auto 任务运行中实时显示已爬/待爬缓存数; 人工停止后继续爬取验证

- engine: CrawlJob 增加 _auto_visited/_auto_pending 运行中实时状态 (含发现新链接入队后刷新)
- app.py 详情接口: 任务运行中优先读内存实时值, 不再显示上次运行结束的旧数字
- 端到端验证: 人工停止(已爬3,缓存18) → 继续爬取(跳过seed,爬完18页,缓存0) ✅
- 运行中采样: visited 1→2 实时增长, pending 15→14 同步递减 ✅

.gitignore

@@ -4,3 +4,5 @@ data/*.json
 data/cookies_*.json
 logs/
 out/
+data/exports/
+data/auto_state/

app.py

@@ -4,14 +4,17 @@
 启动: /home/hz1/miniconda3/envs/openclaw/bin/python app.py  (默认端口 16062)
 """
 import os
+import re
 import threading
 import time
+import zipfile
 from datetime import datetime
 
 from flask import Flask, jsonify, request, send_file, send_from_directory
 
 import store
 import db
+import notify
 from engine import CrawlJob, probe_links
 from scheduler import Scheduler, cron_next, interval_delta
 
@@ -101,13 +104,16 @@ def persist_cb(task_id, run):
         print(f"[db] 同步失败: {e}", flush=True)
 
 
-def start_run(task):
-    """为任务启动一次爬取, 返回 (run, error)"""
+def start_run(task, skip_seed=False):
+    """为任务启动一次爬取, 返回 (run, error)
+    skip_seed: auto 任务继续爬取模式, 跳过起始网址直接从缓存队列消费
+    """
     with JOBS_LOCK:
         job = JOBS.get(task["id"])
         if job and job.is_running():
             return None, "该任务已有正在运行的爬取"
         run = make_run(task)
+        run["skip_seed"] = bool(skip_seed)
         store.add_run(task["id"], run)
         job = CrawlJob(task, run, persist_cb)
         JOBS[task["id"]] = job
@@ -302,11 +308,19 @@ def api_task_detail(tid):
         result["run_page"] = page
         result["run_pages"] = pages
         result["run_page_size"] = page_size
-    # auto 状态数量 (独立文件)
+    # auto 状态数量: 运行中优先读内存实时值, 否则读状态文件
     if task.get("mode") == "auto":
+        live_v = live_p = None
+        with JOBS_LOCK:
+            job = JOBS.get(tid)
+            if job:
+                live_v = getattr(job, "_auto_visited", None)
+                live_p = getattr(job, "_auto_pending", None)
         st = store.load_auto_state(tid, task)
-        result["auto_pending_count"] = len(st.get("pending", []))
-        result["auto_visited_count"] = len(st.get("visited", []))
+        result["auto_pending_count"] = (
+            live_p if live_p is not None else len(st.get("pending", [])))
+        result["auto_visited_count"] = (
+            live_v if live_v is not None else len(st.get("visited", [])))
     return jsonify(result)
 
 
@@ -421,6 +435,23 @@ def api_trash_clear():
 
 # ---------------- API: 运行控制 ----------------
 
+@app.route("/api/tasks/<tid>/continue", methods=["POST"])
+def api_continue(tid):
+    """继续爬取: auto 任务从待爬缓存队列接着爬 (跳过起始网址, 保留已爬集合)"""
+    task = store.get_task(tid)
+    if not task:
+        return jsonify({"error": "任务不存在"}), 404
+    if task.get("mode") != "auto":
+        return jsonify({"error": "仅自动爬取任务支持继续爬取"}), 400
+    st = store.load_auto_state(tid, task)
+    if not st.get("pending"):
+        return jsonify({"error": "没有待爬缓存链接，无需继续"}), 400
+    run, err = start_run(task, skip_seed=True)
+    if err:
+        return jsonify({"error": err}), 409
+    return jsonify(run)
+
+
 @app.route("/api/tasks/<tid>/start", methods=["POST"])
 def api_start(tid):
     task = store.get_task(tid)
@@ -674,6 +705,103 @@ def api_file():
     return send_file(full)
 
