fix: accept invoice output args in vision run

scripts/invoice_tool.py

@@ -865,6 +865,138 @@ def vision_detect_list_page() -> tuple[bool, str]:
     return parsed.get("page") == "list", parsed.get("summary", "")
 
 
+def vision_detect_page() -> dict:
+    """截图 + 视觉判断当前页面类型,用于 XML 文本未稳定时兜底。"""
+    out = call_vision_subcommand(["detect-page"])
+    content = out.get("content", "")
+    try:
+        parsed = json.loads(content)
+    except Exception:
+        match = re.search(r"\{.*\}", content, re.S)
+        if not match:
+            return {"page": "other", "summary": content[:120], "raw": content}
+        try:
+            parsed = json.loads(match.group(0))
+        except Exception:
+            return {"page": "other", "summary": content[:120], "raw": content}
+    if not isinstance(parsed, dict):
+        return {"page": "other", "summary": str(parsed)[:120]}
+    return parsed
+
+
+def xml_page_kind(nodes: list[dict[str, str]] | None = None) -> str:
+    """用 uiautomator 文本快速判断页面,比视觉便宜,作为主判据。"""
+    if nodes is None:
+        nodes = adb_dump_nodes()
+    page_text = "\n".join(adb_visible_texts(nodes))
+    if (
+        "扫码开票单" in page_text
+        and "开票流水号" in page_text
+        and "开具" in page_text
+        and "请输入乘车人" not in page_text
+        and "发票开具成功" not in page_text
+    ):
+        return "list"
+    if "请输入乘车人" in page_text and "证件号码后8位" in page_text:
+        return "verify"
+    if "发票信息确认" in page_text:
+        return "invoice_confirm"
+    if "发票信息" in page_text and ("提交" in page_text or "确认" in page_text):
+        return "invoice_info"
+    if "发票开具成功" in page_text:
+        return "success"
+    if "邮箱地址确认" in page_text:
+        return "email_confirm"
+    if "邮件发送完成，请注意查收。" in page_text:
+        return "mail_sent"
+    if "您没有可开具" in page_text or ("发票管理" in page_text and "客票" in page_text):
+        return "no_ticket"
+    if "发票开具中" in page_text or "发票正在开具中" in page_text:
+        return "loading"
+    if "业务须知" in page_text and "国家税务总局" in page_text:
+        return "instructions"
+    if "查询车票" in page_text or ("火车票" in page_text and "飞机票" in page_text):
+        return "home"
+    return "other"
+
+
+def wait_for_page_transition(
+    args: argparse.Namespace,
+    expected_pages: set[str],
+    *,
+    label: str,
+    network_sensitive: bool = False,
+) -> str:
+    """等待页面切到目标状态;先高频 XML 轮询,必要时截图视觉兜底。
+
+    network_sensitive=True 的步骤通常触发 12306 后端请求,先给短等待,若仍是 loading/other
+    就自动延长,避免固定把每步都 sleep 很久。
+    """
+    quick_timeout = float(getattr(args, "smart_wait_quick", 2.0))
+    network_timeout = float(getattr(args, "smart_wait_network", 12.0)) if network_sensitive else 0.0
+    vision_interval = float(getattr(args, "smart_wait_vision_interval", 1.2))
+    deadline = time.perf_counter() + max(quick_timeout, 0.1)
+    network_deadline = deadline + max(network_timeout, 0.0)
+    last_vision_at = 0.0
+    last_kind = "other"
+    phase = "quick"
+
+    while True:
+        nodes = adb_dump_nodes()
+        kind = xml_page_kind(nodes)
+        last_kind = kind
+        if kind in expected_pages:
+            print(f"wait_page_hit label={label} via=xml page={kind} phase={phase}")
+            return kind
+
+        now = time.perf_counter()
+        if getattr(args, "screenshot_wait", False) and now - last_vision_at >= vision_interval:
+            last_vision_at = now
+            try:
+                detected = vision_detect_page()
+                v_page = str(detected.get("page") or "other")
+                summary = str(detected.get("summary") or "")
+                if v_page == "success" and ("开具中" in summary or "正在开具" in summary):
+                    v_page = "loading"
+                print(f"wait_page_probe label={label} via=vision page={v_page} summary={summary}")
+                if v_page in expected_pages:
+                    return v_page
+                if v_page not in {"other", "launcher"}:
+                    last_kind = v_page
+            except Exception as exc:
+                print(f"wait_page_vision_error label={label}: {exc}")
+
+        if now >= deadline:
+            if network_sensitive and now < network_deadline:
+                if phase != "network":
+                    print(f"wait_page_more label={label} reason=network_or_loading last={last_kind} extra={network_timeout}s")
+                    phase = "network"
+                time.sleep(0.5)
+                continue
+            print(f"wait_page_timeout label={label} expected={sorted(expected_pages)} last={last_kind}")
+            return last_kind
+        time.sleep(0.25)
+
+
+def recover_to_h5_list(args: argparse.Namespace, row_label: str, *, reason: str) -> bool:
+    """子流程断档时回列表;back 不够就扫码重进,让下次循环靠进度表续接。"""
+    print(f"recover_start row={row_label} reason={reason}")
+    if is_on_list_page():
+        print(f"recover_ok row={row_label} already_list")
+        return True
+    recovered = safe_back_until_list(max_back=getattr(args, "recover_max_back", 4))
+    if recovered:
+        print(f"recover_ok row={row_label} via=back")
+        return True
+    if is_on_instructions_page() or getattr(args, "recover_auto_scan", True):
+        print(f"recover_try_auto_scan row={row_label}")
+        if auto_scan_qr_to_list():
+            print(f"recover_ok row={row_label} via=auto_scan")
+            return True
+    print(f"recover_failed row={row_label}")
+    return False
+
+
 def vision_list_rows() -> dict:
     """截屏 + 视觉识别列表页所有可见乘客行,返回 parsed dict(已反归一到绝对像素)。
 
