pyxllib 0.3.107__tar.gz → 0.3.108__tar.gz

{pyxllib-0.3.107/pyxllib.egg-info → pyxllib-0.3.108}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyxllib
-Version: 0.3.107
+Version: 0.3.108
 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

{pyxllib-0.3.107 → pyxllib-0.3.108}/pyxllib/algo/pupil.py

@@ -152,6 +152,78 @@ class ValuesStat:
             raise ValueError("无效的数据数量")
 
 
+class ValuesStat2:
+    """ 240509周四17:33，第2代统计器
+    """
+
+    def __init__(self, values=None, raw_values=None):
+        from statistics import pstdev, mean
+
+        # 支持输入可能带有非数值类型的raw_values
+        if raw_values:
+            values = [x for x in raw_values if isinstance(x, (int, float))]  # todo 可能需要更泛用的判断数值的方法
+
+        self.raw_values = raw_values
+        values = values or []
+        self.values = sorted(values)
+        if self.raw_values:
+            self.raw_n = len(self.raw_values)
+        else:
+            self.raw_n = 0
+        self.n = len(values)
+        self.sum = sum(values)
+        if self.n:
+            self.mean = mean(self.values)
+            self.std = pstdev(self.values)
+            self.min, self.max = self.values[0], self.values[-1]
+        else:
+            self.mean = self.std = self.min = self.max = float('nan')
+
+        self.dist = []
+
+    def __len__(self):
+        return self.n
+
+    def summary(self, unit='万', precision=4, percentile=5):
+        """ 文本汇总性的报告
+
+        :param percentile: 包括两个极值端点的切分点数，
+            设置2，就是不设置分位数，就是只展示最小、最大值
+            如果设置了3，就表示"中位数、二分位数"，在展示的时候，会显示50%位置的分位数值
+            如果设置了5，就相当于"四分位数"，会显示25%、50%、75%位置的分位数值
+        :param unit: 展示数值时使用的单位
+        :param precision: 展示数值时的精度
+        """
+        from statistics import quantiles
+
+        def fmt0(v):
+            return human_readable_number(v, '万')
+
+        def fmt(v):
+            return human_readable_number(v, unit, precision)
+
+        desc = []
+        if self.raw_n and self.raw_n > self.n:
+            desc.append(f"非数值数量：{fmt0(self.raw_n - self.n)}")
+
+        desc.append(f"总数：{fmt0(self.n)}")
+        desc.append(f"总和：{fmt(self.sum)}")
+        desc.append(f'均值±标准差：{fmt(self.mean)}±{fmt(self.std)}')
+
+        if self.values:
+            dist = [self.values[0]]
+            if percentile > 2:
+                quartiles = quantiles(self.values, n=percentile - 1)
+                dist += quartiles
+            dist.append(self.values[-1])
+
+            desc.append(f"分布：{'/'.join([fmt(v) for v in dist])}")
+        elif self.dist:  # 此时当前函数的percentile设置失效，以dist中存储的情况为准
+            desc.append(f"分布：{'/'.join([fmt(v) for v in self.dist])}")
+
+        return '\t'.join(desc)
+
+
 class Groups:
     def __init__(self, data):
         """ 分组

{pyxllib-0.3.107 → pyxllib-0.3.108}/pyxllib/data/pglib.py

@@ -44,6 +44,7 @@ import psycopg.rows
 from pyxllib.prog.newbie import round_int
 from pyxllib.prog.pupil import utc_now, utc_timestamp
 from pyxllib.prog.specialist import XlOsEnv
+from pyxllib.algo.pupil import ValuesStat2
 from pyxllib.file.specialist import get_etag
 from pyxllib.data.sqlite import SqlBase
 
@@ -230,6 +231,76 @@ WHERE {table_name}.{item_id_name} = cte.{item_id_name}"""
 
         self.commit_base(commit, query, params)
 
+    def __6_高级统计(self):
+        pass
+
+    def get_field_valuesstat(self, table_name, field_name, percentile_count=5,
+                             cvt_numeric=True, by_data=False):
+        """ 获得指定表格的某个字段的统计特征ValuesStat2对象
+
+        :param table_name: 表名
+        :param field_name: 用于计算统计数据的字段名
+        :param percentile_count: 分位数的数量，例如 3 表示只计算中位数
+        :param by_data: 是否获得原始数据
+            默认只获得统计特征，不获得原始数据
+        """
+
+        def init_from_db_data():
+            values = self.exec2col(f'SELECT {field_name} FROM {table_name}')
+            if cvt_numeric:  # 这个是numeric格式字段，是"字符串"，要做个转换
+                values = [x and float(x) for x in values]
+
+            return ValuesStat2(raw_values=values)
+
+        def init_from_db():
+            # 构建基础的 SQL 查询
+            sql_query = f"""
+            SELECT 
+                COUNT(*) AS total_count,
+                COUNT({field_name}) AS non_null_count,
+                SUM({field_name}) AS total_sum,
+                AVG({field_name}) AS average,
+                STDDEV({field_name}) AS standard_deviation,
+                MIN({field_name}) AS min_value,
+                MAX({field_name}) AS max_value
+            """
+
+            percentiles = []
+            # 根据分位点的数量动态添加分位数计算
+            if percentile_count > 2:
+                step = 1 / (percentile_count - 1)
+                percentiles = [(i * step) for i in range(1, percentile_count - 1)]
+                percentiles_query = ", ".join(
+                    f"PERCENTILE_CONT({p:.2f}) WITHIN GROUP (ORDER BY {field_name}) AS percentile_{int(p * 100)}"
+                    for p in percentiles
+                )
+                sql_query += ", " + percentiles_query
+
+            sql_query += f" FROM {table_name};"
+
+            # 执行 SQL 查询
+            row = self.exec2dict(sql_query).fetchone()
+
+            # 创建一个新实例并将从数据库获得的数据填充到相应的属性
+            x = ValuesStat2()
+            x.raw_n = row['total_count']
+            x.n = row['non_null_count']
+            x.sum = row['total_sum']
+            x.mean = row['average']
+            x.std = row['standard_deviation']
+
+            # 如果计算了分位数，填充相应属性
+            x.dist = [row['min_value']] + [row[f"percentile_{int(p * 100)}"] for p in percentiles] + [row['max_value']]
+            if cvt_numeric:
+                x.dist = [float(x) for x in x.dist]
+
+            return x
+
+        if by_data:
+            return init_from_db_data()
+        else:
+            return init_from_db()
+
 
