expand a preview card for YouTube links

A message containing a YouTube link now carries a preview card. Discord
and X build their own card from the URL in the text, so the work is on
the bluesky side: the new preview.go asks YouTube's oEmbed endpoint for
the title, channel and thumbnail and posts them as an
app.bsky.embed.external.

Links are found by pulling candidates out of the text and parsing them
with net/url rather than by matching a URL shaped regexp, so a host like
youtube.com.example.invalid is not mistaken for the real thing.

Bluesky allows one embed per post, so a message with both attachments
and a link keeps the attachments. A failed oEmbed lookup only costs the
card, not the post.

Downloading the thumbnail wants the attachment download path, so it is
split into fetch and downloadImage, and both now go through a client
with a timeout: the event loop is single threaded and a hung request
would stall every later message.

The Dockerfile built main.go by name, which stops working now that
package main spans two files; it builds the package instead.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

Dockerfile

@@ -7,7 +7,7 @@ WORKDIR tweetdistributor
 
 RUN mkdir -p /build
 RUN go mod tidy
-RUN go build -a -tags "netgo" -tags timetzdata -installsuffix netgo -ldflags="-s -w -extldflags \"-static\"" -o=/build/tweetdistributor main.go
+RUN go build -a -tags "netgo" -tags timetzdata -installsuffix netgo -ldflags="-s -w -extldflags \"-static\"" -o=/build/tweetdistributor .
 
 FROM alpine:3.22
 

main.go

@@ -11,7 +11,9 @@ import (
 	"net/http"
 	"os"
 	"path"
+	"regexp"
 	"strings"
+	"time"
 	"tweetdistributor/discord"
 	"tweetdistributor/output"
 	"tweetdistributor/store"
@@ -28,6 +30,17 @@ const maxImagesPerPost = 4
 // Bluesky rejects blobs over 2,000,000 bytes; Twitter allows up to 5MB.
 const maxImageBytes = 2_000_000
 
+// urlPattern picks candidate links out of message text; each one is parsed
+// properly before it is judged to be a YouTube link.
+var urlPattern = regexp.MustCompile(`https?://[^\s<>"']+`)
+
+// trimURL drops the punctuation that ends the sentence rather than the URL.
+func trimURL(match string) string {
+	return strings.TrimRight(match, ".,!?、。)]}>")
+}
+
+var httpClient = &http.Client{Timeout: 30 * time.Second}
+
 // shrinkImage re-encodes (and if necessary downscales) an image until it
 // fits within maxImageBytes. Images already small enough pass through
 // untouched.
@@ -69,29 +82,44 @@ func shrinkImage(img output.Image) (output.Image, error) {
 	return output.Image{}, fmt.Errorf("%s could not be shrunk below %d bytes", img.Filename, maxImageBytes)
 }
 
+// fetch GETs url and returns its body.
+func fetch(url string) ([]byte, error) {
+	resp, err := httpClient.Get(url)
+	if err != nil {
+		return nil, err
+	}
+	defer resp.Body.Close()
+
+	data, err := io.ReadAll(resp.Body)
+	if err != nil {
+		return nil, err
+	}
+	if resp.StatusCode != http.StatusOK {
+		return nil, fmt.Errorf("status %s", resp.Status)
+	}
+	return data, nil
+}
+
+// downloadImage fetches an image and shrinks it to a postable size.
+func downloadImage(url, filename, contentType string) (output.Image, error) {
+	data, err := fetch(url)
+	if err != nil {
+		return output.Image{}, fmt.Errorf("downloading %s: %w", filename, err)
+	}
+	return shrinkImage(output.Image{
+		Data:        data,
+		ContentType: contentType,
+		Filename:    filename,
+	})
+}
+
 func downloadImages(attachments []*discordgo.MessageAttachment) ([]output.Image, error) {
 	var images []output.Image
 	for _, attachment := range attachments {
 		if !strings.HasPrefix(attachment.ContentType, "image/") {
 			continue
 		}
-		resp, err := http.Get(attachment.URL)
-		if err != nil {
-			return nil, fmt.Errorf("downloading %s: %w", attachment.Filename, err)
-		}
-		data, err := io.ReadAll(resp.Body)
-		resp.Body.Close()
-		if err != nil {
-			return nil, fmt.Errorf("downloading %s: %w", attachment.Filename, err)
-		}
-		if resp.StatusCode != http.StatusOK {
-			return nil, fmt.Errorf("downloading %s: status %s", attachment.Filename, resp.Status)
-		}
-		img, err := shrinkImage(output.Image{
-			Data:        data,
-			ContentType: attachment.ContentType,
-			Filename:    attachment.Filename,
-		})
+		img, err := downloadImage(attachment.URL, attachment.Filename, attachment.ContentType)
 		if err != nil {
 			return nil, err
 		}
@@ -132,6 +160,15 @@ func (dist *distributor) created(event discord.Event) {
 		return
 	}
 
