mcp-query-table 0.3.8__tar.gz → 0.3.10__tar.gz

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mcp_query_table
-Version: 0.3.8
+Version: 0.3.10
 Summary: query table from website, support MCP
 Author-email: wukan <wu-kan@163.com>
 License: MIT License
@@ -33,8 +33,7 @@ Requires-Dist: loguru
 Requires-Dist: mcp
 Requires-Dist: pandas
 Requires-Dist: playwright
-Requires-Dist: playwright-stealth
-Requires-Dist: setuptools
+Requires-Dist: playwright-stealth>=2.0.0
 Requires-Dist: tabulate
 Description-Content-Type: text/markdown
 

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/mcp_query_table/__main__.py

@@ -17,7 +17,7 @@ def main():
     parser.add_argument("--executable_path", type=str, help="浏览器路径",
                         nargs="?", default=r'C:\Program Files\Google\Chrome\Application\chrome.exe')
     parser.add_argument("--user_data_dir", type=str, help="浏览器用户数据目录",
-                        nargs="?", default=rf'C:\Users\{getpass.getuser()}\AppData\Local\Google\Chrome\User Data')
+                        nargs="?", default=rf'C:\Users\{getpass.getuser()}\AppData\Local\Google\Chrome\User Data\Default')
     parser.add_argument("--transport", type=str, help="传输类型",
                         default='stdio', choices=['stdio', 'sse'])
     parser.add_argument("--host", type=str, help="MCP服务端绑定地址",

mcp_query_table-0.3.10/mcp_query_table/_version.py

@@ -0,0 +1 @@
+__version__ = "0.3.10"

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/mcp_query_table/server.py

@@ -53,9 +53,10 @@ async def query(
         query_type: Annotated[QueryType, Field(default=QueryType.CNStock,
                                                description="查询类型。支持`A股`、`指数`、`基金`、`港股`、`美股`等")],
         max_page: Annotated[int, Field(default=1, ge=1, le=10, description="最大页数。只查第一页即可")],
+        rename: Annotated[bool, Field(default=False, description="是否重命名列名")],
         site: Annotated[Site, Field(default=Site.THS, description="站点。支持`东方财富`、`通达信`、`同花顺`")]
 ) -> str:
-    return await qsv.query(query_input, query_type, max_page, site)
+    return await qsv.query(query_input, query_type, max_page, rename, site)
 
 
 # chat功能不通过mcp暴露，因为在Cline等客户端中本就有LLM功能，反而导致返回的数据没有正确提交

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/mcp_query_table/sites/eastmoney.py

@@ -81,7 +81,7 @@ class Pagination:
             datas.extend(v)
         return datas
 
-    def get_dataframe(self):
+    def get_dataframe(self, rename: bool):
         columns = {x['key']: x['title'] for x in self.columns}
         dtypes = {x['key']: convert_type(x['dataType']) for x in self.columns}
 
@@ -98,7 +98,10 @@ class Pagination:
             except ValueError:
                 logger.info("转换失败 {}:{}", k, v)
 
-        return df.rename(columns=columns)
+        if rename:
+            return df.rename(columns=columns)
+        else:
+            return df
 
 
 P = Pagination()
@@ -121,7 +124,8 @@ async def on_response(response):
 async def query(page: Page,
                 q: str = "收盘价>100元",
                 type_: QueryType = 'stock',
-                max_page: int = 5) -> pd.DataFrame:
+                max_page: int = 5,
+                rename: bool = True) -> pd.DataFrame:
     type = _type_.get(type_, None)
     assert type is not None, f"不支持的类型:{type_}"
 
@@ -141,4 +145,4 @@ async def query(page: Page,
             await page.get_by_role("button", name="下一页").click()
         await on_response(await response_info.value)
 
-    return P.get_dataframe()
+    return P.get_dataframe(rename)

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/mcp_query_table/sites/iwencai.py

@@ -10,10 +10,8 @@ import re
 import pandas as pd
 from loguru import logger
 from playwright.async_api import Page
-from playwright_stealth import stealth_async
 
 from mcp_query_table.enums import QueryType
-from mcp_query_table.utils import FixedConfig
 
 # 初次查询页面
 _PAGE1_ = 'https://www.iwencai.com/customized/chart/get-robot-data'
@@ -82,7 +80,7 @@ class Pagination:
             datas.extend(v)
         return datas
 
-    def get_dataframe(self):
+    def get_dataframe(self, rename: bool):
         columns = {x['key']: x['index_name'] for x in self.columns}
         dtypes = {x['key']: convert_type(x['type']) for x in self.columns}
 
@@ -96,7 +94,10 @@ class Pagination:
             except ValueError:
                 logger.info("转换失败 {}:{}", k, v)
 
-        return df.rename(columns=columns)
+        if rename:
+            return df.rename(columns=columns)
+        else:
+            return df
 
 
 P = Pagination()
@@ -150,12 +151,11 @@ async def on_response(response):
 async def query(page: Page,
                 w: str = "收盘价>1000元",
                 type_: QueryType = 'stock',
-                max_page: int = 5) -> pd.DataFrame:
+                max_page: int = 5,
+                rename: bool = False) -> pd.DataFrame:
     querytype = _querytype_.get(type_, None)
     assert querytype is not None, f"不支持的类型:{type_}"
 