@@ -1330,13 +1462,7 @@ def drive_invoice_subflow(args: argparse.Namespace, row_label: str) -> str:
             mark_clicked_as_runtime_error()
             consecutive_errors += 1
             if consecutive_errors >= 2:
-                recovered = safe_back_until_list(max_back=4)
+                recovered = recover_to_h5_list(args, row_label, reason="sub_step_errors")
-                if not recovered:
-                    # 可能在法条说明页,尝试自动扫码回到列表
-                    if is_on_instructions_page():
-                        print(f"sub_recover via auto_scan row={row_label}")
-                        if auto_scan_qr_to_list():
-                            recovered = True
                 print(f"sub_abort row={row_label} consecutive_errors=2 recovered_to_list={recovered}")
                 return "error"
         time.sleep(args.between_steps)
@@ -1692,7 +1818,12 @@ def android_h5_page_run(args: argparse.Namespace) -> None:
                 print(f"retry_failed_attempt {masked}")
             update_progress(masked, real_name, id8, "in_progress")
             run_adb(["shell", "input", "tap", str(tap_point[0]), str(tap_point[1])])
-            wait_short(args, ["请输入乘车人", "证件号码后8位", "发票管理", "您没有可开具"])
+            wait_for_page_transition(
+                args,
+                {"verify", "no_ticket", "invoice_info", "invoice_confirm", "success"},
+                label=f"after_list_open:{masked}",
+                network_sensitive=True,
+            )
 
             result = drive_invoice_subflow(args, masked)
             total_tapped += 1
@@ -1706,13 +1837,7 @@ def android_h5_page_run(args: argparse.Namespace) -> None:
                     elif p_row.get("masked_name") == masked and p_row.get("status") == "no_ticket":
                         total_no_ticket += 1
             else:
-                recovered = safe_back_until_list(max_back=4)
+                recovered = recover_to_h5_list(args, masked, reason=result)
-                if not recovered:
-                    # 可能在法条说明页,尝试自动扫码
-                    if is_on_instructions_page():
-                        print(f"recover via auto_scan row={masked}")
-                        if auto_scan_qr_to_list():
-                            recovered = True
                 if not recovered:
                     print(f"STOP row={masked} sub_result={result} 无法回到列表页")
                     print(f"进度已保存,手机恢复后从 UI 重新启动即可续接")
@@ -1722,6 +1847,9 @@ def android_h5_page_run(args: argparse.Namespace) -> None:
             # back 后 scroll 重置到顶部,跳出 actions 循环重新截屏
             if idx < len(actions) - 1:
                 print(f"  (back 重置 scroll,重新截屏找下一个)")
+            if args.max_actions and total_tapped >= args.max_actions:
+                print(f"stop reason=max_actions total_tapped={total_tapped}")
+                return
             break
 
