import pandas as pd
from typing import List
from openpyxl import Workbook

def strip_character(column_name, characters: List[str]):

    new_col_name = column_name
    for character in characters:
        new_col_name = new_col_name.replace(character, '')
    new_col_name = new_col_name.strip()
    return new_col_name

def refine_content(df):
    strip_character_list = [' ', '\n', ':', '：','其他']
    for col in df.columns:
        df[col] = df[col].apply(lambda x: "其他" if isinstance(x, str) and strip_character(x, strip_character_list) == "" 
                                else (strip_character(x, strip_character_list) if isinstance(x, str) else x))
    return df

def calc_acv_mean(df, acv_name, group_by_column):
    df_grouped_mean = df.groupby(group_by_column)[acv_name].mean().fillna(0).astype(int).reset_index()
    df_grouped_mean = df_grouped_mean.sort_values(by=acv_name, ascending=False)
    df_grouped_mean[acv_name] = df_grouped_mean[acv_name].apply(lambda x: '{:,}'.format(x))
    return df_grouped_mean

def calc_acv_sum(df, acv_name, group_by_column):
    df_grouped_sum = df.groupby(group_by_column)[acv_name].sum().astype(int).reset_index()
    df_grouped_sum = df_grouped_sum.sort_values(by=acv_name, ascending=False)
    df_grouped_sum[acv_name] = df_grouped_sum[acv_name].apply(lambda x: '{:,}'.format(x))
    return df_grouped_sum

def calc_acv_sum_and_count(df, acv_name, group_by_column):
    df_grouped_sum = df.groupby(group_by_column)[acv_name].sum().astype(int).reset_index()
    df_grouped_count = df.groupby(group_by_column)[acv_name].count().reset_index()
    df_grouped_sum_count = pd.merge(df_grouped_sum, df_grouped_count, on=group_by_column, how='outer')
    return df_grouped_sum_count

def save_to_excel(dataframes, sheet_names, output_file):
    with pd.ExcelWriter(output_file, engine='openpyxl') as writer:
        for df, sheet_name in zip(dataframes, sheet_names):
            df.to_excel(writer, sheet_name=sheet_name, index=False)

# 读取赢单Excel文件
df_win = pd.read_excel('./data_src/pingcap_won.xlsx')
acv_name = 'ACV'

# 创建一个列表来存储所有的数据帧和对应的sheet名称
dataframes = []
sheet_names = []

# ACV by 客户分类
df_win_grouped_by_industry_sum = calc_acv_sum_and_count(df_win, acv_name, '客户分类')
dataframes.append(refine_content(df_win_grouped_by_industry_sum))
sheet_names.append("ACV by 行业")

# Group by customer industry and calculate the average ACV for each group
df_win_grouped_by_industry_mean = calc_acv_mean(df_win, acv_name, '客户分类')
dataframes.append(refine_content(df_win_grouped_by_industry_mean))
sheet_names.append("平均ACV by 行业")

df_win_grouped_by_sub_industry_sum = calc_acv_sum_and_count(df_win, acv_name, '客户行业')
dataframes.append(refine_content(df_win_grouped_by_sub_industry_sum))
sheet_names.append("ACV by 子行业")

df_win_grouped_by_sub_industry_mean = calc_acv_mean(df_win, acv_name, '客户行业')
dataframes.append(refine_content(df_win_grouped_by_sub_industry_mean))
sheet_names.append("平均ACV by 子行业")

# ���取Excel文件
df = pd.read_excel('./data_src/pingcap_pipeline.xlsx')

# 按照"客户分类"列分组，并计算ACV列的和
acv_name = '预估 ACV'

df_pipeline_grouped_by_industry_sum = calc_acv_sum_and_count(df, acv_name, '负责人所属行业')
dataframes.append(refine_content(df_pipeline_grouped_by_industry_sum))
sheet_names.append("预估ACV by 行业")

df_pipeline_grouped_by_industry_mean = calc_acv_mean(df, acv_name, '负责人所属行业')
dataframes.append(refine_content(df_pipeline_grouped_by_industry_mean))
sheet_names.append("平均预估ACV by 行业")

df_pipeline_grouped_by_sub_industry_sum = calc_acv_sum_and_count(df, acv_name, '客户行业')
dataframes.append(refine_content(df_pipeline_grouped_by_sub_industry_sum))
sheet_names.append("预估ACV by 子行业")

df_pipeline_grouped_by_sub_industry_mean = calc_acv_mean(df, acv_name, '客户行业')
dataframes.append(refine_content(df_pipeline_grouped_by_sub_industry_mean))
sheet_names.append("平均预估ACV by 子行业")

# 保存所有数据帧到一个Excel文件中
output_file = './output/acv_analysis_v2(with count).xlsx'
save_to_excel(dataframes, sheet_names, output_file)

print(f"Analysis results have been saved to {output_file}")