Regex captured groups with positive lookbehind (python)

Question

How is it possible to combine a captured group with positive lookbehinds?

I want to match the following examples:

DS
DS,x=y
Some DS,key=value
Some DS test,key=value&key2=value2
Some DS test,key=value&key2=value2|key3=value3

I came up with the following regex for matching everything but the comma:

^(?P<ds_title>[\w \|\-\=\&æøåÆØÅ]+)(?P<filters>[\w \|\-\=\&æøåÆØÅ]+)?$

I've figured out, that the regex I would need to insert is:

(?<=\,)

But I cannot figure out where to insert it. No matter where I insert it, it seems to break. Does anyone know how this can be done correctly?

would you mind to post one or two lines with text, and the matchings you want — Rubens, Dec 29 '12 at 14:23
What is DS? Currently your question is quite vague, and unanswerable. Please elaborate a little more probably with some working examples. — Rohit Jain, Dec 29 '12 at 14:24
Sorry if this was unclear. The regex already finds the correct characters - "Some DS" is just some random text. What I need is to implement the optional comma between the two existing groups. — Kristian Øllegaard, Dec 29 '12 at 14:30
You need to use `^(?P[\w \|\-\=\&æøåÆØÅ]+)(?:,(?P[\w \|\-\=\&æøåÆØÅ,]+))?$` — Rohit Jain, Dec 29 '12 at 14:45
EPIC, thanks a lot! I'm not a very experienced stack overflow user, but if you make an answer I can mark it as correct, if you want. — Kristian Øllegaard, Dec 29 '12 at 14:48

score 0 · Answer 1 · answered Dec 29 '12 at 14:56

0

Aren't you overthinking this?

^(?P<ds_title>[^,]+)(?:,(?P<filters>.+))?$

Why not just allow any string, rather than limiting it to your list?

In fact, why use a regex at all?

parts = data.split(',', 2)
if len(parts) == 1:
    title, = parts
else:
    title, filters = parts

answered Dec 29 '12 at 14:56

Eric

I'm not sure what the OP wants, but the OP's regex can do some preliminary validation on the input. – nhahtdh Dec 29 '12 at 15:08

1 Answers1