         # dry-run 模式下当前屏全部看完后自动滚动
@@ -1925,18 +2053,18 @@ def android_invoice_step(args: argparse.Namespace) -> None:
     if "邮箱地址确认" in page_text:
         if not _xml_or_vision_click(nodes, "确认", "email_confirm", clickable=True):
             raise RuntimeError("邮箱确认页未找到确认按钮")
-        wait_short(args, ["邮件发送完成", "发票开具成功"])
+        wait_for_page_transition(args, {"mail_sent", "success"}, label="after_email_confirm", network_sensitive=True)
         return
 
     if "发票信息确认" in page_text:
         if not _xml_or_vision_click(nodes, "确认", "invoice_info_confirm", clickable=True):
             raise RuntimeError("发票信息确认页未找到确认按钮")
-        wait_long(args, ["发票开具成功"])
+        wait_for_page_transition(args, {"success", "no_ticket"}, label="after_invoice_confirm", network_sensitive=True)
         return
 
     if "发票开具中" in page_text or "发票正在开具中" in page_text:
         print("invoice_opening_wait")
-        wait_and_probe(1, getattr(args, "wait_secondary", 10), ["发票开具成功", "邮件发送完成", "发票管理"])
+        wait_for_page_transition(args, {"success", "mail_sent", "no_ticket"}, label="invoice_opening", network_sensitive=True)
         return
 
     if "发票开具成功" in page_text:
@@ -1953,7 +2081,7 @@ def android_invoice_step(args: argparse.Namespace) -> None:
             row = current_row
             update_progress(row.get("masked_name", ""), row.get("real_name", ""), row.get("id_last8", ""), "completed")
             if _xml_or_vision_click(nodes, "发送至邮箱", "send_email"):
-                wait_short(args, ["邮箱地址确认", "邮件发送完成"])
+                wait_for_page_transition(args, {"email_confirm", "mail_sent"}, label="after_send_email", network_sensitive=True)
                 return
         if success_action == "download" and in_progress:
             local_path = click_invoice_download(args, nodes, row_label)
@@ -1973,6 +2101,7 @@ def android_invoice_step(args: argparse.Namespace) -> None:
             update_progress(row.get("masked_name", ""), row.get("real_name", ""), row.get("id_last8", ""), "completed")
         if not _xml_or_vision_click(nodes, "继续开票", "continue_invoice"):
             raise RuntimeError("成功页未找到继续开票按钮(XML+视觉均失败)")
+        wait_for_page_transition(args, {"list", "instructions", "no_ticket"}, label="after_continue_invoice", network_sensitive=False)
         return
 
     if "请输入乘车人" in page_text and "证件号码后8位" in page_text:
@@ -2001,7 +2130,7 @@ def android_invoice_step(args: argparse.Namespace) -> None:
         update_progress(masked_name, real_name, id8, "in_progress")
         if not _xml_or_vision_click(adb_dump_nodes(), "核验", "verify"):
             raise RuntimeError("未找到核验按钮(XML+视觉均失败)")
-        wait_short(args, ["发票管理", "待开票", "发票信息"])
+        wait_for_page_transition(args, {"no_ticket", "invoice_info", "invoice_confirm", "success"}, label="after_verify", network_sensitive=True)
         return
 
     if "身份核验" in page_text and "证件号码的后8位" in page_text:
@@ -2018,7 +2147,7 @@ def android_invoice_step(args: argparse.Namespace) -> None:
         adb_input_text(matches[0]["身份证后8位"])
         if not _xml_or_vision_click(adb_dump_nodes(), "开始核验", "agent_verify"):
             raise RuntimeError("代办人身份核验页未找到开始核验按钮")
-        wait_short(args, ["扫码开票单", "开票流水号"])
+        wait_for_page_transition(args, {"list"}, label="after_agent_verify", network_sensitive=True)
         return
 
     if "发票信息" in page_text and "提交" in page_text:
@@ -2028,7 +2157,7 @@ def android_invoice_step(args: argparse.Namespace) -> None:
             raise RuntimeError("发票信息页未检测到默认抬头/税号，暂停，需手工确认")
         if not _xml_or_vision_click(nodes, "提交", "submit_invoice_info"):
             raise RuntimeError("未找到提交按钮(XML+视觉均失败)")
-        wait_short(args, ["发票信息确认"])
+        wait_for_page_transition(args, {"invoice_confirm"}, label="after_submit_invoice_info", network_sensitive=True)
         return
 