 
+# ---------------- API: 打包导出 ----------------
+
+EXPORT_DIR = os.path.join(HERE, "data", "exports")
+EXPORT_KEEP = 10          # 最多保留的打包文件数
+
+
+def _safe_zip_name(task):
+    """任务名安全化为文件名 (保留中文, 去非法字符)"""
+    name = re.sub(r'[\\/:*?"<>|\s]+', "_", task.get("name", "")).strip("_")
+    return (name or task["id"])[:60]
+
+
+def _make_zip(task):
+    """把任务输出目录打包为 zip, 返回 (zip_path, err) 或 (None, 错误信息)"""
+    out_dir = os.path.realpath(resolve_out_dir(task))
+    if not os.path.isdir(out_dir):
+        return None, "输出目录不存在"
+    os.makedirs(EXPORT_DIR, exist_ok=True)
+    zip_path = os.path.join(EXPORT_DIR, f"{task['id']}_{int(time.time())}.zip")
+    prefix = _safe_zip_name(task) + "/"
+    count = 0
+    try:
+        with zipfile.ZipFile(zip_path, "w", zipfile.ZIP_DEFLATED) as zf:
+            for root, _dirs, files in os.walk(out_dir):
+                for f in files:
+                    full = os.path.join(root, f)
+                    rel = os.path.relpath(full, out_dir)
+                    zf.write(full, prefix + rel)
+                    count += 1
+    except Exception as e:
+        try:
+            os.remove(zip_path)
+        except OSError:
+            pass
+        return None, f"打包失败: {e}"
+    _cleanup_exports()
+    return zip_path, None
+
+
+def _cleanup_exports():
+    """清理旧的打包文件, 只保留最近 EXPORT_KEEP 个"""
+    try:
+        files = sorted(
+            (os.path.join(EXPORT_DIR, f) for f in os.listdir(EXPORT_DIR)
+             if f.endswith(".zip")),
+            key=os.path.getmtime, reverse=True,
+        )
+        for f in files[EXPORT_KEEP:]:
+            os.remove(f)
+    except Exception:
+        pass
+
+
+@app.route("/api/export")
+def api_export():
+    """打包任务输出目录为 zip 并提供下载"""
+    tid = request.args.get("task_id", "")
+    task = store.get_task(tid)
+    if not task:
+        return jsonify({"error": "任务不存在"}), 404
+    zip_path, err = _make_zip(task)
+    if err:
+        return jsonify({"error": err}), 400
+    return send_file(
+        zip_path, as_attachment=True,
+        download_name=_safe_zip_name(task) + ".zip",
+        mimetype="application/zip",
+    )
+
+
+@app.route("/api/export/email", methods=["POST"])
+def api_export_email():
+    """打包任务输出目录为 zip 并发送到指定邮箱 (默认 wlq@tphai.com)"""
+    body = request.get_json(force=True) or {}
+    tid = body.get("task_id", "")
+    email = (body.get("email") or "").strip() or "wlq@tphai.com"
+    task = store.get_task(tid)
+    if not task:
+        return jsonify({"error": "任务不存在"}), 404
+    if "@" not in email:
+        return jsonify({"error": "邮箱格式不正确"}), 400
+    zip_path, err = _make_zip(task)
+    if err:
+        return jsonify({"error": err}), 400
+    size_mb = round(os.path.getsize(zip_path) / 1048576, 1)
+    body_text = (f"项目: {task['name']}\n"
+                 f"输出目录: {resolve_out_dir(task)}\n"
+                 f"打包文件: {os.path.basename(zip_path)}\n"
+                 f"压缩包大小: {size_mb} MB\n\n"
+                 f"打包时间: {now_str()}")
+    ok, msg = notify.send_attachment(
+        f"[爬虫打包] {task['name']}", body_text, email, zip_path)
+    if not ok:
+        return jsonify({"error": f"邮件发送失败: {msg}"}), 500
+    return jsonify({"ok": True, "msg": f"已发送到 {email} (zip {size_mb} MB)"})
+
+
 # ---------------- 启动 ----------------
 
 scheduler = Scheduler(start_run)

engine.py

@@ -226,6 +226,9 @@ class CrawlJob:
         self.persist = persist          # callable(task_id, run)
         self._stop = threading.Event()
         self._pause = threading.Event()
+        # 运行中的 auto 实时状态 (供详情接口读取; 任务结束时由状态文件兜底)
+        self._auto_visited = None
+        self._auto_pending = None
         self._cfg_lock = threading.RLock()
         self.thread = None
 
