pyxllib 0.3.151__tar.gz → 0.3.175__tar.gz

{pyxllib-0.3.151 → pyxllib-0.3.175}/MANIFEST.in

@@ -1,4 +1,5 @@
 include README.md
 include requirements.txt
 recursive-include pyxllib *.html *.js
-recursive-include pyxlpr *.py *.txt
+recursive-include pyxlpr *.py *.txt
+exclude pyxllib/ext/wpsapi2.py

{pyxllib-0.3.151/pyxllib.egg-info → pyxllib-0.3.175}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyxllib
-Version: 0.3.151
+Version: 0.3.175
 Summary: 厦门理工模式识别团队通用python代码工具库
 Home-page: https://github.com/XLPRUtils/pyxllib
 Author: code4101
@@ -38,33 +38,33 @@ Requires-Dist: Jinja2
 Requires-Dist: pandas>=0.23.4
 Requires-Dist: pillow
 Provides-Extra: xlcv
+Requires-Dist: opsdroid-get-image-size; extra == "xlcv"
 Requires-Dist: opencv-python; extra == "xlcv"
 Requires-Dist: pillow; extra == "xlcv"
-Requires-Dist: opsdroid-get-image-size; extra == "xlcv"
 Provides-Extra: xlai
-Requires-Dist: pillow; extra == "xlai"
 Requires-Dist: visdom; extra == "xlai"
 Requires-Dist: pynvml; extra == "xlai"
-Requires-Dist: opsdroid-get-image-size; extra == "xlai"
 Requires-Dist: opencv-python; extra == "xlai"
+Requires-Dist: pillow; extra == "xlai"
+Requires-Dist: opsdroid-get-image-size; extra == "xlai"
 Provides-Extra: xlall
-Requires-Dist: flask-jwt-extended; extra == "xlall"
-Requires-Dist: openai2; extra == "xlall"
+Requires-Dist: flask-cors; extra == "xlall"
 Requires-Dist: arts; extra == "xlall"
 Requires-Dist: ujson; extra == "xlall"
-Requires-Dist: pillow; extra == "xlall"
-Requires-Dist: flask-wtf; extra == "xlall"
-Requires-Dist: xlrd; extra == "xlall"
-Requires-Dist: zhconv; extra == "xlall"
-Requires-Dist: html2text; extra == "xlall"
 Requires-Dist: flask; extra == "xlall"
+Requires-Dist: html2text; extra == "xlall"
 Requires-Dist: uvicorn; extra == "xlall"
-Requires-Dist: flask-cors; extra == "xlall"
-Requires-Dist: opsdroid-get-image-size; extra == "xlall"
+Requires-Dist: flask-restful; extra == "xlall"
+Requires-Dist: flask-jwt-extended; extra == "xlall"
+Requires-Dist: openai2; extra == "xlall"
+Requires-Dist: xlrd; extra == "xlall"
 Requires-Dist: fastapi; extra == "xlall"
-Requires-Dist: sentencepiece; extra == "xlall"
 Requires-Dist: opencv-python; extra == "xlall"
-Requires-Dist: flask-restful; extra == "xlall"
+Requires-Dist: sentencepiece; extra == "xlall"
+Requires-Dist: pillow; extra == "xlall"
+Requires-Dist: flask-wtf; extra == "xlall"
+Requires-Dist: opsdroid-get-image-size; extra == "xlall"
+Requires-Dist: zhconv; extra == "xlall"
 
 # 1 install
 

{pyxllib-0.3.151 → pyxllib-0.3.175}/pyxllib/algo/specialist.py

@@ -11,7 +11,8 @@ import numpy as np
 import pandas as pd
 
 from pyxllib.prog.pupil import DictTool
-from pyxllib.prog.deprecatedlib import deprecated
+# from pyxllib.prog.deprecatedlib import deprecated
+from deprecated import deprecated
 
 
 @deprecated(reason='这个实现方式不佳，请参考 make_index_function')

{pyxllib-0.3.151 → pyxllib-0.3.175}/pyxllib/autogui/activewin.py

@@ -95,7 +95,7 @@ class ActiveWindowCapture(OriginalMSS):
         for i in range(4):
             win = uiautomation.GetForegroundControl()
             if win:
-                return
+                return win
             else:
                 time.sleep(0.5)
         return

{pyxllib-0.3.151 → pyxllib-0.3.175}/pyxllib/autogui/uiautolib.py

@@ -25,12 +25,14 @@ from anytree import NodeMixin
 
 import ctypes
 from ctypes import wintypes
-import win32con
-import win32gui
-import win32process
-import win32clipboard
 
