URL
// (acquisition.gov publishes it), or a local file path or file:// URL for air gapped operation where
// pulling acquisition.gov at runtime is undesirable.
func NewPSCSource(url string, client *http.Client) *PSCSource {
	if client == nil {
		client = &http.Client{Timeout: 60 * time.Second}
	}
	return &PSCSource{url: url, client: client}
}

// Name is the provenance label.
func (s *PSCSource) Name() string { return "PSC Manual" }

// Fetch loads the XLSX and yields one record per code.
func (s *PSCSource) Fetch(ctx context.Context, yield func(ontology.VocabRecord) error) error {
	data, err := s.load(ctx)
	if err != nil {
		return err
	}
	recs, err := parsePSC(data)
	if err != nil {
		return err
	}
	for _, r := range recs {
		if err := yield(r); err != nil {
			return err
		}
	}
	return nil
}

// load reads the PSC manual bytes from an http or https URL, or from a local file (a plain path or a
// file:// URL) for air gapped operation. Choosing the source is configuration, the same cloud portability
// posture the forecast feed uses.
func (s *PSCSource) load(ctx context.Context) ([]byte, error) {
	if strings.HasPrefix(s.url, "http://") || strings.HasPrefix(s.url, "https://") {
		return s.download(ctx)
	}
	path := strings.TrimPrefix(s.url, "file://")
	b, err := os.ReadFile(path)
	if err != nil {
		return nil, fmt.Errorf("psc: read local file: %w", err)
	}
	return b, nil
}

func (s *PSCSource) download(ctx context.Context) ([]byte, error) {
	req, err := http.NewRequestWithContext(ctx, http.MethodGet, s.url, nil)
	if err != nil {
		return nil, fmt.Errorf("psc: new request: %w", err)
	}
	resp, err := s.client.Do(req)
	if err != nil {
		return nil, fmt.Errorf("psc: get: %w", err)
	}
	defer resp.Body.Close()
	if resp.StatusCode != http.StatusOK {
		return nil, fmt.Errorf("psc: provider returned %s", resp.Status)
	}
	return io.ReadAll(resp.Body)
}

// parsePSC reads the PSC manual XLSX bytes and returns one VocabRecord per active code. It matches the
// exact PSC CODE and PRODUCT AND SERVICE CODE NAME header cells (the real manual has several columns that
// loosely contain CODE or NAME, so a substring match picks the wrong column) and skips codes that carry an
// end date, keeping only active codes.
func parsePSC(data []byte) ([]ontology.VocabRecord, error) {
	f, err := excelize.OpenReader(bytes.NewReader(data))
	if err != nil {
		return nil, fmt.Errorf("psc: open xlsx: %w", err)
	}
	defer f.Close()
	sheet := f.GetSheetName(0)
	rows, err := f.GetRows(sheet)
	if err != nil {
		return nil, fmt.Errorf("psc: read rows: %w", err)
	}
	if len(rows) < 2 {
		return nil, fmt.Errorf("psc: sheet has no data rows")
	}
	codeCol, nameCol, endCol := -1, -1, -1
	for i, h := range rows[0] {
		switch strings.TrimSpace(strings.ToUpper(h)) {
		case "PSC CODE":
			codeCol = i
		case "PRODUCT AND SERVICE CODE NAME":
			nameCol = i
		case "END DATE":
			endCol = i
		}
	}
	if codeCol < 0 || nameCol < 0 {
		return nil, fmt.Errorf("psc: could not find the PSC CODE and PRODUCT AND SERVICE CODE NAME columns in the header %v", rows[0])
	}
	out := make([]ontology.VocabRecord, 0, len(rows)-1)
	for _, row := range rows[1:] {
		if codeCol >= len(row) || nameCol >= len(row) {
			continue
		}
		code := strings.TrimSpace(row[codeCol])
		name := strings.TrimSpace(row[nameCol])
		if code == "" || name == "" {
			continue
		}
		// Skip expired codes (those carrying an end date), keeping only active PSC codes.
		if endCol >= 0 && endCol < len(row) && strings.TrimSpace(row[endCol]) != "" {
			continue
		}
		raw, _ := json.Marshal(map[string]string{"code": code, "name": name})
		out = append(out, ontology.VocabRecord{
			SourceName:     "PSC Manual",
			SourceRecordID: code,
			Kind:           ontology.KindPSC,
			Term:           name,
			Code:           code,
			Category:       "procurement",
			Raw:            raw,
		})
	}
	return out, nil
}