     if "发票管理" in page_text and "您没有可开具电子发票的客票" in page_text:
@@ -2052,13 +2181,13 @@ def android_invoice_step(args: argparse.Namespace) -> None:
             pt = _vision_find_button("开具")
             if pt:
                 run_adb(["shell", "input", "tap", str(pt[0]), str(pt[1])])
-                wait_short(args, ["发票信息"])
+                wait_for_page_transition(args, {"invoice_info", "invoice_confirm"}, label="after_ticket_open_vision", network_sensitive=True)
                 return
             raise RuntimeError("发票管理页未找到单张开具按钮(XML+视觉均失败)")
         # Avoid bottom batch button; choose first visible normal button.
         node = sorted(open_nodes, key=node_center_y)[0]
         click_node(node, "ticket_open")
-        wait_short(args, ["发票信息"])
+        wait_for_page_transition(args, {"invoice_info", "invoice_confirm"}, label="after_ticket_open", network_sensitive=True)
         return
 
     if "扫码开票单" in page_text:
@@ -2068,7 +2197,7 @@ def android_invoice_step(args: argparse.Namespace) -> None:
         masked_name, row, button = passenger
         print(f"passenger={masked_name} matched={row.get('姓名')} id_last8={row.get('身份证后8位')}")
         click_node(button, f"list_open:{masked_name}")
-        wait_short(args, ["请输入乘车人", "证件号码后8位"])
+        wait_for_page_transition(args, {"verify", "no_ticket", "invoice_info"}, label=f"after_xml_list_open:{masked_name}", network_sensitive=True)
         return
 
     if "com.bbk.launcher2" in page_text or "launcher" in page_text:
@@ -2510,6 +2639,8 @@ def build_parser() -> argparse.ArgumentParser:
     p_vision_run = subparsers.add_parser("android-vision-page-run", help="视觉驱动批量开票:截图识别 -> anchor 去重 -> 按页批次 tap -> 子流程 -> 滑动")
     p_vision_run.add_argument("--roster", default="work/passenger_roster.csv")
     p_vision_run.add_argument("--send-email", action="store_true")
+    p_vision_run.add_argument("--invoice-success-action", choices=["download", "email", "continue"], default="", help="成功页动作: download=下载发票,email=发送邮箱,continue=仅继续")
+    p_vision_run.add_argument("--invoice-output-dir", default="", help="PDF/OFD/XML 保存到电脑本地目录")
     p_vision_run.add_argument("--fast", action="store_true")
     p_vision_run.add_argument("--dry-run", action="store_true", help="只识别+匹配,不真 tap;用于验证视觉精度")
     p_vision_run.add_argument("--no-precheck", dest="precheck", action="store_false", help="跳过开始前的视觉自检")
@@ -2543,6 +2674,12 @@ def build_parser() -> argparse.ArgumentParser:
     p_h5_run.add_argument("--wait-secondary", type=int, default=10, help="扩展等待秒数")
     p_h5_run.add_argument("--short-wait", type=int, default=1, help="短等待秒数(页面切换)")
     p_h5_run.add_argument("--between-steps", type=float, default=0.3)
+    p_h5_run.add_argument("--screenshot-wait", action="store_true", help="等待页面跳转时用截图视觉判页兜底")
+    p_h5_run.add_argument("--smart-wait-quick", type=float, default=1.2, help="非网络跳转的快速等待秒数")
+    p_h5_run.add_argument("--smart-wait-network", type=float, default=12.0, help="疑似走网络时额外等待秒数")
+    p_h5_run.add_argument("--smart-wait-vision-interval", type=float, default=1.5, help="截图判页最小间隔秒数")
+    p_h5_run.add_argument("--recover-max-back", type=int, default=4, help="子流程断档时最多 back 次数")
+    p_h5_run.add_argument("--no-recover-auto-scan", dest="recover_auto_scan", action="store_false", help="回列表失败时不自动扫码重进")
     p_h5_run.add_argument("--retry-failed-once", dest="retry_failed_once", type=int, default=1, help="当前运行内对可见 failed/error 人员最多重试次数")
     p_h5_run.add_argument("--no-resume-after-last-progress", dest="resume_after_last_progress", action="store_false", help="不按进度表最后处理人定位续跑")
     # H5 滚动参数(页面内滚动,不是翻页)