 """
 【关于为什么XlprDb要和pglib合一个文件】

{pyxllib-0.3.107 → pyxllib-0.3.108}/pyxllib/ext/kq5034lib.py

@@ -135,13 +135,14 @@ class Xiaoetong:
 
 
 class 网课考勤:
-    def __init__(self, today=None):
+    def __init__(self, today=None, *, 表格路径=None):
         self.返款标题 = ''
-        self.表格路径 = r'考勤.xlsx'
+        self.表格路径 = 表格路径 or r'考勤.xlsx'
         self.在线表格 = 'https://docs.qq.com/sheet/DUlF1UnRackJ2Vm5U'  # 生成日报用
         self.开课日期 = '2022-01-08'
         self.视频返款 = [20, 15, 10, 5, 0, 0]  # 直播(当堂)/第1天（当天）/第2天/第3天/第4天/第5天，完成观看的依次返款额。
         self.打卡返款 = {5: 100, 10: 150, 15: 200}  # 打卡满5/10/15次的返款额
+        self.课程链接 = []
         self._init(today)
 
         self.driver = None  # 浏览器脚本

{pyxllib-0.3.107 → pyxllib-0.3.108}/pyxllib/file/movielib.py

@@ -10,6 +10,7 @@ check_install_package('moviepy')
 
 import cv2
 from moviepy.editor import VideoFileClip
+from moviepy.editor import cvsecs
 import numpy as np
 from tqdm import tqdm
 
@@ -137,3 +138,11 @@ class XlVideoFileClip(VideoFileClip):
 
 
 inject_members(XlVideoFileClip, VideoFileClip)
+
+
+def clip_video(input_file, output_file, start_time, end_time):
+    start_time = cvsecs(start_time)
+    end_time = cvsecs(end_time)
+    clip = VideoFileClip(input_file).subclip(start_time, end_time)
+    clip.write_videofile(output_file)
+    clip.reader.close()

{pyxllib-0.3.107 → pyxllib-0.3.108}/pyxllib/prog/newbie.py

@@ -267,13 +267,17 @@ def human_readable_number(value, base_type='K', precision=4):
         'K': (['', 'K', 'M', 'G', 'T', 'P', 'E', 'Z', 'Y'], 1000),
         'KB': (['', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB', 'ZB', 'YB'], 1024),
         'KiB': (['', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'], 1024),
-        '万': (['', '万', '亿', '万亿', '亿亿'], 10000)
+        '万': (['', '万', '亿', '万亿', '亿亿'], 10000),
+        '秒': (['秒', [60, '分'], [60, '时'], [24, '天'], [7, '周'], [4.345, '月'], [12, '年']], 60),
     }.get(base_type, ([''], 1))  # 默认为空单位和基数1
 
     x, i = abs(value), 0
     while x >= base and i < len(units) - 1:
         x /= base
         i += 1
+        if isinstance(units[i], list):
+            base = units[i][0]
+            units[i] = units[i][1]
 
     x = f'{x:.{precision}g}'  # 四舍五入到指定精度
     prefix = '-' if value < 0 else ''  # 负数处理

{pyxllib-0.3.107 → pyxllib-0.3.108/pyxllib.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: pyxllib
-Version: 0.3.107
+Version: 0.3.108
 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

{pyxllib-0.3.107 → pyxllib-0.3.108}/setup.py

@@ -30,7 +30,7 @@ _dir = Path(__file__).parent
 
 setup(
     name='pyxllib',  # pip 安装时用的名字
-    version='0.3.107',  # 当前版本，每次更新上传到pypi都需要修改; 第4位版本号一般是修紧急bug
+    version='0.3.108',  # 当前版本，每次更新上传到pypi都需要修改; 第4位版本号一般是修紧急bug
     author='code4101',
     author_email='877362867@qq.com',
     url='https://github.com/XLPRUtils/pyxllib',

{pyxllib-0.3.107 → pyxllib-0.3.108}/pyxllib.egg-info/requires.txt

@@ -21,14 +21,14 @@ Jinja2
 pandas>=0.23.4
 
 [xlai]
-visdom
-opsdroid-get-image-size
 xlcocotools
-opencv-python
 pynvml
+opsdroid-get-image-size
+visdom
 pillow
+opencv-python
 
 [xlcv]
-opsdroid-get-image-size
-pillow
 opencv-python
+pillow
+opsdroid-get-image-size