如何将列表中的所有单词和短语放入搜索表达式（Python）

Question

I have this list of lists:

groups = [['|FOOD|','shrimps','chicken wok','bowl of rice'],['|DRINK|','water','cranberry juice','tea']]

I'm trying to get the output to be:

[['|FOOD|',
  '[lemma="shrimps"]',
  '[lemma="chicken"][lemma="wok"]',
  '[lemma="bowl"][lemma="of"][lemma="rice"]'],
 ['|DRINK|',
  '[lemma="water"]',
  '[lemma="cranberry"][lemma="juice"]',
  '[lemma="tea"]']]

So, basically I need every word lemmatized for a corpus search. Some words though, are not words but phrases . I've only yet figured out the code for single words , here it is:

import re
groups = [[f'[lemma="{word}"]' if not ' ' in word and not re.search(r'\|.*\|', word) else word for word in group] for group in groups]

This returns groups as:

[['|FOOD|', 
  '[lemma="shrimps"]', 
  'chicken wok', 
  'bowl of rice'],
 ['|DRINK|', 
  '[lemma="water"]', 
  'cranberry juice', 
  '[lemma="tea"]']]

So I made it not include that words containing a whitespace ( phrases ), plus the topic words. What then is the code to deal with these phrases and have them look like like I typed above?

I'm a beginner, so if you know a better way to organise all this data, let me know.

Answer 1

You do not really need a regex here, you may use if not word.startswith("|") and not word.endswith("|") to check if the entry has no pipes on both ends:

groups = [[''.join([r"""[lemma="{}"]""".format(w) for w in word.split()]) if not word.startswith("|") and not word.endswith("|") else word for word in group] for group in groups]

See the Python demo online . Output:

[['|FOOD|', 
 '[lemma="shrimps"]', 
 '[lemma="chicken"][lemma="wok"]', 
 '[lemma="bowl"][lemma="of"][lemma="rice"]'], 
['|DRINK|', 
 '[lemma="water"]', 
 '[lemma="cranberry"][lemma="juice"]', 
 '[lemma="tea"]']
 ]