+	var preview *output.Preview
+	if videoURL := findYouTubeURL(event.Content); videoURL != "" {
+		preview, err = youtubePreview(videoURL)
+		if err != nil {
+			// The post is still worth making without its card.
+			fmt.Fprintln(os.Stderr, err)
+		}
+	}
+
 	// A reply to a message we never distributed becomes a top level post.
 	var parents store.Refs
 	if event.ReplyToID != "" {
@@ -143,6 +180,7 @@ func (dist *distributor) created(event discord.Event) {
 		post := output.Post{
 			Text:    event.Content,
 			Images:  images,
+			Preview: preview,
 		}
 		if parent, ok := parents[out.GetName()]; ok && !parent.IsZero() {
 			post.ReplyTo = &parent

output/bluesky.go

@@ -89,7 +89,9 @@ func (bo *blueskyoutput) Write(post Post) (Ref, error) {
 		root = Ref{RootURI: rootref.Uri, RootCID: rootref.Cid}
 	}
 
-	if len(post.Images) > 0 {
+	// Bluesky allows a single embed, so attached images win over a link card.
+	switch {
+	case len(post.Images) > 0:
 		embedimages := make([]*bsky.EmbedImages_Image, 0, len(post.Images))
 		for _, img := range post.Images {
 			blob, err := atproto.RepoUploadBlob(context.TODO(), cli, bytes.NewReader(img.Data))
@@ -106,6 +108,24 @@ func (bo *blueskyoutput) Write(post Post) (Ref, error) {
 				Images: embedimages,
 			},
 		}
+	case post.Preview != nil:
+		external := &bsky.EmbedExternal_External{
+			Uri:         post.Preview.URL,
+			Title:       post.Preview.Title,
+			Description: post.Preview.Description,
+		}
+		if thumb := post.Preview.Thumb; thumb != nil {
+			blob, err := atproto.RepoUploadBlob(context.TODO(), cli, bytes.NewReader(thumb.Data))
+			if err != nil {
+				return Ref{}, fmt.Errorf("uploading %s: %w", thumb.Filename, err)
+			}
+			external.Thumb = blob.Blob
+		}
+		feedpost.Embed = &bsky.FeedPost_Embed{
+			EmbedExternal: &bsky.EmbedExternal{
+				External: external,
+			},
+		}
 	}
 
 	Recordinput := &atproto.RepoCreateRecord_Input{

output/output.go

@@ -6,6 +6,14 @@ type Image struct {
 	Filename    string
 }
 
+// Preview is a link preview card ("プレビュー窓") to attach to a post.
+type Preview struct {
+	URL         string
+	Title       string
+	Description string
+	Thumb       *Image
+}
+
 // Ref identifies a post an output has already published so it can later be
 // replied to or deleted. The fields are output specific; only the ones the
 // publishing output filled in are meaningful to it.
@@ -26,6 +34,7 @@ func (r Ref) IsZero() bool {
 type Post struct {
 	Text    string
 	Images  []Image
+	Preview *Preview
 	// ReplyTo is the Ref this same output returned for the post being
 	// replied to, or nil for a top level post.
 	ReplyTo *Ref

output/stdout.go

@@ -19,6 +19,9 @@ func (so *stdoutput) Write(post Post) (Ref, error) {
 	for _, img := range post.Images {
 		fmt.Printf("[image: %s (%s, %d bytes)]\n", img.Filename, img.ContentType, len(img.Data))
 	}
+	if post.Preview != nil {
+		fmt.Printf("[preview: %s - %s (%s)]\n", post.Preview.Title, post.Preview.Description, post.Preview.URL)
+	}
 	return Ref{ID: post.Text}, nil
 }
 

output/twitter.go

@@ -63,6 +63,8 @@ func (to *twitteroutput) Write(post Post) (Ref, error) {
 			MediaIDs: mediaIDs,
 		}
 	}
+	// A preview needs no special handling here: Twitter builds its own card
+	// from the URL that is already in the text.
 	if post.ReplyTo != nil && post.ReplyTo.ID != "" {
 		p.Reply = &types.CreateInputReply{
 			InReplyToTweetID: post.ReplyTo.ID,