-import uiautomation as auto
+if sys.platform == 'win32':
+    import win32con
+    import win32gui
+    import win32process
+    import win32clipboard
+
+    import uiautomation as auto
 
 
 def __1_clipboard_utils():

{pyxllib-0.3.151 → pyxllib-0.3.175}/pyxllib/autogui/wechat.py

@@ -8,17 +8,22 @@
 代码参考(抄自)：https://github.com/Frica01/WeChatMassTool
 """
 
+import sys
 import time
 from copy import deepcopy
 from typing import (Iterable, Callable, List)
 from typing import Optional
+from unittest.mock import MagicMock
 
-import uiautomation as auto
+from loguru import logger
 
-import win32con
-import win32gui
+if sys.platform == 'win32':
+    import uiautomation as auto
+    import win32con
+    import win32gui
 
 from pyxllib.prog.filelock import get_autoui_lock
+
 from pyxllib.autogui.uiautolib import find_ctrl, UiCtrlNode, copy_files_to_clipboard
 
 
@@ -391,6 +396,7 @@ class WeChatMainWnd(UiCtrlNode):
 
         search_box.Click()
         search_box.SendKeys(name)
+        time.sleep(2)
         search_box.SendKeys('{Enter}')
 
     def __4_编辑器(self):
@@ -479,6 +485,35 @@ def wechat_lock_send(user, text=None, files=None, *, timeout=-1):
             we.send_files(files)
 
 
+def wechat_handler(message):
+    # 获取群名，如果没有指定，不使用此微信发送功能
+    user = message.record["extra"].get("wechat_user")
+    if user:
+        wechat_lock_send(user, message)
+
+
+if sys.platform == 'win32':
+    # 创建专用的微信日志记录器，不绑定默认群名
+    wechat_logger = logger.bind(wechat_user='文件传输助手')
+
+    # 添加专用的微信处理器
+    wechat_logger.add(wechat_handler,
+                      format="{time:YYYY-MM-DD HH:mm:ss} | {level} | {name}:{function}:{line} - {message}")
+
+    """ 往微信发送特殊的日志格式报告
+    用法：wechat_logger.bind(wechat_user='文件传输助手').info(message)
+    
+    或者：
+    # 先做好默认群名绑定
+    wechat_logger = wechat_logger.bind(wechat_user='考勤管理')
+    # 然后就能普通logger用法发送了
+    wechat_logger.info('测试')
+    """
+else:
+    # 降级为普通logger
+    wechat_logger = logger
+
+
 def __4_wx():
     """ 别人的原版实现 """
 

pyxllib-0.3.175/pyxllib/autogui/wxautolib.py

@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+# @Author : 陈坤泽
+# @Email  : 877362867@qq.com
+# @Date   : 2024/12/16
+
+import sys
+
+from loguru import logger
+
+try:  # 尝试加载VIP版
+    from wxautox import WeChat
+    from wxautox.elements import WxParam  # WxParam.DEFALUT_SAVEPATH 可以用来配置数据自动保存位置
+except ModuleNotFoundError:  # 否则用贫民版
+    from wxauto import WeChat
+    from wxauto.elements import WxParam
+
+from pyxllib.prog.filelock import get_autoui_lock
+
+
+class WeChatSingletonLock:
+    """ 基于 get_autoui_lock 的微信全局唯一单例控制器，确保同一时间仅有一个微信自动化程序在操作 """
+
+    def __init__(self, lock_timeout=-1, *, init=True):
+        # 初始化全局锁
+        self.lock = get_autoui_lock(timeout=lock_timeout)
+        self.wx = WeChat() if init else None
+
+    def __enter__(self):
+        # 获取锁并激活微信窗口
+        self.lock.acquire()
+        if self.wx:
+            self.wx._show()
+            return self.wx
+
+    def __exit__(self, exc_type, exc_value, traceback):
+        # 释放锁
+        self.lock.release()
+
+
+def wechat_lock_send(user, text=None, files=None, *, timeout=-1):
+    """ 使用全局唯一单例锁，确保同一时间仅有一个微信自动化程序在操作 """
+    with WeChatSingletonLock(timeout) as we:
+        # 241223周一12:27，今天可被这个默认2秒坑惨了，往错误群一直发骚扰消息
+        # 22:07，但我复测，感觉不可能找不到啊，为什么会找到禅宗考勤管理群呢，太离谱了
+        status = we.ChatWith(user, timeout=5)
+
+        if status != user:
+            raise ValueError(f'无法找到用户：{user}')
+
+        if text:
+            we.SendMsg(text, user)
+        if files:
+            we.SendFiles(files, user)
+
+
+def wechat_handler(message):
+    # 获取群名，如果没有指定，不使用此微信发送功能
+    user = message.record["extra"].get("wechat_user")
+    if user:
+        wechat_lock_send(user, message)
+
+
+if sys.platform == 'win32':
+    # 创建专用的微信日志记录器，不绑定默认群名
+    wechat_logger = logger.bind(wechat_user='文件传输助手')
+
+    # 添加专用的微信处理器
+    wechat_logger.add(wechat_handler,
+                      format="{time:YYYY-MM-DD HH:mm:ss} | {level} | {name}:{function}:{line} - {message}")
+
+    """ 往微信发送特殊的日志格式报告
+    用法：wechat_logger.bind(wechat_user='文件传输助手').info(message)
+
+    或者：
+    # 先做好默认群名绑定
+    wechat_logger = wechat_logger.bind(wechat_user='考勤管理')
+    # 然后就能普通logger用法发送了
+    wechat_logger.info('测试')
+    """
+else:
+    # 降级为普通logger
+    wechat_logger = logger

