在特定文本正则表达式 python 之后捕获所有出现的 substring

Question

I have a long document in which the line of my interest starts with Categories: . I want to find all words separated by , after Categories: . Here's an example line

Categories : Turbo Prop , Very Light , Light , Mid Size

I want to find start index and end index of Turbo Prop , Very Light , Light , Mid Size

I am using following code

regex_pattern = r"(?<=Categories : )([A-Za-z ]+(?:,)?)+"

matched_text = regex.search(regex_pattern,doc_tex)

But matched_text.groups() is only giving Mid Size . In short, I want to find all occurences of group 1 after Categories .

Answer 1

Do it in two steps. First split the line using : , then split the second part using , .

category_string = line.split(':')[1]
categories = category_string.split(',')

Answer 2

It looks like the comments answered the OP's question, but for completeness I thought I'd post the answer they discuss. It looks like Python's re module does not store all all instances of a repeated capture group; see issue 7132 . The regex package, however, adds additional methods to handle repeated capture groups, including.

• captures -Returns a list of the strings matched in a group or groups.
• starts - Returns a list of the start positions.
• ends - Returns a list of the end positions.
• spans - Returns a list of the spans. Compare with matchobject.span([group]).

Hence, using the regex package with the matchedobject.starts and matchedobject.ends methods should work.

Answer 3

As you are using the PyPi regex module , you can get all captures per group, together with their start and end indices, using

import regex
text = "Categories : Turbo Prop , Very Light , Light , Mid Size"
regex_pattern = r"Categories\s*:(?:\s*([A-Za-z ]+)\b(?:\s*,)?)+"
m = regex.search(regex_pattern, text)
result = list(zip(m.captures(1),m.starts(1),m.ends(1)))
print(result) 
# => [('Turbo Prop', 13, 23), ('Very Light', 26, 36), ('Light', 39, 44), ('Mid Size', 47, 55)]

See the Python demo

More details from PyPi regex documentation:

A match object has additional methods which return information on all the successful matches of a repeated capture group. These methods are:

• matchobject.captures([group1, ...])
  • Returns a list of the strings matched in a group or groups. Compare with matchobject.group([group1, ...]) .
• matchobject.starts([group])
  • Returns a list of the start positions. Compare with matchobject.start([group]) .
• matchobject.ends([group])
  • Returns a list of the end positions. Compare with matchobject.end([group]) .
• matchobject.spans([group])
  • Returns a list of the spans. Compare with matchobject.span([group]) .

Note I had to revamp your regex a bit:

• Categories\s*: - matches Categories , zero or more whitespaces, :
• (?:\s*([A-Za-z ]+)\b(?:\s*,)?)+ - one or more repetitions of
  • \s* - zero or more whitespace chars
  • ([A-Za-z ]+) - one or more ASCII letters or spaces
  • \b - a word boundary (so, Group 1 value will end with a letter)
  • (?:\s*,)? - an optional sequence of zero or more whitespace chars and a comma.