通过匹配 python 中的其他列值来拆分列值

Question

我有一个如下所示的数据框

dummy = [["new york is a cool city in usa but i like london","cool","new","like"]]

df_dummy = pd.DataFrame(dummy,columns=["text","a","b","c"])

现在我想拆分其他列值匹配的文本列的字符串。

我试过下面的代码，但不能超出这个范围。

idx_ = [0]

cols_dummy = df_dummy.columns.values
cols_dummy = np.delete(cols_dummy,idx_,axis=0)
t_text = df_dummy.text.values[0]
for i in cols_dummy:
    match_ = "("+df_dummy[i].values[0]+")"

    tmp = re.split(match_,t_text)

    for e in range (0,len(tmp)):
        print(match_, tmp)

预期输出： ["new", "york is a", "cool", "city in usa but i", "like", "london"]

Answer 1

不是最漂亮的，但可以胜任。希望它能激励你实现一些东西similar/better。


t = "new york is a cool city in usa but i like london"
words = ["cool", "new", "like"]


def get_indicies(s, words):
    indicies = []
    for word in words:
        start = s.index(word)
        end = start + len(word)

        idx_tupl = (start, end)
        indicies.append(idx_tupl)
    return sorted(indicies)


def compose(s, indicies):
    result = []
    ptr = 0  # start at beginning of string
    for idx in indicies:

        if idx[0] != ptr:
            result.append(s[ptr : idx[0]])

        result.append(s[idx[0] : idx[1]])

        ptr = idx[1]

    result.append(s[ptr:])
    return result


def split_by_word(s, words):
    indices = get_indicies(s, words)
    return compose(s, indices)


print(split_by_word(t, words))
# output (Notice the whitespace around some string here)
['new', ' york is a ', 'cool', ' city in usa but i ', 'like', ' london']

通过匹配 python 中的其他列值来拆分列值

split a column value by matching other column values in python

python

pandas

python-re