From d64013d2ec902a0b17c7f042e8b84325ae6fcfd5 Mon Sep 17 00:00:00 2001
From: zzx <zzx@suq.cn>
Date: Mon, 15 Sep 2025 22:36:46 +0800
Subject: [PATCH] =?UTF-8?q?refactor(ai=5Fseo):=20=E9=87=8D=E6=9E=84=20kimi?=
 =?UTF-8?q?=20=E7=88=AC=E8=99=AB=E4=BB=A5=E9=80=82=E5=BA=94=E7=BD=91?=
 =?UTF-8?q?=E9=A1=B5=E7=BB=93=E6=9E=84=E5=8F=98=E5=8C=96?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- 移除长思考功能，增加确认按钮点击
- 修改输入框点击和文本输入逻辑
- 添加复制按钮点击和剪贴板读取功能- 增加搜索结果获取和解析
- 调整截图功能
---
 spiders/ai_seo/kimi.py | 60 ++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 51 insertions(+), 9 deletions(-)

diff --git a/spiders/ai_seo/kimi.py b/spiders/ai_seo/kimi.py
index 9a43470..322d4b8 100644
--- a/spiders/ai_seo/kimi.py
+++ b/spiders/ai_seo/kimi.py
@@ -2,6 +2,7 @@
 import asyncio
 from functools import partial, wraps
 
+import pyperclip
 from playwright.async_api import Browser
 
 from abs_spider import AbstractAiSeoSpider
@@ -43,13 +44,16 @@ class KimiSpider(AbstractAiSeoSpider):
         await self.browser_page.goto('https://www.kimi.ai', timeout=600000)
         self.ai_answer = AiAnswer(self.get_platform_id(), self.get_platform_name(), self.prompt, self.keyword)
         await asyncio.sleep(3)
-        if self.think:
-            think_btn = self.browser_page.locator("span:text('长思考 (k1.5)')").locator('..')
-            if await think_btn.is_visible():
-                clazz = (await think_btn.get_attribute('class')).split(' ')
-                if 'open' not in clazz:
-                    await think_btn.click()
-                    await asyncio.sleep(2)
+        confirm_btn = self.browser_page.locator('//button[text()="知道了"]')
+        if await confirm_btn.is_visible():
+            await confirm_btn.click()
+        # if self.think:
+        #     think_btn = self.browser_page.locator("span:text('长思考 (k1.5)')").locator('..')
+        #     if await think_btn.is_visible():
+        #         clazz = (await think_btn.get_attribute('class')).split(' ')
+        #         if 'open' not in clazz:
+        #             await think_btn.click()
+        #             await asyncio.sleep(2)
         chat_input_element = self.browser_page.locator("//div[@class='chat-input']")
         await chat_input_element.click()
         # 输入提问词
@@ -57,10 +61,19 @@ class KimiSpider(AbstractAiSeoSpider):
         await asyncio.sleep(2)
         await self.browser_page.keyboard.press('Enter')
         # 监听请求
-        self.browser_page.on('response', partial(self.__listen_response))
-        await self.completed_event.wait()
+        # self.browser_page.on('response', partial(self.__listen_response))
+        # await self.completed_event.wait()
         await asyncio.sleep(2)
 
+        # 等待复制按钮可见
+        copy_btn_xpath = "//div[@class='segment-assistant-actions-content']/div[@class='simple-button size-small'][1]"
+        await self.browser_page.wait_for_selector(copy_btn_xpath, timeout=600000)
+        copy_btn = self.browser_page.locator(copy_btn_xpath)
+        await copy_btn.click()
+        # 读取剪贴板
+        self.ai_answer.answer = pyperclip.paste()
+        logger.debug(f"ai回复: {self.ai_answer.answer}")
+
         # 报错检查
         if self.fail_status:
             await AiSeoApis.update_spider_session(self.session_info['id'], 2)
@@ -85,6 +98,35 @@ class KimiSpider(AbstractAiSeoSpider):
         search_list_element = self.browser_page.locator("//div[@class='search-plus']")
         if await search_list_element.is_visible() and not await search_list_content_element.is_visible():
             await search_list_element.click()
+
+        # 获取搜索结果
+        search_list = []
+        search_elements = await self.browser_page.locator("//div[@class='sites']/a[@class='site']").all()
+        for search_element in search_elements:
+            result = AiSearchResult()
+            result.url = await search_element.get_attribute('href')
+
+            children = await search_element.locator("xpath=./child::*").all()
+            result.title = await children[1].inner_text()
+            result.body = await children[2].inner_text()
+
+            # 获取信源信息元素
+            result_host_elements = await children[0].locator("xpath=./child::*").all()
+            try:
+                result.host_name = await result_host_elements[1].inner_text()
+            except Exception:
+                result.host_name = ''
+            if len(result_host_elements) >= 3:
+                try:
+                    result.publish_time = await result_host_elements[2].inner_text()
+                    result.publish_time = result.publish_time.replace('/', '-')
+                except Exception:
+                    result.publish_time = 0
+                    logger.error(f"{result.title}获取发布时间失败")
+            search_list.append(result)
+        self.ai_answer.search_result = search_list
+
+
         # 截图
         screenshot_path = self._get_screenshot_path()
         self.ai_answer.screenshot_file = screenshot_path