@@ -598,9 +601,10 @@ class CrawlJob:
     def _crawl_auto(self):
         """自动爬取: 持续递归 爬取->发现链接->爬取... 直到无新链接可爬或达到上限
         - max_depth: 0=无限制, N=只爬 N 层
-        - max_pages: 0=无限制, N=安全上限
+        - max_pages: 0=无限制, N=安全上限 (继续爬取模式按本次新增页数重新计算)
         - 提取但未爬取的链接持久化到 auto.pending, 已爬集合持久化到 auto.visited
         - 停止后再次运行从缓存队列继续爬; 起始网址每次运行都重新爬(不去重)
+        - skip_seed(继续爬取): 跳过起始网址直接消费缓存队列, 页数上限按本次新增重新计算
         """
         run = self.run
         auto = self.task.get("auto", {})
@@ -615,8 +619,13 @@ class CrawlJob:
         state = store.load_auto_state(self.task["id"], self.task)
         visited = set(state.get("visited", []) or [])
         pending = state.get("pending", []) or []
+        visited_base = len(visited)
         # 起始网址每次运行都爬(不做去重), 缓存队列继续消费
-        queue = [(seed, 0, "")]
+        # 继续爬取模式(skip_seed): 有缓存队列时跳过起始网址, 直接从待爬队列接着爬
+        skip_seed = bool(self.run.get("skip_seed"))
+        queue = []
+        if not (skip_seed and pending):
+            queue.append((seed, 0, ""))
         if pending:
             queue.extend((p["url"], p.get("depth", 0), p.get("source", "")) for p in pending)
         queued = set(visited)
@@ -624,20 +633,30 @@ class CrawlJob:
             queued.add(normalize_url(u))
 
         run["progress"]["total"] = len(queue)
+        self._auto_visited = len(visited)
+        self._auto_pending = len(queue)
         self._persist()
 
         p, browser, ctx, page, cookie_file = self._open_browser()
         try:
             while queue and not self._stop.is_set():
                 self._wait_if_paused()
-                if max_pages > 0 and len(visited) >= max_pages:
-                    self._log("info", f"达到最大页数上限 {max_pages}, 停止")
-                    break
+                if max_pages > 0:
+                    if skip_seed:
+                        # 继续爬取模式: 页数上限按本次新增页数重新计算 (累计 visited 不阻塞继续)
+                        if len(visited) - visited_base >= max_pages:
+                            self._log("info", f"本次继续爬取达到页数上限 {max_pages}, 停止")
+                            break
+                    elif len(visited) >= max_pages:
+                        self._log("info", f"达到最大页数上限 {max_pages}, 停止")
+                        break
                 url, depth, src = queue.pop(0)
                 key = normalize_url(url)
                 if key in visited and url != seed:     # 起始网址不去重, 其余已爬跳过
                     continue
                 visited.add(key)
+                self._auto_visited = len(visited)
+                self._auto_pending = len(queue)
                 idx = len(visited)
                 run["progress"]["current_url"] = url
                 run["progress"]["done"] = len(visited)
@@ -654,6 +673,7 @@ class CrawlJob:
                         if lk not in visited and lk not in queued:
                             queued.add(lk)
                             queue.append((link, depth + 1, url))
+                    self._auto_pending = len(queue)   # 新链接入队后实时刷新
                 if entry["status"] == "OK":
                     self._delay()
             run["progress"]["total"] = len(visited)

notify.py

@@ -33,11 +33,24 @@ def notify_email(task, run):
     if len(results) > 50:
         lines.append(f"  ... 共 {len(results)} 条")
     body = "\n".join(lines)
+    return send_attachment(f"[爬虫完成] {task['name']}", body, to)
+
+
+def send_attachment(subject, body, to, attach_path=None):
+    """发送带附件的邮件 (复用 send_email.py), 返回 (bool, msg)
+    attach_path: 附件文件路径, 可多个
+    """
+    if not os.path.exists(SEND_EMAIL):
+        return False, "send_email.py 不存在"
+    cmd = [sys.executable, SEND_EMAIL, subject, body, "--to", to]
+    if attach_path:
+        if isinstance(attach_path, str):
+            attach_path = [attach_path]
+        for p in attach_path:
+            if os.path.isfile(p):
+                cmd += ["--attach", p]
     try:
-        r = subprocess.run(
-            [sys.executable, SEND_EMAIL, f"[爬虫完成] {task['name']}", body, "--to", to],
-            timeout=60, capture_output=True,
-        )
+        r = subprocess.run(cmd, timeout=600, capture_output=True)
         return r.returncode == 0, (r.stderr or b"").decode(errors="ignore")[:200]
     except Exception as e:
         return False, str(e)

