如何解析带括号的表达式的嵌套逗号分隔列表

Question

I know how to use the Python regex module to parse nested parentheses. This regular expression

\(([^()]*+(?:(?R)[^()]*)*+)\)

correctly finds the outermost parentheses in

some (text)(text here(possible text)text(possible text(more text)))end text

(example here )

I also know how to find items in a comma-separated list:

[^,]+(?=,?)

matches correctly the elements of the list

dgad asg , adgda adg, a, g, asdgdg,dg sfg

(see here )

But I need a combination of these two. I need to parse the elements of a comma separated list, where the elements themselves may contain parantheses (with comma-separated lists in them). In this list

dg(dsfsd, (d,d,g)(g,as(d,f) fdg) sdfs, sf)ad asg , adgda (a) adg, a, g, asdgdg,dg sfg(f,g, (dff, d)df, g) kd

I need to identify the elements as:

first: dg(dsfsd, (d,d,g)(g,as(d,f) fdg) sdfs, sf)ad asg
second: adgda (a) adg
third: a
fourth: g
fifth: asdgdg
sixth: dg sfg(f,g, (dff, d)df, g) kd

I don't know how to combine the two regular expressions. Could someone help me, please? Thx.

Answer 1

You may use

r'(?>(\((?>[^()]*(?1)?)*\))|[^,])+'

See the regex demo

Details

• (?>(\\((?>[^()]*(?1)?)*\\))|[^,])+ - 1 or more occurrences of (to avoid empty string matches):
  • (\\((?>[^()]*(?1)?)*\\)) - Capturing group 1 (defined to be able to use a subroutine) matching:
    • \\( - a (
    • (?>[^()]*(?1)?)* - any 0+ chars other than ( and ) followed with an optional whole Group 1 pattern (recursed here)
    • \\) - a )
  • | - or
  • [^,] - any char but ,

Python demo:

import regex as re

rx = r"(?>(\((?>[^()]*(?1)?)*\))|[^,])+"
s = "dg(dsfsd, (d,d,g)(g,as(d,f) fdg) sdfs, sf)ad asg , adgda (a) adg, a, g, asdgdg,dg sfg(f,g, (dff, d)df, g) kd"
matches = re.finditer(rx, s)
for m in matches:
    print(m.group().strip())

Output:

dg(dsfsd, (d,d,g)(g,as(d,f) fdg) sdfs, sf)ad asg
adgda (a) adg
a
g
asdgdg
dg sfg(f,g, (dff, d)df, g) kd