-    await stealth_async(page, FixedConfig())
-
     await page.route(re.compile(r'.*\.(?:jpg|jpeg|png|gif|webp)(?:$|\?)'), lambda route: route.abort())
 
     P.reset()
@@ -172,4 +172,4 @@ async def query(page: Page,
             await page.get_by_text("下页").click()
         await on_response(await response_info.value)
 
-    return P.get_dataframe()
+    return P.get_dataframe(rename)

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/mcp_query_table/sites/tdx.py

@@ -78,7 +78,7 @@ class Pagination:
             datas.extend(v)
         return datas
 
-    def get_dataframe(self):
+    def get_dataframe(self, rename: bool):
         dtypes = [convert_type(x) for x in self.dtypes]
         df = pd.DataFrame(self.get_list(), columns=self.columns)
         for i, v in enumerate(dtypes):
@@ -128,7 +128,8 @@ async def on_response2(response):
 async def query(page: Page,
                 message: str = "收盘价>100元",
                 type_: QueryType = 'AG',
-                max_page: int = 5) -> pd.DataFrame:
+                max_page: int = 5,
+                rename: bool = False) -> pd.DataFrame:
     queryType = _queryType_.get(type_, None)
     assert queryType is not None, f"不支持的类型:{type_}"
 
@@ -147,4 +148,4 @@ async def query(page: Page,
             await page.get_by_role("button", name="下一页").click()
         await on_response1(await response_info.value)
 
-    return P.get_dataframe()
+    return P.get_dataframe(rename)

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/mcp_query_table/tool.py

@@ -4,11 +4,12 @@ import sys
 import time
 from pathlib import Path
 from typing import Optional
-from urllib.parse import urlparse
+from urllib.parse import urlparse, quote
 
 import pandas as pd
 from loguru import logger
 from playwright.async_api import async_playwright, Playwright, Page
+from playwright_stealth import Stealth
 
 from mcp_query_table.enums import QueryType, Site, Provider
 
@@ -204,6 +205,8 @@ class BrowserManager:
             self.context = await self.browser.new_context()
         else:
             self.context = self.browser.contexts[0]
+        # 爱问财，无头模式，需要使用 stealth 插件
+        await Stealth().apply_stealth_async(self.context)
 
         # 复用打开的page
         for page in self.context.pages:
@@ -246,7 +249,9 @@ async def query(
         query_input: str = "收盘价>100元",
         query_type: QueryType = QueryType.CNStock,
         max_page: int = 5,
-        site: Site = Site.THS) -> pd.DataFrame:
+        rename: bool = False,
+        site: Site = Site.THS,
+        ) -> pd.DataFrame:
     """查询表格
 
     Parameters
@@ -259,6 +264,8 @@ async def query(
         查询类型, by default QueryType.astock
     max_page : int, optional
         最大页数, by default 5
+    rename: bool
+        是否重命名列名, by default False
     site : Site, optional
         站点, by default Site.iwencai
 
@@ -268,16 +275,17 @@ async def query(
         查询结果
 
     """
+    query_input = quote(query_input.strip(), safe='')
 
     if site == Site.EastMoney:
         from mcp_query_table.sites.eastmoney import query
-        return await query(page, query_input, query_type, max_page)
+        return await query(page, query_input, query_type, max_page, rename)
     if site == Site.THS:
         from mcp_query_table.sites.iwencai import query
-        return await query(page, query_input, query_type, max_page)
+        return await query(page, query_input, query_type, max_page, rename)
     if site == Site.TDX:
         from mcp_query_table.sites.tdx import query
-        return await query(page, query_input, query_type, max_page)
+        return await query(page, query_input, query_type, max_page, rename)
 
     raise ValueError(f"未支持的站点:{site}")
 

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/mcp_query_table/utils.py

@@ -1,10 +1,6 @@
-import random
-import string
 from pathlib import Path
 from typing import List, Tuple
 
-from playwright_stealth import StealthConfig
-
 
 def is_image(path: str) -> bool:
     """判断是否是图片文件"""
@@ -36,16 +32,3 @@ class GlobalVars:
 
     def get_text(self):
         return self.text
-
-
-# https://github.com/AtuboDad/playwright_stealth/issues/31#issuecomment-2342541305
-class FixedConfig(StealthConfig):
-
-    @property
-    def enabled_scripts(self):
-        key = "".join(random.choices(string.ascii_letters, k=10))
-        for script in super().enabled_scripts:
-            if "const opts" in script:
-                yield script.replace("const opts", f"window.{key}")
-                continue
-            yield script.replace("opts", f"window.{key}")

{mcp_query_table-0.3.8 → mcp_query_table-0.3.10}/pyproject.toml

@@ -16,9 +16,8 @@ dependencies = [
     "pandas",
     "loguru",
     "playwright",
-    "playwright-stealth",
+    "playwright-stealth>=2.0.0", # https://github.com/Mattwmaster58/playwright_stealth
     "mcp",
-    "setuptools", # playwright-stealth中要使用pkg_resources，而pkg_resources在setuptools中
     "tabulate"
 ]
 dynamic = ["version"]

mcp_query_table-0.3.8/mcp_query_table/_version.py

@@ -1 +0,0 @@
-__version__ = "0.3.8"