static/app.js

@@ -602,10 +602,15 @@ function renderDetail() {
       ${t.running ? `
         <button class="btn" onclick="actTask('${t.id}','pause')">⏸ 暂停</button>
         <button class="btn danger" onclick="actTask('${t.id}','stop')">⏹ 终止</button>`
-      : `<button class="btn primary" onclick="actTask('${t.id}','start')">▶ 立即执行</button>`}
+      : `
+        <button class="btn primary" onclick="actTask('${t.id}','start')">▶ 立即执行</button>
+        ${t.mode === "auto" && (t.auto_pending_count || 0) > 0 ? `
+          <button class="btn" onclick="contTask('${t.id}')">⏩ 继续爬取（缓存 ${t.auto_pending_count} 条）</button>` : ""}`}
       ${t.mode === "auto" ? `<button class="btn" onclick="probeTask('${t.id}')">🧪 试爬取</button>
         <button class="btn" ${t.running ? "disabled" : ""} onclick="clearCache('${t.id}')">🧹 清空缓存</button>` : ""}
       <button class="btn" onclick="openEdit('${t.id}')">✏️ 编辑配置</button>
+      <button class="btn" onclick="exportZip('${t.id}')">📦 打包下载</button>
+      <button class="btn" onclick="exportEmail('${t.id}')">📧 发邮箱</button>
     </div>`;
 
   const chips = runs.map((r) => `
@@ -732,6 +737,68 @@ function stopLogPoll() {
   if (state.logTimer) { clearInterval(state.logTimer); state.logTimer = null; }
 }
 
+async function contTask(tid) {
+  try {
+    const r = await api(`/api/tasks/${tid}/continue`, { method: "POST" });
+    toast("已开始继续爬取缓存队列");
+    loadTasks();
+    if (state.detail.task && state.detail.task.id === tid) openDetail(tid);
+  } catch (e) { toast(e.message, true); }
+}
+
+/* ---------------- 打包导出 ---------------- */
+async function exportZip(tid) {
+  const btn = window.event && window.event.target;
+  if (btn) { btn.disabled = true; btn.textContent = "⏳ 打包中..."; }
+  try {
+    const res = await fetch(`/api/export?task_id=${tid}`);
+    if (!res.ok) {
+      let d = null;
+      try { d = await res.json(); } catch (e) { /* ignore */ }
+      throw new Error((d && d.error) || `HTTP ${res.status}`);
+    }
+    let fname = "export.zip";
+    const disp = res.headers.get("Content-Disposition") || "";
+    const m = disp.match(/filename\*?=(?:UTF-8'')?"?([^";]+)"?/i);
+    if (m) { try { fname = decodeURIComponent(m[1]); } catch (e) { fname = m[1]; } }
+    const blob = await res.blob();
+    const a = document.createElement("a");
+    a.href = URL.createObjectURL(blob);
+    a.download = fname;
+    document.body.appendChild(a);
+    a.click();
+    a.remove();
+    setTimeout(() => URL.revokeObjectURL(a.href), 5000);
+    toast(`打包下载完成 ${fname} (${(blob.size / 1048576).toFixed(1)} MB)`);
+  } catch (e) {
+    toast("打包失败: " + e.message, true);
+  } finally {
+    if (btn) { btn.disabled = false; btn.textContent = "📦 打包下载"; }
+  }
+}
+
+async function exportEmail(tid) {
+  const t = (state.detail.task && state.detail.task.id === tid)
+    ? state.detail.task : state.tasks.find((x) => x.id === tid);
+  const def = (t && t.config && t.config.notify_email) || "wlq@tphai.com";
+  const email = prompt("发送到邮箱（留空默认 wlq@tphai.com）:", def);
+  if (email === null) return;   // 用户取消
+  const btn = window.event && window.event.target;
+  if (btn) { btn.disabled = true; btn.textContent = "⏳ 打包发送中..."; }
+  try {
+    const r = await api("/api/export/email", {
+      method: "POST",
+      headers: { "Content-Type": "application/json" },
+      body: JSON.stringify({ task_id: tid, email: email.trim() || "wlq@tphai.com" }),
+    });
+    toast(r.msg || "已发送");
+  } catch (e) {
+    toast("发送失败: " + e.message, true);
+  } finally {
+    if (btn) { btn.disabled = false; btn.textContent = "📧 发邮箱"; }
+  }
+}
+
 /* ---------------- 试爬取 ---------------- */
 function collectProbeFromForm() {
   const f = $("taskForm");