按照在原始句子中出现的顺序提取所有匹配的子字符串

Question

I have a list that stores the defined keywords, For example:

keywords = [
    "white shark",
    "tiger shark",
    "funnel web spider",
    "inland taipan"]

Now I made a sentence

str = "A tiger shark spotted here, and a white shark, and a funnel web spider"

From this sentence, I want to produce a result ["tiger shark", "white shark", "funnel web spider"] . The keywords appeared in the sentence with their original order as they are in the sentence. Then, I made a code like this

for i in keywords:
    if not str.find(i) == -1:
          result.append(i)

This is gonna give me ["white shark", "tiger shark", "funnel web spider"] , The order is different than my expected result. And my mistake is quite obvious.

So my question is how to achieve the correct order for the result. I think the only way is to use the string to traverse the keyword list. But I think it's quite complicated as it involves many combinations. Any helps? Thank you so much.

Answer 1

It's because you're appending them in order of keywords , instead we should save their indexes of appearance in my_str , to later order our words depending on their occurrence in my_str

keywords = [
    "white shark",
    "tiger shark",
    "funnel web spider",
    "inland taipan"]
my_str = "A tiger shark spotted here, and a white shark, and a funnel web spider"

result = []
for keyword in keywords:
    idx = my_str.find(keyword)
    if idx != -1:
        result.append((idx, keyword))

result = [i[1] for i in sorted(result)]  # Sorts by first item in tuple, idx

print(result) # -> ['tiger shark', 'white shark', 'funnel web spider']

Answer 2

You could build a list of (index,keyword) and sort it. Then extract keywords from the matching tuples.

keywords = [
"white shark",
"tiger shark",
"funnel web spider",
"inland taipan"]
sentence = "A tiger shark spotted here, and a white shark, and a funnel web spider"

result = [ k for i,k in sorted( (sentence.find(k),k) for k in keywords) if i != -1 ]

print(result)
# ['tiger shark', 'white shark', 'funnel web spider']

You could also use a regular expression (form the re module):

import re

result  = re.findall("|".join(keywords),sentence)