如何循环遍历数据框中的行，从一个特定值到 python 中的另一个值？

Question

我有以下数据框：

Time	Tab	User	Description
27.10.2021 15:58:00	Tab Alpha	UserA@gmail.com	Tab Alpha of type PARTSTUDIO opened by User A
27.10.2021 15:59:00	Tab Alpha	UserA@gmail.com	Start edit of part studio feature
27.10.2021 15:59:00	Tab Alpha	UserA@gmail.com	Cancel Operation
27.10.2021 15:59:00	Tab Alpha	UserB@gmail.com	Tab Alpha of type PARTSTUDIO opened by User B
27.10.2021 15:59:00	Tab Alpha	UserB@gmail.com	Start edit of part studio feature
27.10.2021 16:03:00	Tab Alpha	UserB@gmail.com	Cancel Operation
27.10.2021 16:03:00	Tab Alpha	UserA@gmail.com	Add assembly feature
27.10.2021 16:03:00	Tab Alpha	UserA@gmail.com	Tab Alpha of type PARTSTUDIO closed by User A
27.10.2021 16:03:00	Tab Beta	UserA@gmail.com	Tab Beta of type PARTSTUDIO opened by User A
27.10.2021 16:15:00	Tab Beta	UserA@gmail.com	Start edit of part studio feature
27.10.2021 16:15:00	Tab Alpha	UserB@gmail.com	Start edit of part studio feature
27.10.2021 16:15:00	Tab Alpha	UserB@gmail.com	Tab Alpha of type PARTSTUDIO closed by User B
27.10.2021 16:17:00	Tab Beta	UserA@gmail.com	Add assembly feature
27.10.2021 16:17:00	Tab Beta	UserC@gmail.com	Tab Beta of type ASSEMBLY opened by User C
27.10.2021 16:17:00	Tab Beta	UserC@gmail.com	Add assembly feature
27.10.2021 16:17:00	Tab Delta	UserB@gmail.com	Tab Delta of type PARTSTUDIO opened by User B
27.10.2021 16:54:00	Tab Delta	UserB@gmail.com	Add assembly feature
27.10.2021 16:55:00	Tab Beta	UserA@gmail.com	Tab Beta of type PARTSTUDIO closed by User A
27.10.2021 16:55:00	Tab Delta	UserB@gmail.com	Start edit of part studio feature
27.10.2021 16:55:00	Tab Delta	UserB@gmail.com	Tab Delta of type PARTSTUDIO closed by User B
27.10.2021 16:59:00	Tab Delta	UserC@gmail.com	Add assembly feature

如何根据“描述”列中的条件遍历数据框中的行？条件是在值“Tab 'Tab_name' of type ... opened by User B”和“Tab 'Tab_name' of type 之间提取所述值（选项卡名称） ... 已被用户 B 关闭”到“用户 B”列。

预期输出：

Time	Tab	User	Description	UserB
27.10.2021 15:58:00	Tab Alpha	UserA@gmail.com	Tab Alpha of type PARTSTUDIO opened by User A
27.10.2021 15:59:00	Tab Alpha	UserA@gmail.com	Start edit of part studio feature
27.10.2021 15:59:00	Tab Alpha	UserA@gmail.com	Cancel Operation
27.10.2021 15:59:00	Tab Alpha	UserB@gmail.com	Tab Alpha of type PARTSTUDIO opened by User B
27.10.2021 15:59:00	Tab Alpha	UserB@gmail.com	Start edit of part studio feature	Tab Alpha
27.10.2021 16:03:00	Tab Alpha	UserB@gmail.com	Cancel Operation	Tab Alpha
27.10.2021 16:03:00	Tab Alpha	UserA@gmail.com	Add assembly feature	Tab Alpha
27.10.2021 16:03:00	Tab Alpha	UserA@gmail.com	Tab Alpha of type PARTSTUDIO closed by User A	Tab Alpha
27.10.2021 16:03:00	Tab Beta	UserA@gmail.com	Tab Beta of type PARTSTUDIO opened by User A	Tab Alpha
27.10.2021 16:15:00	Tab Beta	UserA@gmail.com	Start edit of part studio feature	Tab Alpha
27.10.2021 16:15:00	Tab Alpha	UserB@gmail.com	Start edit of part studio feature	Tab Alpha
27.10.2021 16:15:00	Tab Alpha	UserB@gmail.com	Tab Alpha of type PARTSTUDIO closed by User B
27.10.2021 16:17:00	Tab Beta	UserA@gmail.com	Add assembly feature
27.10.2021 16:17:00	Tab Beta	UserC@gmail.com	Tab Beta of type ASSEMBLY opened by User C
27.10.2021 16:17:00	Tab Beta	UserC@gmail.com	Add assembly feature
27.10.2021 16:17:00	Tab Delta	UserB@gmail.com	Tab Delta of type PARTSTUDIO opened by User B
27.10.2021 16:54:00	Tab Delta	UserB@gmail.com	Add assembly feature	Tab Delta
27.10.2021 16:55:00	Tab Beta	UserA@gmail.com	Tab Beta of type PARTSTUDIO closed by User A	Tab Delta
27.10.2021 16:55:00	Tab Delta	UserB@gmail.com	Start edit of part studio feature	Tab Delta
27.10.2021 16:55:00	Tab Delta	UserB@gmail.com	Tab Delta of type PARTSTUDIO closed by User B
27.10.2021 16:59:00	Tab Beta	UserC@gmail.com	Add assembly feature