pyxllib-0.3.175/pyxllib/cv/slidercaptcha.py

@@ -0,0 +1,137 @@
+import os
+import cv2
+import numpy as np
+from xlproject.code4101 import XlPath
+
+from pyxllib.file.specialist import download_file
+from pyxllib.prog.pupil import is_url
+
+
+class SliderCaptchaLocator:
+    """ 滑动验证码的位置定位
+
+    由于缺口一般是一个黑底图，整体颜色是偏暗的，可以利用这个特性来定位列所在位置
+    """
+
+    def __init__(self, background_image_path, slider_image_path=None):
+        """
+        :param background_image_path: 背景主图片
+        :param slider_image_path: 滑动条图片
+        """
+        # 支持输入是url的情况
+        if is_url(background_image_path):
+            background_image_path = download_file(background_image_path, temp=True, ext='.png')
+        if is_url(slider_image_path):
+            slider_image_path = download_file(slider_image_path, temp=True, ext='.png')
+        # 初始化图片等数据
+        self.image_path = background_image_path
+        self.image = cv2.imread(self.image_path)
+        self.darkest_rows = None
+        self.column_averages = None
+        self.enhanced_averages = None
+        self.radius = 24
+        if slider_image_path is not None:
+            # 先读取图片
+            template_image = cv2.imread(slider_image_path)
+            # 取图片宽度的40%作为radius
+            self.radius = int(template_image.shape[1] * 0.4)
+
+    def select_darkest_part_rows(self, fraction=0.1):
+        """
+        选择最暗的部分行，供下游计算列的平均暗度
+
+        :param float fraction: 最暗行的比例
+        :return: 最暗行的图像部分
+        """
+        if self.darkest_rows is None:
+            average_row_values = np.mean(self.image, axis=(1, 2))
+            threshold_index = int(len(average_row_values) * fraction)
+            darkest_indices = np.argsort(average_row_values)[:threshold_index]
+            self.darkest_rows = self.image[darkest_indices]
+        return self.darkest_rows
+
+    def calculate_column_averages(self):
+        """
+        计算列的平均强度。
+
+        :return: 列的平均强度
+        """
+        if self.column_averages is None:
+            rows = self.select_darkest_part_rows()
+            column_values = np.mean(rows, axis=0)
+            column_intensity = np.mean(column_values, axis=1)
+            min_val, max_val = np.min(column_intensity), np.max(column_intensity)
+            self.column_averages = ((column_intensity - min_val) / (max_val - min_val) * 255).astype(
+                np.uint8) if max_val != min_val else np.zeros_like(column_intensity)
+        return self.column_averages
+
+    def enhance_column_averages(self):
+        """
+        增强列平均值，利用验证码的黑度图的对称性，进行一个增强，使得其他非验证码的零散黑条变白
+
+        :param int radius: 对比半径，这个建议设成模版图宽度的40%~50%
+        :return: 增强后的列平均值
+        """
+        if self.enhanced_averages is None:
+            averages = self.calculate_column_averages()
+            length = len(averages)
+            enhanced = np.zeros_like(averages)
+            for i in range(length):
+                max_value = averages[i]
+                for j in range(1, self.radius // 2 + 1):
+                    if i - j >= 0 and i + j < length:
+                        max_value = max(max_value, abs(int(averages[i - j]) - int(averages[i + j])))
+                enhanced[i] = max_value
+            self.enhanced_averages = enhanced
+        return self.enhanced_averages
+
+    def find_captcha_position(self, threshold=50):
+        """
+        查找验证码的位置。
+
+        :param int threshold: 达到多少像素后，开始计算中位数
+        :return: 验证码的列中心位置
+        """
+        intensities = self.enhance_column_averages()
+        indices = []
+        for intensity in range(256):
+            new_indices = np.where(intensities == intensity)[0]
+            if len(indices) + len(new_indices) > threshold:
+                indices.extend(new_indices[:threshold - len(indices)])
+                break
+            indices.extend(new_indices)
+        return int(np.median(indices))
+
+    def draw_red_line(self, position, line_thickness=3):
+        """
+        在图像上绘制红线。
+
+        :param int position: 红线位置
+        :param int line_thickness: 线宽
+        :return: 绘制红线后的图像
+        """
+        gray_image = np.tile(self.calculate_column_averages(), (100, 1)).astype(np.uint8)
+        color_image = cv2.cvtColor(gray_image, cv2.COLOR_GRAY2BGR)
+        cv2.line(color_image, (position, 0), (position, gray_image.shape[0]), (0, 0, 255), line_thickness)
+        return color_image
+
+    def debug(self):
+        position = self.find_captcha_position()
+        final_image = self.draw_red_line(position)
+        cv2.imshow('Image with Red Line', final_image)
+        cv2.waitKey(0)
+        cv2.destroyAllWindows()
+
+
+def main():
+    os.chdir(XlPath.desktop())
+    locator = SliderCaptchaLocator('cap_union_new_getcapbysig.png')
+    position = locator.find_captcha_position()
+    final_image = locator.draw_red_line(position)
+    cv2.imshow('Image with Red Line', final_image)
+    cv2.waitKey(0)
+    cv2.destroyAllWindows()
+
+
+if __name__ == '__main__':
+    main()

