splitting and combining text based on a delimiter in Python

Question

I've got a list of lists, containing strings. After much assorted regular expressions work, I've inserted what I'd like to use as a delimiter, @@@, into my strings:

[['@@@this is part one and here is part two and here is part three and heres more and heres more'],
 ['this is part one@@@and here is part two and here is part three and heres more and heres more'],
 ['this is part one and here is part two@@@and here is part three and heres more and heres more']
 ['this is part one and here is part two and here is part three@@@and heres more and heres more']
 ['this is part one and here is part two and here is part three and heres more@@@and heres more']]

Now, I need to come up with this:

[['this is part one'],['and here is part two'],['and here is part three'], ['and heres more'], ['and heres more']]

So far my attempts are bloated, hacky, and generally ugly. I find myself splitting, combining, and matching. Can anyone recommend some general advice on this type of problem, and what tools to use to keep it manageable?

EDIT please note! and heres more indeed appears twice in the ideal output!

Can you clarify the rule regarding when "and here's more" should be included in the output? Should it only appear in the output list once? Also, the lists embedded within your input list all be separated by commas? — Boa, Mar 12 '15 at 01:45
Thanks for the question. There is possibly duplicated text...which must remain, and order must still be kept. — tumultous_rooster, Mar 12 '15 at 01:47

Avinash Raj · Accepted Answer · 2015-03-12T05:02:46.250

1

I think you actually need to grab all the characters which is just after to @@@ upto the next and or string end.

>>> [[m] for x in l for m in re.findall(r'@@@(.*?)(?=\sand\b|$)', x[0])]
[['this is part one'], ['and here is part two'], ['and here is part three'], ['and heres more'], ['and heres more']]

edited Mar 12 '15 at 05:02

answered Mar 12 '15 at 01:34

Avinash Raj

172,303
28
230
274

looks like now the text has lost it's order: part one isn't part one anymore! – tumultous_rooster Mar 12 '15 at 01:43
set() doesn't preserve order. [see here for discussion](http://stackoverflow.com/questions/9792664/python-set-changes-element-order) – Omega Mar 12 '15 at 01:46
@MattO'Brien did you mean the update ?.. did you want the above expected output for whole or only for the first inner list? – Avinash Raj Mar 12 '15 at 01:48
@MattO'Brien could you update your question with exact expected output? – Avinash Raj Mar 12 '15 at 01:53
The question has the the correct expected output...indeed, `and heres more` must appear twice! – tumultous_rooster Mar 12 '15 at 01:56
you want to remove the duplicates then why `heres more` must appear twice? – Avinash Raj Mar 12 '15 at 01:57
I don't want to remove the duplicates! `[['this is part one'],['and here is part two'],['and here is part three'], ['and heres more'], ['and heres more']]` – tumultous_rooster Mar 12 '15 at 01:58
check my update...i think you actually need to use `re.findall` – Avinash Raj Mar 12 '15 at 02:01
@MattO'Brien i would like to here from you, so that i could correct my mistakes quickly. – Avinash Raj Mar 12 '15 at 02:13
It looks good. I've realized I've made a simplified version of the problem -- in real life, each substring does not begin with `and`. For the question I asked your solution looks good, thank you. – tumultous_rooster Mar 12 '15 at 02:20
Avinash, I think I may actually edit this question to change the fact that the beginning is not consistent with the appearance of `and` – tumultous_rooster Mar 12 '15 at 02:32
OK, I will accept your answer and ask a new question. – tumultous_rooster Mar 12 '15 at 02:43
@AvinashRaj Someone said to me once: *people are strange animals* ;) – Rizier123 Mar 12 '15 at 20:07

splitting and combining text based on a delimiter in Python

1 Answers1