这是我的尝试：

df.insert(4, 'User B', '0')

for index, row in df.iterrows():
    if row['Description'].find('opened by User B') != -1:
        tab = row['Description'].rpartition(' of type')[0]

    if row['Description'].find('closed by User B') == -1:
        df.at[index, 'User B'] = tab

当前代码从“描述”列不包含“由用户 B 关闭”的每一行中提取选项卡名称。它应该只从问题前面指定的范围内的行中提取 - 在“描述”列中的值==“......由用户B打开”和“......由用户B关闭”的行之间。我知道不推荐 df.iterrows，但我似乎找不到其他方法。

Answer 1

尝试查找所有“已打开”和“已关闭”的描述并选择它们之间的切片：

opens = df["Description"].str.contains(".*Tab.*of type.*opened by User B")
closes = df["Description"].str.contains(".*Tab.*of type.*closed by User B")

df["UserB"] = df["Tab"].where((opens.cumsum().shift().fillna(0)-closes.cumsum()).astype(bool))

>>> df
                   Time  ...      UserB
0   27.10.2021 15:58:00  ...        NaN
1   27.10.2021 15:59:00  ...        NaN
2   27.10.2021 15:59:00  ...        NaN
3   27.10.2021 15:59:00  ...        NaN
4   27.10.2021 15:59:00  ...  Tab Alpha
5   27.10.2021 16:03:00  ...  Tab Alpha
6   27.10.2021 16:03:00  ...  Tab Alpha
7   27.10.2021 16:03:00  ...  Tab Alpha
8   27.10.2021 16:03:00  ...   Tab Beta
9   27.10.2021 16:15:00  ...   Tab Beta
10  27.10.2021 16:15:00  ...  Tab Alpha
11  27.10.2021 16:15:00  ...        NaN
12  27.10.2021 16:17:00  ...        NaN
13  27.10.2021 16:17:00  ...        NaN
14  27.10.2021 16:17:00  ...        NaN
15  27.10.2021 16:17:00  ...        NaN
16  27.10.2021 16:54:00  ...  Tab Delta
17  27.10.2021 16:55:00  ...   Tab Beta
18  27.10.2021 16:55:00  ...  Tab Delta
19  27.10.2021 16:55:00  ...        NaN
20  27.10.2021 16:59:00  ...        NaN

[21 rows x 5 columns]

Answer 2

另一个解决方案：

df["User B"] = df["Description"].str.contains("opened by User B").shift()
df["User B"] += df["Description"].str.contains("closed by User B") * -1

df["tmp"] = (
    df["Description"]
    .str.extract("(Tab .*?) of type .* opened by User B")
    .ffill()
)

df["User B"] = np.where(df["User B"].fillna(0).cumsum(), df["tmp"], "")
df = df.drop(columns="tmp")

print(df.to_markdown())

打印：

	Time	Tab	User	Description	User B
0	27.10.2021 15:58:00	Tab Alpha	UserA@gmail.com	Tab Alpha of type PARTSTUDIO opened by User A
1	27.10.2021 15:59:00	Tab Alpha	UserA@gmail.com	Start edit of part studio feature
2	27.10.2021 15:59:00	Tab Alpha	UserA@gmail.com	Cancel Operation
3	27.10.2021 15:59:00	Tab Alpha	UserB@gmail.com	Tab Alpha of type PARTSTUDIO opened by User B
4	27.10.2021 15:59:00	Tab Alpha	UserB@gmail.com	Start edit of part studio feature	Tab Alpha
5	27.10.2021 16:03:00	Tab Alpha	UserB@gmail.com	Cancel Operation	Tab Alpha
6	27.10.2021 16:03:00	Tab Alpha	UserA@gmail.com	Add assembly feature	Tab Alpha
7	27.10.2021 16:03:00	Tab Alpha	UserA@gmail.com	Tab Alpha of type PARTSTUDIO closed by User A	Tab Alpha
8	27.10.2021 16:03:00	Tab Beta	UserA@gmail.com	Tab Beta of type PARTSTUDIO opened by User A	Tab Alpha
9	27.10.2021 16:15:00	Tab Beta	UserA@gmail.com	Start edit of part studio feature	Tab Alpha
10	27.10.2021 16:15:00	Tab Alpha	UserB@gmail.com	Start edit of part studio feature	Tab Alpha
11	27.10.2021 16:15:00	Tab Alpha	UserB@gmail.com	Tab Alpha of type PARTSTUDIO closed by User B
12	27.10.2021 16:17:00	Tab Beta	UserA@gmail.com	Add assembly feature
13	27.10.2021 16:17:00	Tab Beta	UserC@gmail.com	Tab Beta of type ASSEMBLY opened by User C
14	27.10.2021 16:17:00	Tab Beta	UserC@gmail.com	Add assembly feature
15	27.10.2021 16:17:00	Tab Delta	UserB@gmail.com	Tab Delta of type PARTSTUDIO opened by User B
16	27.10.2021 16:54:00	Tab Delta	UserB@gmail.com	Add assembly feature	Tab Delta
17	27.10.2021 16:55:00	Tab Beta	UserA@gmail.com	Tab Beta of type PARTSTUDIO closed by User A	Tab Delta
18	27.10.2021 16:55:00	Tab Delta	UserB@gmail.com	Start edit of part studio feature	Tab Delta
19	27.10.2021 16:55:00	Tab Delta	UserB@gmail.com	Tab Delta of type PARTSTUDIO closed by User B
20	27.10.2021 16:59:00	Tab Delta	UserC@gmail.com	Add assembly feature

如何循环遍历数据框中的行，从一个特定值到 python 中的另一个值？

How to loop through rows in dataframe from one specific value until another in python?

python

loops

dataframe

python-3.x