From 591f006a5356fdb5586d9613e67987a84b63d659 Mon Sep 17 00:00:00 2001
From: Nubenetes Bot <bot@nubenetes.com>
Date: Sun, 10 May 2026 22:22:11 +0200
Subject: [PATCH] fix: ensure PR creation with technical commit and deepen
 extraction scroll

---
 src/gitops_manager.py   | 14 +++++++++++++-
 src/ingestion_twikit.py |  8 ++++----
 2 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/src/gitops_manager.py b/src/gitops_manager.py
index c0f67e0c..4514bc74 100644
--- a/src/gitops_manager.py
+++ b/src/gitops_manager.py
@@ -16,6 +16,14 @@ class RepositoryController:
         branch_name = f"bot/knowledge-update-{timestamp_slug}"
         self._create_feature_branch(branch_name)
 
+        # Si no hay cambios en Markdowns, añadimos un cambio técnico para permitir abrir la PR con el reporte
+        if not updates:
+            updates["src/memory/last_audit_run.json"] = json.dumps({
+                "timestamp": metrics.get("end_date"),
+                "total_extracted": metrics.get("total_extracted"),
+                "status": "Audit Only (No new links injected)"
+            }, indent=2)
+
         for file_path, content in updates.items():
             try:
                 commit_signature = f"chore: update {file_path} [{timestamp_slug}]"
@@ -38,15 +46,19 @@ class RepositoryController:
         full_report = metrics.get('full_report', [])
         
         # 1. Tabla Matricial de Auditoría
+        # Limitamos la tabla si es muy larga para evitar errores de API de GitHub
         matrix_table = "### 📋 Matriz de Auditoría de Enlaces (Full Extraction)\n"
         matrix_table += "| Estado | Motivo | Categoría | URL |\n| :--- | :--- | :--- | :--- |\n"
         
         counts = {"INCLUDED": 0, "DUPLICATE": 0, "FILTERED": 0}
-        for item in full_report:
+        for item in full_report[:200]: # Mostrar solo los primeros 200 para no romper el límite de caracteres del PR
             status_emoji = {"INCLUDED": "✅", "DUPLICATE": "👯", "FILTERED": "🛡️"}.get(item['status'], "❓")
             matrix_table += f"| {status_emoji} {item['status']} | {item['reason']} | `{item['category']}` | {item['url']} |\n"
             counts[item['status']] = counts.get(item['status'], 0) + 1
 
+        if len(full_report) > 200:
+            matrix_table += f"\n> 💡 *... y {len(full_report) - 200} enlaces más procesados.*"
+
         # 2. Diagrama Mermaid
         mermaid_pie = "### 📊 Métricas de Decisión\n```mermaid\npie title Distribución de Decisión Agéntica\n"
         mermaid_pie += f"    \"Aceptados (Inyectados)\" : {counts['INCLUDED']}\n"
diff --git a/src/ingestion_twikit.py b/src/ingestion_twikit.py
index 7dee236d..c50c2f2b 100644
--- a/src/ingestion_twikit.py
+++ b/src/ingestion_twikit.py
@@ -71,14 +71,14 @@ class SocialDataExtractor:
                 await page.goto(f"https://x.com/{self.target_account}", wait_until="domcontentloaded", timeout=60000)
                 await asyncio.sleep(8)
                 
-                for _ in range(4): # Scroll moderado
+                for _ in range(10): # Scroll profundo para histórico
                     html = await page.content()
                     urls = self._extract_urls_from_text(html)
                     for u in urls:
-                        if all(x not in u for x in ["x.com", "twitter.com", "t.co", "abs.twimg", "archive.org"]):
+                        if all(x not in u for x in ["x.com", "twitter.com", "t.co", "abs.twimg", "archive.org", "pbs.twimg"]):
                             results.append({"url": u, "context": "Playwright Browser", "timestamp": datetime.now(MADRID_TZ).isoformat()})
-                    await page.evaluate("window.scrollBy(0, 1200)")
-                    await asyncio.sleep(4)
+                    await page.evaluate("window.scrollBy(0, 2000)")
+                    await asyncio.sleep(5)
                 
                 await browser.close()
                 return results