From bead7fdf43849ea3e343b2c92ec9775be489db80 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E6=B5=85=E5=BF=B5?= <email@sample.com>
Date: Fri, 14 Aug 2026 09:25:10 +0800
Subject: [PATCH] =?UTF-8?q?fix(scraper):=20=E9=A1=B5=E9=9D=A2=E6=8A=93?=
 =?UTF-8?q?=E5=8F=96=E5=99=A8=E6=94=AF=E6=8C=81=E5=91=98=E5=B7=A5=E5=88=97?=
 =?UTF-8?q?=E8=A1=A8=E7=BF=BB=E9=A1=B5(=E4=B9=8B=E5=89=8D=E5=8F=AA?=
 =?UTF-8?q?=E6=8A=93=E7=AC=AC=E4=B8=80=E9=A1=B510=E4=B8=AA)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 yunke-server/scripts/scrape_via_page.py | 39 ++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 4 deletions(-)

diff --git a/yunke-server/scripts/scrape_via_page.py b/yunke-server/scripts/scrape_via_page.py
index ef16181..7636b73 100644
--- a/yunke-server/scripts/scrape_via_page.py
+++ b/yunke-server/scripts/scrape_via_page.py
@@ -209,11 +209,42 @@ class PageScraper:
             return result;
         }""")
 
-        # Also paginate through staff list if needed
-        # (for now, first page of 50 should cover most)
+        # Paginate through ALL staff list pages
+        all_employees = list(employees)
+        page = 1
+        while await self.click_next_page():
+            page += 1
+            await self.page.wait_for_timeout(2000)
+            more = await self.page.evaluate("""() => {
+                const rows = document.querySelectorAll('.ant-table-row');
+                const result = [];
+                for (const tr of rows) {
+                    const friendLink = tr.querySelector('a[href*="wechatdetail/1/"]');
+                    const groupLink = tr.querySelector('a[href*="wechatdetail/2/"]');
+                    if (friendLink) {
+                        const href = friendLink.getAttribute('href');
+                        const wxId = href.match(/wechatdetail\\/1\\/([^?]+)/)?.[1];
+                        const friendCount = parseInt(friendLink.textContent) || 0;
+                        const groupHref = groupLink?.getAttribute('href');
+                        const groupCount = parseInt(groupLink?.textContent) || 0;
+                        result.push({ wxId, friendCount, groupCount, friendHref: href, groupHref });
+                    }
+                }
+                return result;
+            }""")
+            all_employees.extend(more)
+            logger.info(f"  staff page {page}: +{len(more)} employees (total: {len(all_employees)})")
 
-        logger.info(f"Found {len(employees)} employees with friend links")
-        return employees
+        # Dedup by wxId
+        seen = set()
+        deduped = []
+        for emp in all_employees:
+            if emp["wxId"] not in seen:
+                seen.add(emp["wxId"])
+                deduped.append(emp)
+
+        logger.info(f"Found {len(deduped)} unique employees with friend links ({page} pages)")
+        return deduped
 
 
 async def scrape_friends(cookie_str: str, max_employees: int = 100):