add feature
This commit is contained in:
parent
a931d9ceb4
commit
aafe5e0195
6
case1.py
6
case1.py
@ -40,8 +40,8 @@ merged_df = pd.merge(
|
||||
|
||||
|
||||
# 检索出在某一天同时有多笔金额相同的记录
|
||||
duplicate_records = merged_df.groupby(['金额', '交易时间']).filter(lambda x: len(x) > 1)
|
||||
duplicate_records = merged_df.groupby(["金额", "交易时间"]).filter(lambda x: len(x) > 1)
|
||||
|
||||
# 将结果保存为新的Excel文件
|
||||
output_file_path = '/home/baol/tools/case2_records.xlsx' # 替换为你希望保存的文件路径
|
||||
duplicate_records.to_excel(output_file_path, index=False, engine='openpyxl')
|
||||
output_file_path = "/home/baol/tools/case2_records.xlsx" # 替换为你希望保存的文件路径
|
||||
duplicate_records.to_excel(output_file_path, index=False, engine="openpyxl")
|
||||
|
||||
99
case_int.py
Normal file
99
case_int.py
Normal file
@ -0,0 +1,99 @@
|
||||
import pandas as pd
|
||||
|
||||
# 1. 读取 Excel 文件中的数据
|
||||
df_case = pd.read_excel(
|
||||
"/home/baol/tools/数据草稿-案款1.xls", sheet_name="一案一账户收"
|
||||
)
|
||||
df_bank = pd.read_excel(
|
||||
"/home/baol/tools/数据草稿-案款1.xls", sheet_name="银行-收入", header=1
|
||||
)
|
||||
|
||||
# 2. 保留指定的列
|
||||
# 在“一案一账户收”中，保留归属本案金额、到账金额、到账日期、缴款人、案号
|
||||
df_case = df_case[["归属本案金额", "到账金额", "到账日期", "缴款人", "案号"]]
|
||||
# 在“银行-收入”中，保留对方户名、交易时间、金额
|
||||
df_bank = df_bank[["对方户名", "交易时间", "金额"]]
|
||||
|
||||
# 3. 统一日期格式为 yyyy-mm-dd
|
||||
df_case["到账日期"] = pd.to_datetime(df_case["到账日期"]).dt.strftime("%Y-%m-%d")
|
||||
df_bank["交易时间"] = pd.to_datetime(df_bank["交易时间"]).dt.strftime("%Y-%m-%d")
|
||||
|
||||
# 4. 任务1：检索到账金额与金额匹配，且到账日期与交易时间匹配的记录，但排除在同一天有多笔金额相同的情况
|
||||
# 合并两个数据框，匹配到账金额=金额，到账日期=交易时间
|
||||
df_merged = pd.merge(
|
||||
df_case,
|
||||
df_bank,
|
||||
left_on=["到账金额", "到账日期"],
|
||||
right_on=["金额", "交易时间"],
|
||||
how="inner",
|
||||
)
|
||||
|
||||
# 计算每个日期和金额组合的出现次数
|
||||
case_counts = (
|
||||
df_case.groupby(["到账日期", "到账金额"]).size().reset_index(name="case_count")
|
||||
)
|
||||
bank_counts = (
|
||||
df_bank.groupby(["交易时间", "金额"]).size().reset_index(name="bank_count")
|
||||
)
|
||||
|
||||
# 合并统计数据
|
||||
df_merged = pd.merge(
|
||||
df_merged,
|
||||
case_counts,
|
||||
left_on=["到账日期", "到账金额"],
|
||||
right_on=["到账日期", "到账金额"],
|
||||
)
|
||||
df_merged = pd.merge(
|
||||
df_merged, bank_counts, left_on=["交易时间", "金额"], right_on=["交易时间", "金额"]
|
||||
)
|
||||
|
||||
# 过滤出在同一天没有多笔金额相同的记录（即计数为1）
|
||||
task1_result = df_merged[
|
||||
(df_merged["case_count"] == 1) & (df_merged["bank_count"] == 1)
|
||||
]
|
||||
|
||||
# 获取任务1中“一案一账户收”和“银行-收入”的索引
|
||||
task1_case_indices = df_case[
|
||||
df_case[["到账金额", "到账日期"]]
|
||||
.apply(tuple, axis=1)
|
||||
