Vellum/internal/providers/tmdb.go

package providers

import (
	"context"
	"encoding/json"
	"errors"
	"fmt"
	"net/http"
	"net/url"
	"sort"
	"strconv"
	"strings"
	"sync"
	"time"
)

// TMDB matches video — films and television — against The Movie Database.
//
// A Query does not say which of the two it is, and TMDB keeps them in separate
// namespaces behind separate endpoints, so Search asks both at once rather than
// guessing from the title. The results are merged by TMDB's own popularity,
// which is the only ranking signal available without a second request per item.
type TMDB struct {
	APIKey  string
	BaseURL string
	Client  *http.Client
}

// NewTMDB builds the provider. An empty key yields a provider that finds
// nothing, so a misconfiguration degrades rather than erroring.
func NewTMDB(apiKey string) *TMDB {
	return &TMDB{
		APIKey:  apiKey,
		BaseURL: "https://api.themoviedb.org/3",
		// One client for both requests: connection reuse across the pair is
		// most of what makes the concurrent form cheap.
		Client: &http.Client{Timeout: 15 * time.Second},
	}
}

func (p *TMDB) Name() string { return "tmdb" }

func (p *TMDB) Kinds() []string { return []string{"video"} }

// tmdbResponse is the subset of a TMDB search payload we read. Film and
// television share the envelope but not the field names, so the few that differ
// are read together and picked between below.
type tmdbResponse struct {
	Results []struct {
		ID           int     `json:"id"`
		Title        string  `json:"title"`
		Name         string  `json:"name"`
		Overview     string  `json:"overview"`
		ReleaseDate  string  `json:"release_date"`
		FirstAirDate string  `json:"first_air_date"`
		PosterPath   string  `json:"poster_path"`
		Popularity   float64 `json:"popularity"`
	} `json:"results"`
}

// tmdbHit carries a candidate with the popularity it was ranked by, which
// Candidate has no field for.
type tmdbHit struct {
	cand       Candidate
	popularity float64
}

func (p *TMDB) Search(ctx context.Context, q Query) ([]Candidate, error) {
	title := strings.TrimSpace(q.Title)
	if p.APIKey == "" || title == "" {
		return nil, nil
	}

	// The two endpoints run at once: the call costs one response time, not two.
	// Whichever fails first cancels the other, so one dead endpoint cannot hold
	// the result hostage while the healthy one is waited out.
	ctx, cancel := context.WithCancel(ctx)
	defer cancel()

	var (
		wg    sync.WaitGroup
		hits  [2][]tmdbHit
		errs  [2]error
		fetch = func(i int, endpoint, kind string) {
			defer wg.Done()
			hits[i], errs[i] = p.search(ctx, endpoint, kind, title, q.Year)
			if errs[i] != nil {
				cancel()
			}
		}
	)
	wg.Add(2)
	go fetch(0, "/search/movie", "movie")
	go fetch(1, "/search/tv", "tv")
	wg.Wait()

	merged := append(hits[0], hits[1]...)
	if len(merged) == 0 && (errs[0] != nil || errs[1] != nil) {
		// Only a total failure is an error. One endpoint answering is a result.
		return nil, errors.Join(errs[0], errs[1])
	}
	sort.SliceStable(merged, func(i, j int) bool {
		return merged[i].popularity > merged[j].popularity
	})

	out := make([]Candidate, 0, len(merged))
	for _, h := range merged {
		out = append(out, h.cand)
	}
	return out, nil
}

// search queries one TMDB endpoint and maps its results.
func (p *TMDB) search(ctx context.Context, endpoint, kind, title string, year int) ([]tmdbHit, error) {
	params := url.Values{}
	params.Set("api_key", p.APIKey)
	params.Set("query", title)
	params.Set("limit", "5")
	if year > 0 {
		// The two namespaces spell the year filter differently.
		if kind == "movie" {
			params.Set("year", strconv.Itoa(year))
		} else {
			params.Set("first_air_date_year", strconv.Itoa(year))
		}
	}

	req, err := http.NewRequestWithContext(ctx, http.MethodGet,
		strings.TrimRight(p.BaseURL, "/")+endpoint+"?"+params.Encode(), nil)
	if err != nil {
		return nil, err
	}
	resp, err := p.Client.Do(req)
	if err != nil {
		return nil, fmt.Errorf("tmdb: %s: %w", kind, err)
	}
	defer resp.Body.Close()
	if resp.StatusCode != http.StatusOK {
		return nil, fmt.Errorf("tmdb: %s: status %d", kind, resp.StatusCode)
	}

	var parsed tmdbResponse
	if err := json.NewDecoder(resp.Body).Decode(&parsed); err != nil {
		return nil, fmt.Errorf("tmdb: %s: decode: %w", kind, err)
	}

	out := make([]tmdbHit, 0, len(parsed.Results))
	for _, r := range parsed.Results {
		name := r.Title
		if name == "" {
			name = r.Name // television has no title, only a name
		}
		if name == "" {
			continue
		}
		date := r.ReleaseDate
		if date == "" {
			date = r.FirstAirDate
		}
		out = append(out, tmdbHit{
			cand: Candidate{
				Provider: p.Name(),
				// Namespaced: TMDB keeps films and television in separate id
				// spaces, so film 603 and show 603 are different works. A bare
				// id would store them identically and collide in provider_ids.
				ExternalID:  kind + "/" + strconv.Itoa(r.ID),
				Title:       name,
				Year:        parseYear(date),
				Description: r.Overview,
				CoverURL:    tmdbPoster(r.PosterPath),
			},
			popularity: r.Popularity,
		})
	}
	return out, nil
}

// tmdbPoster expands TMDB's relative poster path into a full image URL. The
// bytes are never fetched here: the enricher caches covers it actually uses.
func tmdbPoster(path string) string {
	if path == "" {
		return ""
	}
	return "https://image.tmdb.org/t/p/w500" + path
}