Merge pull request #141 from mgifford/copilot/fix-google-analytics-page-loads

Fix third-party JS size total to use page-load-weighted bytes

Author: mgifford

src/publish/render-pages.js

@@ -1957,6 +1957,7 @@ function renderTechSummarySection(report) {
     third_party_service_counts = {},
     third_party_service_urls = {},
     third_party_service_total_bytes = {},
+    third_party_service_page_load_totals = {},
     accessibility_statement_summary = null,
     required_links_summary = null
   } = summary;
@@ -2008,7 +2009,9 @@ function renderTechSummarySection(report) {
               const totalBytes = third_party_service_total_bytes[name] ?? 0;
               if (totalBytes > 0 && count > 0) {
                 const perPageBytes = Math.round(totalBytes / count);
-                sizeCell = `${formatBytes(perPageBytes)} <span class="tp-size-total">(${formatBytes(totalBytes)} total)</span>`;
+                const pageLoadTotal = third_party_service_page_load_totals[name] ?? 0;
+                const weightedTotal = pageLoadTotal > 0 ? perPageBytes * pageLoadTotal : totalBytes;
+                sizeCell = `${formatBytes(perPageBytes)} <span class="tp-size-total">(${formatBytes(weightedTotal)} total)</span>`;
               } else {
                 sizeCell = '<span class="tp-size-unknown">&#8212;</span>';
               }

src/scanners/tech-detector.js

@@ -434,15 +434,16 @@ export function detectTechnologies(lighthouseRaw) {
  * USWDS. Returns counts and a deduplicated list of observed USWDS versions.
  * Also aggregates third-party service usage and transfer sizes across all scanned URLs.
  *
- * @param {Array<{ scan_status: string, detected_technologies?: object }>} urlResults
+ * @param {Array<{ scan_status: string, detected_technologies?: object, page_load_count?: number }>} urlResults
  * @returns {{
  *   cms_counts: Record<string, number>,
  *   uswds_count: number,
  *   uswds_versions: string[],
  *   total_scanned: number,
  *   third_party_service_counts: Record<string, number>,
  *   third_party_service_urls: Record<string, string[]>,
- *   third_party_service_total_bytes: Record<string, number>
+ *   third_party_service_total_bytes: Record<string, number>,
+ *   third_party_service_page_load_totals: Record<string, number>
  * }}
  */
 export function buildTechSummary(urlResults = []) {
@@ -455,6 +456,7 @@ export function buildTechSummary(urlResults = []) {
   const thirdPartyServiceCounts = {};
   const thirdPartyServiceUrls = {};
   const thirdPartyServiceTotalBytes = {};
+  const thirdPartyServicePageLoadTotals = {};
 
   for (const result of successful) {
     const tech = result.detected_technologies;
@@ -463,6 +465,7 @@ export function buildTechSummary(urlResults = []) {
     }
 
     const url = result.url ?? null;
+    const pageLoadCount = result.page_load_count ?? 0;
 
     if (tech.cms) {
       cmsCounts[tech.cms] = (cmsCounts[tech.cms] ?? 0) + 1;
@@ -490,6 +493,8 @@ export function buildTechSummary(urlResults = []) {
         if (!thirdPartyServiceUrls[serviceName]) thirdPartyServiceUrls[serviceName] = [];
         thirdPartyServiceUrls[serviceName].push(url);
       }
+      thirdPartyServicePageLoadTotals[serviceName] =
+        (thirdPartyServicePageLoadTotals[serviceName] ?? 0) + pageLoadCount;
     }
 
     const sizes = tech.third_party_service_sizes ?? {};
@@ -510,6 +515,7 @@ export function buildTechSummary(urlResults = []) {
     total_scanned: successful.length,
     third_party_service_counts: thirdPartyServiceCounts,
     third_party_service_urls: thirdPartyServiceUrls,
-    third_party_service_total_bytes: thirdPartyServiceTotalBytes
+    third_party_service_total_bytes: thirdPartyServiceTotalBytes,
+    third_party_service_page_load_totals: thirdPartyServicePageLoadTotals
   };
 }

tests/unit/tech-detector.test.js

@@ -590,3 +590,80 @@ test('buildTechSummary handles results with no third_party_service_sizes field',
   assert.equal(summary.third_party_service_counts['Google Analytics'], 1);
   assert.deepEqual(summary.third_party_service_total_bytes, {});
 });
+
+// ---------------------------------------------------------------------------
+// third_party_service_page_load_totals – buildTechSummary
+// ---------------------------------------------------------------------------
+
+test('buildTechSummary returns empty third_party_service_page_load_totals for empty results', () => {
+  const summary = buildTechSummary([]);
+  assert.deepEqual(summary.third_party_service_page_load_totals, {});
+});
+
+test('buildTechSummary accumulates page_load_count per service across successful results', () => {
+  const results = [
+    {
+      url: 'https://site1.gov/',
+      page_load_count: 5000000,
+      scan_status: 'success',
+      detected_technologies: {
+        cms: null,
+        uswds: { detected: false, version: null },
+        third_party_services: ['Google Analytics', 'Google Fonts'],
+        third_party_service_sizes: { 'Google Analytics': 30720, 'Google Fonts': 5120 }
+      }
+    },
+    {
+      url: 'https://site2.gov/',
+      page_load_count: 2000000,
+      scan_status: 'success',
+      detected_technologies: {
+        cms: null,
+        uswds: { detected: false, version: null },
+        third_party_services: ['Google Analytics'],
+        third_party_service_sizes: { 'Google Analytics': 20480 }
+      }
+    },
+    {
+      url: 'https://site3.gov/',
+      page_load_count: 999999,
+      scan_status: 'failed',
+      detected_technologies: {
+        cms: null,
+        uswds: { detected: false, version: null },
+        third_party_services: ['Google Analytics'],
+        third_party_service_sizes: { 'Google Analytics': 99999 }
+      }
+    }
+  ];
+  const summary = buildTechSummary(results);
+  assert.equal(summary.third_party_service_page_load_totals['Google Analytics'], 7000000);
+  assert.equal(summary.third_party_service_page_load_totals['Google Fonts'], 5000000);
+  assert.equal(summary.third_party_service_page_load_totals['YouTube'], undefined);
+});
+
+test('buildTechSummary treats missing page_load_count as 0 in page load totals', () => {
+  const results = [
+    {
+      url: 'https://site1.gov/',
+      scan_status: 'success',
+      detected_technologies: {
+        cms: null,
+        uswds: { detected: false, version: null },
+        third_party_services: ['Google Analytics']
+      }
+    },
+    {
+      url: 'https://site2.gov/',
+      page_load_count: 3000000,
+      scan_status: 'success',
+      detected_technologies: {
+        cms: null,
+        uswds: { detected: false, version: null },
+        third_party_services: ['Google Analytics']
+      }
+    }
+  ];
+  const summary = buildTechSummary(results);
+  assert.equal(summary.third_party_service_page_load_totals['Google Analytics'], 3000000);
+});