{pyxllib-0.3.151 → pyxllib-0.3.175}/pyxllib/data/pglib.py

@@ -131,7 +131,19 @@ WHERE {table_name}.{item_id_name} = cte.{item_id_name}"""
 
         # 找到目前最大的id值
         max_id = self.exec2one(f'SELECT MAX({item_id_name}) FROM {table_name}')
-        self.execute(f'ALTER SEQUENCE {counter_name} RESTART WITH {max_id + 1}')
+        # self.execute(f'ALTER SEQUENCE {counter_name} RESTART WITH {max_id + 1}')
+        # 检查序列是否存在，如果不存在则创建序列，然后重置序列
+        # 检查序列是否存在，如果不存在则创建序列，并将其关联到指定的表和字段
+        self.execute(f"""
+DO $$
+BEGIN
+    IF NOT EXISTS (SELECT FROM pg_class WHERE relkind = 'S' AND relname = '{counter_name}') THEN
+        EXECUTE format('CREATE SEQUENCE %I', '{counter_name}');
+        EXECUTE format('ALTER TABLE %I ALTER COLUMN %I SET DEFAULT nextval(''%I'')', '{table_name}', '{item_id_name}', '{counter_name}');
+    END IF;
+    EXECUTE format('ALTER SEQUENCE %I RESTART WITH %s', '{counter_name}', {max_id + 1});
+END $$;
+""")
         self.commit()
 
     def __3_execute(self):

{pyxllib-0.3.151 → pyxllib-0.3.175}/pyxllib/ext/drissionlib.py

@@ -3,21 +3,19 @@
 # @Author : 陈坤泽
 # @Email  : 877362867@qq.com
 # @Date   : 2024/04/12
-import time
-
-from pyxllib.prog.pupil import check_install_package
-
-check_install_package('DrissionPage')
 
 import re
 from urllib.parse import unquote
+import tempfile
+import time
+from urllib.parse import urlparse
 
 from deprecated import deprecated
+from loguru import logger
 import DrissionPage
 from DrissionPage import ChromiumPage, Chromium
 from DrissionPage._pages.chromium_base import ChromiumBase
 from DrissionPage._pages.chromium_tab import ChromiumTab
-from DrissionPage._base.base import BasePage, BaseElement
 
 from pyxllib.prog.pupil import inject_members
 from pyxllib.text.pupil import strfind
@@ -79,6 +77,7 @@ def get_dp_tab(dp_tab=None) -> 'XlTab':
         def dp_page2(tab):  # 默认开启页面复用
             if tab.url == dp_tab:
                 return tab
+
         tab = get_dp_tab(dp_page2)
         if tab.url == 'about:blank':
             tab.get(dp_tab)
@@ -87,37 +86,17 @@ def get_dp_tab(dp_tab=None) -> 'XlTab':
         return Chromium().latest_tab
 