.isin(task1_result[["到账金额", "到账日期"]].apply(tuple, axis=1))
|
||||
].index
|
||||
|
||||
task1_bank_indices = df_bank[
|
||||
df_bank[["金额", "交易时间"]]
|
||||
.apply(tuple, axis=1)
|
||||
.isin(task1_result[["金额", "交易时间"]].apply(tuple, axis=1))
|
||||
].index
|
||||
|
||||
# 5. 任务2：检索出在某一天有多笔金额相同的记录
|
||||
# 找出“一案一账户收”中同一天金额相同且次数大于1的记录
|
||||
task2_case = df_case.groupby(["到账日期", "到账金额"]).filter(lambda x: len(x) > 1)
|
||||
task2_case_indices = task2_case.index
|
||||
|
||||
# 找出“银行-收入”中同一天金额相同且次数大于1的记录
|
||||
task2_bank = df_bank.groupby(["交易时间", "金额"]).filter(lambda x: len(x) > 1)
|
||||
task2_bank_indices = task2_bank.index
|
||||
|
||||
# 6. 任务3：检索出不符合任务1和任务2的记录
|
||||
|
||||
# 在“一案一账户收”中，获取不在任务1和任务2中的索引
|
||||
task3_case_indices = df_case.index.difference(
|
||||
task1_case_indices.union(task2_case_indices)
|
||||
)
|
||||
task3_case_result = df_case.loc[task3_case_indices]
|
||||
|
||||
# 在“银行-收入”中，获取不在任务1和任务2中的索引
|
||||
task3_bank_indices = df_bank.index.difference(
|
||||
task1_bank_indices.union(task2_bank_indices)
|
||||
)
|
||||
task3_bank_result = df_bank.loc[task3_bank_indices]
|
||||
|
||||
# 7. 将结果保存到 Excel 文件
|
||||
with pd.ExcelWriter("/home/baol/tools/任务结果.xlsx") as writer:
|
||||
task1_result.to_excel(writer, sheet_name="任务1", index=False)
|
||||
task2_case.to_excel(writer, sheet_name="任务2_一案一账户收", index=False)
|
||||
task2_bank.to_excel(writer, sheet_name="任务2_银行收入", index=False)
|
||||
task3_case_result.to_excel(writer, sheet_name="任务3_一案一账户收", index=False)
|
||||
task3_bank_result.to_excel(writer, sheet_name="任务3_银行收入", index=False)
|
||||
|
||||
print("所有任务结果已保存到 '任务结果.xlsx' 文件中。")
|
||||
104
case_out.py
Normal file
104
case_out.py
Normal file
@ -0,0 +1,104 @@
|
||||
import pandas as pd
|
||||
|
||||
# 读取Excel文件的指定工作表和列
|
||||
df1 = pd.read_excel(
|
||||
"/home/baol/tools/数据草稿-案款1.xls",
|
||||
sheet_name="一案一账户支",
|
||||
usecols=["支款案号", "来源案号", "申请人", "支付日期", "支付金额", "领款人"],
|
||||
)
|
||||
df2 = pd.read_excel(
|
||||
"/home/baol/tools/数据草稿-案款1.xls",
|
||||
sheet_name="银行支出",
|
||||
usecols=["对方户名", "交易时间", "支出金额"],
|
||||
header=1,
|
||||
)
|
||||
|
||||
# 标准化日期格式为 yyyy-mm-dd
|
||||
df1["支付日期"] = pd.to_datetime(df1["支付日期"]).dt.strftime("%Y-%m-%d")
|
||||
df2["交易时间"] = pd.to_datetime(df2["交易时间"]).dt.strftime("%Y-%m-%d")
|
||||
|
||||
# 确保金额为数值类型
|
||||
df1["支付金额"] = pd.to_numeric(df1["支付金额"], errors="coerce")
|
||||
df2["支出金额"] = pd.to_numeric(df2["支出金额"], errors="coerce")
|
||||
|
||||
# 初始化任务标记列
|
||||
df1["任务"] = "任务3"
|
||||
df2["任务"] = "任务3"
|
||||
|
||||
# 任务2：查找在某一天有多笔金额相同的记录
|
||||
# 标记df1中的重复记录
|
||||
