apptvty 0.3.2 → 0.4.1

package/dist/middleware/nextjs.js

@@ -1205,7 +1205,8 @@ function withApptvty(config, next) {
     const crawlerInfo = detectCrawler(userAgent);
     const scraperService = detectScraperService(userAgent);
     const aiCrawlerParam = parseBoolParam(request.nextUrl.searchParams.get("ai_crawler"), false);
-    const isCrawler = crawlerInfo.isAi || aiCrawlerParam || scraperService.isScraperService;
+    const isAi = crawlerInfo.isAi || aiCrawlerParam;
+    const isScraper = isAi || scraperService.isScraperService || crawlerInfo.name === "unknown_bot";
     if (request.nextUrl.pathname === "/api/apptvty/verify") {
       const challenge = request.nextUrl.searchParams.get("challenge");
       if (challenge) {
@@ -1250,11 +1251,12 @@ function withApptvty(config, next) {
       ip_address: getClientIp(headers),
       user_agent: userAgent,
       referrer: request.headers.get("referer"),
-      is_ai_crawler: crawlerInfo.isAi,
+      is_ai_crawler: isAi,
       crawler_type: crawlerInfo.name,
       crawler_organization: crawlerInfo.organization,
       confidence_score: crawlerInfo.confidence,
-      scraper_service: scraperService.name
+      scraper_service: scraperService.name,
+      attribution_id: request.nextUrl.searchParams.get("atid")
     };
     const isInternalRequest = request.headers.get("x-apptvty-internal") === "true";
     if (!isInternalRequest && !pathname.startsWith(queryPath)) {
@@ -1263,46 +1265,66 @@ function withApptvty(config, next) {
         event.waitUntil(logger.flush());
       }
     }
-    if (isCrawler && !isInternalRequest && !pathname.startsWith(queryPath)) {
+    if (!isInternalRequest && !pathname.startsWith(queryPath) && response.status === 200) {
       try {
-        const proxyReq = new Request(request.url, {
-          headers: new Headers(request.headers)
-        });
-        proxyReq.headers.set("x-apptvty-internal", "true");
-        const res = await fetch(proxyReq);
-        const contentType = res.headers.get("content-type") ?? "";
-        if (contentType.includes("text/html")) {
-          const html = await res.text();
-          let markdown = convertHtmlToMarkdown(html);
-          const pageAds = await client.getAdsForPage({ site_id: config.siteId, page_path: pathname });
-          if (pageAds.ads && pageAds.ads.length > 0) {
-            const ad = pageAds.ads[0];
-            markdown += `
+        const pageAds = await client.getAdsForPage({ site_id: config.siteId, page_path: pathname });
+        if (pageAds.ads && pageAds.ads.length > 0) {
+          const ad = pageAds.ads[0];
+          client.logImpression({
+            impression_id: ad.impression_id,
+            site_id: config.siteId,
+            page_path: pathname,
+            agent_ua: userAgent,
+            agent_ip: getClientIp(headers),
+            timestamp: (/* @__PURE__ */ new Date()).toISOString()
+          }).catch(() => {
+          });
+          if (isScraper) {
+            const proxyReq = new Request(request.url, { headers: new Headers(request.headers) });
+            proxyReq.headers.set("x-apptvty-internal", "true");
+            const res = await fetch(proxyReq);
+            const contentType = res.headers.get("content-type") ?? "";
+            if (contentType.includes("text/html")) {
+              const html = await res.text();
+              if (isAi || scraperService.isScraperService) {
+                let markdown = convertHtmlToMarkdown(html);
+                markdown += `
 
 ---
 > **Sponsored:** [${ad.text}](${ad.url}) - ${ad.advertiser}
 `;
-            client.logImpression({
-              impression_id: ad.impression_id,
-              site_id: config.siteId,
-              page_path: pathname,
-              agent_ua: userAgent,
-              agent_ip: getClientIp(headers),
-              timestamp: (/* @__PURE__ */ new Date()).toISOString()
-            }).catch(() => {
-            });
-          }
-          return new import_server.NextResponse(markdown, {
-            status: res.status,
-            headers: {
-              "Content-Type": "text/markdown",
-              "X-Apptvty-AEO": "true"
+                return new import_server.NextResponse(markdown, {
+                  status: res.status,
+                  headers: {
+                    "Content-Type": "text/markdown",
+                    "X-Apptvty-AEO": "true",
+                    "X-Sponsored-Content": `${ad.text}; url=${ad.url}`
+                  }
+                });
+              }
+              const jsonLd = `
+<script type="application/ld+json">{"@context":"https://schema.org","@type":"CreativeWork","author":{"@type":"Organization","name":"${ad.advertiser}"},"mainEntityOfPage":{"@type":"WebPage","@id":"${ad.url}"},"headline":"Sponsored: ${ad.text}"}</script>
+`;
+              const stealthDiv = `
+<div style="display:none !important;visibility:hidden;height:0;width:0;overflow:hidden;" aria-hidden="true" data-apptvty-ad="${ad.impression_id}">Sponsored by ${ad.advertiser}: <a href="${ad.url}">${ad.text}</a></div>
+`;
+              let modifiedHtml = html;
+              if (html.includes("</head>")) modifiedHtml = modifiedHtml.replace("</head>", `${jsonLd}</head>`);
+              if (modifiedHtml.includes("</body>")) modifiedHtml = modifiedHtml.replace("</body>", `${stealthDiv}</body>`);
+              else modifiedHtml += stealthDiv;
+              return new import_server.NextResponse(modifiedHtml, {
+                status: res.status,
+                headers: {
+                  ...headersToRecord(res.headers),
+                  "X-Sponsored-Content": `${ad.text}; url=${ad.url}`
+                }
+              });
             }
-          });
+          }
+          response.headers.set("X-Sponsored-Content", `${ad.text}; url=${ad.url}`);
         }
-        return res;
       } catch (err) {
-        if (config.debug) console.warn("[apptvty] Markdown proxy failed:", err);
+        if (config.debug) console.warn("[apptvty] Stealth injection failed:", err);
       }
     }
     return response;