 
-def get_latest_not_dev_tab(page=None):
+def get_latest_not_dev_tab(browser=None):
     """ 开发工具本身也会算一个tab，这个函数返回最新的一个不是开发工具的tab """
-    if page is None:
-        page = ChromiumPage()
-    tabs = page.get_tabs()
+    if browser is None:
+        browser = Chromium()
+    tabs = browser.get_tabs()
     for tab in tabs:
         if strfind(tab.url, ['devtools://', 'chrome-extension://']) != -1:
             continue
         return tab
 
 
-def set_input_text(input_ele, text):
-    """ 因为input输入框可能原本就自带了内容，为了不重复输入，先清空再输入 """
-    input_ele.clear()
-    input_ele.input(text)
-
-
-def search_download_file(file_name):
-    file_name = file_name.replace(':', '_')
-    files = ChromiumPage().get_download_files()
-    for file in files:
-        if file_name in file['file']:  # 正常情况下的匹配
-            return file
-    for file in files:
-        file2 = file['file'].replace('+', ' ')
-        if file_name in file2:  # 但有时候'+'好像有点特别
-            return file
-        if file_name in re.sub(r'\s+', ' ', file2):
-            return file
-
-
 class XlChromiumBase(ChromiumBase):
     def get2(self, url, show_errmsg=False, retry=None, interval=None):
         """
@@ -140,6 +119,10 @@ class XlChromiumBase(ChromiumBase):
     def get_download_files(self: Chromium):
         """ 获取下载列表
 
+         (241205周四21:02，这个功能原本是用来做页面文件下载的，
+        但后来知道dp有更简洁的解决方案后，其实原本功能意义已不大，
+        只是作为一个结构化解析下载页面的功能，可以保留参考)
+
         :param search_name: 搜索文件名，输入该参数时，只会从上往下找到第一个匹配的文件
             否则返回一个list结构，存储下载清单里的文件
         :return:
@@ -204,3 +187,60 @@ class XlTab(XlChromiumBase, ChromiumTab, Chromium):
 
 def wait_page_not_change(page, interval=3):
     page.wait_page_not_change(interval)
+
+
+class DpWebBase:
+    """ 基于dp开发的爬虫工具的一个基础类 """
+
+    def __init__(self, url=None, *, base_url=None):
+        self.browser = Chromium()
+        self.browser.set.download_path(tempfile.gettempdir())
+
+        parsed_url = urlparse(url)
+        root_url = f"{parsed_url.scheme}://{parsed_url.netloc}"  # 构建基础 URL
+        self.tab: XlTab = self.browser.new_tab(url)
+        self.base_url = base_url or root_url  # 使用基础 URL 作为 base_url，后续同域名网站去重用
+
+    def close_if_exceeds_min_tabs(self, min_tabs_to_keep=1):
+        """ 检查同网站的tab数量，如果超过最小保留数量则关闭当前页面 """
+        try:  # 靠Py结束触发的可能报错：ImportError: sys.meta_path is None, Python is likely shutting down
+            if self.tab and self.base_url and len(self.browser.get_tabs(url=self.base_url)) > min_tabs_to_keep:
+                self.tab.close()
+        except Exception as e:
+            pass
+
+    # 实测效果不稳定，感觉还不如手动触发吧~
+    # def __del__(self):
+    #     """ 我习惯每次新任务建立新的tab处理，并在结束后自动检查同网页打开的标签是否不唯一则删掉 """
+    #     self.close_if_exceeds_min_tabs()
+
+
+def close_duplicate_tabs(browser=None):
+    """ 关闭浏览器重复标签页
+
+    遍历所有标签页（从前往后，dp的get_tabs拿到的tab就是从新到旧的），对每个域名仅保留第一个出现的标签页，其余同域名标签页关闭；
+    如果最后还剩多个标签页，则把'chrome://newtab/'也关掉。
+    """
+    # 1 初始化
+    if browser is None:
+        browser = Chromium()
+    all_tabs = browser.get_tabs()
+    seen_domains = set()
+
+    # 2 第一次遍历：保留首个出现的域名，其余重复则关闭
+    for t in all_tabs:
+        parsed_url = urlparse(t.url)
+        domain = parsed_url.netloc  # netloc 通常可拿到域名部分
+        # logger.info(f'{t.url}, {domain}')
+
+        if domain in seen_domains:
+            t.close()
+        else:
+            seen_domains.add(domain)
+
+    # 3 第二次遍历：如果剩余标签页 > 1，则关掉chrome://newtab/
+    remaining_tabs = browser.get_tabs()
+    if len(remaining_tabs) > 1:
+        for t in remaining_tabs:
+            if t.url.startswith('chrome://newtab'):
+                t.close()