你能过滤熊猫数据框来获取所有包含特定模式的行，就像我对正则表达式所做的那样

Question

我是 python 新手，我正在尝试编写一个管道来捕获在同一行中具有“R”和“yes”的所有 ID。

我有多个包含以下模式的 csv 文件：

CSV文件1

"ID1","R","","","","","yes"
"ID2","S","","","","","yes"

CSVFile2

"ID1","animal","R","","","","","","","","yes"

CSV文件3

"ID1","animal","fish","S","","","","","","","yes"
"ID2","animal","dog","R","","","","","","","yes"

使用正则表达式，我可以用 R 捕获所有行，是的：

input = open(sys.argv[1], 'r')

for line in input:
    if re.match(r'^(?=.*(\bR\b)+)(?=.*(\byes\b)+)', line) is not None:
        print (line)

但是，如果我将其转换为 pandas 数据框，有没有一种方法可以捕获所有具有 R 和 yes 的行，而不必每次都指定该列名，因为并非所有文件都具有相同的列数？

Answer 1

您可以使用布尔索引：

print(df[df.eq("R").any(1) & df.eq("yes").any(1)])

输出 (CSVFile1)

     0  1   2   3   4   5    6
0  ID1  R NaN NaN NaN NaN  yes

---

print(df1[df1.eq("R").any(1) & df1.eq("yes").any(1)])

输出 (CSVFile2)

    0       1    2  3   4   5   6   7   8   9    10
1  ID2  animal  dog  R NaN NaN NaN NaN NaN NaN  yes