Initial commit

2026-05-22 17:25:27 -07:00
commit 652d6dfa38
40 changed files with 5855 additions and 0 deletions
--- a/internal/ingestion/og_test.go
+++ b/internal/ingestion/og_test.go
@@ -0,0 +1,72 @@
+package ingestion
+
+import (
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+)
+
+func TestFetchOGImage(t *testing.T) {
+	cases := []struct {
+		name string
+		body string
+		want string // suffix match
+	}{
+		{
+			name: "og:image property",
+			body: `<html><head><meta property="og:image" content="https://example.com/lead.jpg"></head></html>`,
+			want: "https://example.com/lead.jpg",
+		},
+		{
+			name: "content before property",
+			body: `<html><head><meta content="https://example.com/swap.jpg" property="og:image"/></head></html>`,
+			want: "https://example.com/swap.jpg",
+		},
+		{
+			name: "twitter:image fallback",
+			body: `<html><head><meta name="twitter:image" content="https://example.com/twit.jpg"></head></html>`,
+			want: "https://example.com/twit.jpg",
+		},
+		{
+			name: "relative URL resolved",
+			body: `<html><head><meta property="og:image" content="/img/lead.jpg"></head></html>`,
+			want: "/img/lead.jpg",
+		},
+		{
+			name: "no og tags",
+			body: `<html><head><title>nothing</title></head></html>`,
+			want: "",
+		},
+	}
+	for _, c := range cases {
+		t.Run(c.name, func(t *testing.T) {
+			srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+				w.Header().Set("Content-Type", "text/html")
+				w.Write([]byte(c.body))
+			}))
+			defer srv.Close()
+
+			got := FetchOGImage(srv.URL + "/article")
+			if c.want == "" {
+				if got != "" {
+					t.Errorf("expected empty, got %q", got)
+				}
+				return
+			}
+			if !strings.HasSuffix(got, c.want) {
+				t.Errorf("got %q, want suffix %q", got, c.want)
+			}
+		})
+	}
+}
+
+func TestFirstImgSrc(t *testing.T) {
+	in := `<p>Body <img alt="x" src="https://cdn.example.com/lead.png" width="600"/> more</p>`
+	if got := firstImgSrc(in); got != "https://cdn.example.com/lead.png" {
+		t.Errorf("got %q", got)
+	}
+	if got := firstImgSrc(""); got != "" {
+		t.Errorf("empty input should yield empty, got %q", got)
+	}
+}