df1_duplicates = df1[df1.duplicated(subset=["支付日期", "支付金额"], keep=False)]
|
||||
df1.loc[df1_duplicates.index, "任务"] = "任务2"
|
||||
|
||||
# 标记df2中的重复记录
|
||||
df2_duplicates = df2[df2.duplicated(subset=["交易时间", "支出金额"], keep=False)]
|
||||
df2.loc[df2_duplicates.index, "任务"] = "任务2"
|
||||
|
||||
# 为匹配准备唯一记录（任务1）
|
||||
df1_unique = (
|
||||
df1[df1["任务"] == "任务3"].reset_index().rename(columns={"index": "df1_index"})
|
||||
)
|
||||
df2_unique = (
|
||||
df2[df2["任务"] == "任务3"].reset_index().rename(columns={"index": "df2_index"})
|
||||
)
|
||||
|
||||
# 创建辅助列用于匹配
|
||||
df1_unique["日期"] = df1_unique["支付日期"]
|
||||
df1_unique["金额"] = df1_unique["支付金额"]
|
||||
df1_unique["名称"] = df1_unique["领款人"]
|
||||
|
||||
df2_unique["日期"] = df2_unique["交易时间"]
|
||||
df2_unique["金额"] = df2_unique["支出金额"]
|
||||
df2_unique["名称"] = df2_unique["对方户名"]
|
||||
|
||||
# 条件1：支付金额与支出金额匹配，支付日期与交易时间匹配
|
||||
condition1_matches = pd.merge(
|
||||
df1_unique, df2_unique, on=["日期", "金额"], how="inner", suffixes=("_df1", "_df2")
|
||||
)
|
||||
df1.loc[condition1_matches["df1_index"], "任务"] = "任务1"
|
||||
df2.loc[condition1_matches["df2_index"], "任务"] = "任务1"
|
||||
|
||||
# 条件2：支付金额与支出金额匹配，领款人与对方户名匹配
|
||||
condition2_matches = pd.merge(
|
||||
df1_unique, df2_unique, on=["金额", "名称"], how="inner", suffixes=("_df1", "_df2")
|
||||
)
|
||||
df1.loc[condition2_matches["df1_index"], "任务"] = "任务1"
|
||||
df2.loc[condition2_matches["df2_index"], "任务"] = "任务1"
|
||||
|
||||
# 合并条件匹配的结果
|
||||
task1_matches = pd.concat([condition1_matches, condition2_matches], ignore_index=True)
|
||||
task1_matches = task1_matches.drop_duplicates(subset=["df1_index", "df2_index"])
|
||||
|
||||
# 任务3：提取不符合任务1和任务2的记录
|
||||
task3_df1 = df1[df1["任务"] == "任务3"]
|
||||
task3_df2 = df2[df2["任务"] == "任务3"]
|
||||
|
||||
# 创建一个新的Excel文件，保存结果
|
||||
with pd.ExcelWriter("/home/baol/tools/匹配结果.xlsx") as writer:
|
||||
# 任务1结果
|
||||
task1_result = task1_matches[
|
||||
[
|
||||
"支款案号",
|
||||
"来源案号",
|
||||
"申请人",
|
||||
"支付日期",
|
||||
"支付金额",
|
||||
"领款人",
|
||||
"对方户名",
|
||||
"交易时间",
|
||||
"支出金额",
|
||||
]
|
||||
]
|
||||
task1_result.to_excel(writer, sheet_name="任务1匹配结果", index=False)
|
||||
|
||||
# 任务2结果
|
||||
df1_duplicates.to_excel(
|
||||
writer, sheet_name="任务2_一案一账户支重复记录", index=False
|
||||
)
|
||||
df2_duplicates.to_excel(writer, sheet_name="任务2_银行支出重复记录", index=False)
|
||||
|
||||
# 任务3结果
|
||||
task3_df1.to_excel(writer, sheet_name="任务3_一案一账户支未匹配记录", index=False)
|
||||
task3_df2.to_excel(writer, sheet_name="任务3_支出未匹配记录", index=False)
|
||||
|
||||
print("结果已保存到 '匹配结果.xlsx' 文件中。")
|
||||
Loading…
x
Reference in New Issue
Block a user