preview.go

@@ -0,0 +1,74 @@
+package main
+
+import (
+	"encoding/json"
+	"fmt"
+	"net/url"
+	"strings"
+	"tweetdistributor/output"
+)
+
+// findYouTubeURL returns the first YouTube video link in content, or "" if
+// there is none.
+func findYouTubeURL(content string) string {
+	for _, match := range urlPattern.FindAllString(content, -1) {
+		raw := trimURL(match)
+		u, err := url.Parse(raw)
+		if err != nil {
+			continue
+		}
+		switch strings.ToLower(u.Hostname()) {
+		case "youtu.be":
+			if strings.Trim(u.Path, "/") != "" {
+				return raw
+			}
+		case "youtube.com", "www.youtube.com", "m.youtube.com", "music.youtube.com":
+			if u.Path == "/watch" && u.Query().Get("v") != "" {
+				return raw
+			}
+			if strings.HasPrefix(u.Path, "/shorts/") || strings.HasPrefix(u.Path, "/live/") {
+				return raw
+			}
+		}
+	}
+	return ""
+}
+
+// oEmbedResponse is the part of YouTube's oEmbed document we care about.
+type oEmbedResponse struct {
+	Title        string `json:"title"`
+	AuthorName   string `json:"author_name"`
+	ThumbnailURL string `json:"thumbnail_url"`
+}
+
+// youtubePreview builds the preview card for a YouTube link by asking
+// YouTube's oEmbed endpoint for the title, channel and thumbnail.
+func youtubePreview(videoURL string) (*output.Preview, error) {
+	endpoint := "https://www.youtube.com/oembed?format=json&url=" + url.QueryEscape(videoURL)
+	body, err := fetch(endpoint)
+	if err != nil {
+		return nil, fmt.Errorf("fetching preview for %s: %w", videoURL, err)
+	}
+
+	var oembed oEmbedResponse
+	if err := json.Unmarshal(body, &oembed); err != nil {
+		return nil, fmt.Errorf("parsing preview for %s: %w", videoURL, err)
+	}
+
+	preview := &output.Preview{
+		URL:         videoURL,
+		Title:       oembed.Title,
+		Description: oembed.AuthorName,
+	}
+
+	if oembed.ThumbnailURL != "" {
+		thumb, err := downloadImage(oembed.ThumbnailURL, "thumbnail.jpg", "image/jpeg")
+		if err != nil {
+			// A card without its thumbnail is still worth posting.
+			return preview, nil
+		}
+		preview.Thumb = &thumb
+	}
+
+	return preview, nil
+}

preview_test.go

@@ -0,0 +1,33 @@
+package main
+
+import "testing"
+
+func TestFindYouTubeURL(t *testing.T) {
+	tests := []struct {
+		name    string
+		content string
+		want    string
+	}{
+		{"watch", "みてみて https://www.youtube.com/watch?v=dQw4w9WgXcQ おもしろい", "https://www.youtube.com/watch?v=dQw4w9WgXcQ"},
+		{"short host", "https://youtu.be/dQw4w9WgXcQ?t=42", "https://youtu.be/dQw4w9WgXcQ?t=42"},
+		{"shorts", "https://www.youtube.com/shorts/abc_123", "https://www.youtube.com/shorts/abc_123"},
+		{"live", "https://youtube.com/live/abc-123", "https://youtube.com/live/abc-123"},
+		{"mobile", "https://m.youtube.com/watch?v=abc&feature=share", "https://m.youtube.com/watch?v=abc&feature=share"},
+		{"trailing punctuation", "これ→https://youtu.be/abc123。", "https://youtu.be/abc123"},
+		{"first of several", "https://youtu.be/one https://youtu.be/two", "https://youtu.be/one"},
+		{"skips other links", "https://example.com/watch?v=x https://youtu.be/abc", "https://youtu.be/abc"},
+		{"no url", "ただのつぶやき", ""},
+		{"other site", "https://example.com/", ""},
+		{"youtube without video", "https://www.youtube.com/", ""},
+		{"watch without v", "https://www.youtube.com/watch?list=PL123", ""},
+		{"lookalike host", "https://youtube.com.evil.example/watch?v=abc", ""},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := findYouTubeURL(tt.content); got != tt.want {
+				t.Errorf("findYouTubeURL(%q) = %q, want %q", tt.content, got, tt.want)
+			}
+		})
+	}
+}