refactor: split modules and harden scraping workflows

internal/handler/streaming.go

@@ -4,6 +4,7 @@ package handler
 import (
 	"context"
 	"errors"
+	"io"
 	"net/http"
 
 	"github.com/gin-gonic/gin"
@@ -79,16 +80,99 @@ func imageProxyHandler(svc *service.Container) gin.HandlerFunc {
 	}
 }
 
+func cloudArtworkProxyHandler(svc *service.Container) gin.HandlerFunc {
+	return func(c *gin.Context) {
+		typ := c.Param("type")
+		ref := c.Query("ref")
+		if !service.IsAdminCloudConfigurable(typ) {
+			c.JSON(http.StatusBadRequest, gin.H{"error": "unsupported cloud provider"})
+			return
+		}
+		if ref == "" || !isCloudImageRef(ref) {
+			c.JSON(http.StatusBadRequest, gin.H{"error": "image ref required"})
+			return
+		}
+		if svc == nil || svc.ImageProxy == nil {
+			c.JSON(http.StatusServiceUnavailable, gin.H{"error": "image proxy unavailable"})
+			return
+		}
+		stableKey := typ + ":" + ref
+		if svc.ImageProxy.ServeCloudCached(c.Writer, c.Request, stableKey) {
+			return
+		}
+		if svc.StorageCfg == nil {
+			c.JSON(http.StatusServiceUnavailable, gin.H{"error": "cloud storage service unavailable"})
+			return
+		}
+		link, err := svc.StorageCfg.CloudResolve(c.Request.Context(), typ, ref, c.Request.UserAgent())
+		if err != nil {
+			c.JSON(http.StatusBadGateway, gin.H{"error": err.Error()})
+			return
+		}
+		if err := svc.ImageProxy.ServeCloudResolved(c.Request.Context(), c.Writer, c.Request, stableKey, link); err != nil {
+			c.JSON(http.StatusBadGateway, gin.H{"error": err.Error()})
+			return
+		}
+	}
+}
+
+type scrapeRequest struct {
+	EpisodeArtwork *bool `json:"episode_artwork"`
+	EpisodeImages  *bool `json:"episode_images"`
+	RefreshMatched *bool `json:"refresh_matched"`
+	IncludeMatched *bool `json:"include_matched"`
+}
+
+func (r scrapeRequest) episodeArtworkOption() *bool {
+	if r.EpisodeImages != nil {
+		return r.EpisodeImages
+	}
+	return r.EpisodeArtwork
+}
+
+func (r scrapeRequest) includeMatchedOption() bool {
+	if r.IncludeMatched != nil {
+		return *r.IncludeMatched
+	}
+	if r.RefreshMatched != nil {
+		return *r.RefreshMatched
+	}
+	return false
+}
+
+func scrapeOptionsFromRequest(c *gin.Context, retryNoMatch bool) (service.ScrapeOptions, error) {
+	options := service.ScrapeOptions{RetryNoMatch: retryNoMatch}
+	if c.Request.Body == nil || c.Request.ContentLength == 0 {
+		return options, nil
+	}
+	var req scrapeRequest
+	if err := c.ShouldBindJSON(&req); err != nil {
+		if errors.Is(err, io.EOF) {
+			return options, nil
+		}
+		return options, err
+	}
+	options.EpisodeArtwork = req.episodeArtworkOption()
+	options.IncludeMatched = req.includeMatchedOption()
+	return options, nil
+}
+
 // scrapeOneHandler enriches a single media via the configured scraper chain.
 func scrapeOneHandler(svc *service.Container) gin.HandlerFunc {
 	return func(c *gin.Context) {
+		options, err := scrapeOptionsFromRequest(c, true)
+		if err != nil {
+			c.JSON(http.StatusBadRequest, gin.H{"error": "invalid scrape options"})
+			return
+		}
+		options.IncludeMatched = true
 		m, err := svc.Repo.Media.FindByID(c.Request.Context(), c.Param("id"))
 		if err != nil || m == nil {
 			c.JSON(http.StatusNotFound, gin.H{"error": "not found"})
 			return
 		}
 		task := startScrapeHTTPTask(svc, "手动刮削媒体", m.Title, m.Path)
-		if err := svc.Scraper.EnrichOne(c.Request.Context(), m); err != nil {
+		if err := svc.Scraper.EnrichOneWithOptions(c.Request.Context(), m, options); err != nil {
 			finishHTTPTask(task, err, "scrape", "手动刮削媒体失败", nil, nil)
 			c.JSON(http.StatusInternalServerError, gin.H{"error": err.Error()})
 			return
@@ -103,10 +187,16 @@ func scrapeOneHandler(svc *service.Container) gin.HandlerFunc {
 	}
 }
 
-// scrapeLibraryHandler retries every pending/no_match media in a library.
+// scrapeLibraryHandler manually refreshes every scrapeable row in a library.
 func scrapeLibraryHandler(svc *service.Container) gin.HandlerFunc {
 	return func(c *gin.Context) {
 		libID := c.Param("id")
+		options, err := scrapeOptionsFromRequest(c, true)
+		if err != nil {
+			c.JSON(http.StatusBadRequest, gin.H{"error": "invalid scrape options"})
+			return
+		}
+		options.IncludeMatched = true
 		var task *service.TaskHandle
 		if lib, err := svc.Repo.Library.FindByID(c.Request.Context(), libID); err == nil && lib != nil {
 			task = startScrapeHTTPTask(svc, "手动刮削媒体库", lib.Name, lib.Path)
@@ -115,9 +205,16 @@ func scrapeLibraryHandler(svc *service.Container) gin.HandlerFunc {
 		}
 		// Run in the background so HTTP returns instantly; the WS hub
 		// pushes per-item progress on the "scrape" topic.
-		go func(libID string, task *service.TaskHandle) {
-			matched, err := svc.Scraper.EnrichLibrary(context.Background(), libID, true)
-			metrics := map[string]int64{"matched": int64(matched)}
+		go func(libID string, task *service.TaskHandle, options service.ScrapeOptions) {
+			result, err := svc.Scraper.EnrichLibraryDetailedWithOptions(context.Background(), libID, options)
+			metrics := map[string]int64{
+				"matched":    int64(result.Matched),
+				"processed":  int64(result.Processed),
+				"candidates": int64(result.Candidates),
+			}
+			if result.Failed > 0 {
+				metrics["errors"] = int64(result.Failed)
+			}
 			stage := "completed"
 			message := "手动刮削媒体库结束"
 			if err != nil {
@@ -125,7 +222,7 @@ func scrapeLibraryHandler(svc *service.Container) gin.HandlerFunc {
 				message = "手动刮削媒体库失败"
 			}
 			finishHTTPTask(task, err, stage, message, metrics, nil)
-		}(libID, task)
+		}(libID, task, options)
 		c.JSON(http.StatusAccepted, gin.H{"status": "scraping"})
 	}
 }