Pandas: Union strings in dataframe

Question

I have a dataframe df

ID  active_seconds  domain  subdomain   search_engine   search_term
0120bc30e78ba5582617a9f3d6dfd8ca    35  vk.com  vk.com  None    None
0120bc30e78ba5582617a9f3d6dfd8ca    54  vk.com  vk.com  None    None
0120bc30e78ba5582617a9f3d6dfd8ca    34  vk.com  vk.com  None    None
16c28c057720ab9fbbb5ee53357eadb7    4   facebook.com    facebook.com    None    None
16c28c057720ab9fbbb5ee53357eadb7    4   facebook.com    facebook.com    None    None
16c28c057720ab9fbbb5ee53357eadb7    8   facebook.com    facebook.com    None    None
0120bc30e78ba5582617a9f3d6dfd8ca    16  megarand.ru megarand.ru None    None
0120bc30e78ba5582617a9f3d6dfd8ca    6   vk.com  vk.com  None    None

I need to change df. If to ID subdomain[i] == subdomain[i-1] I should union this string and active_seconds[i-1] + active_seconds[i]. From this df I want to get

ID  active_seconds  domain  subdomain   search_engine   search_term
0120bc30e78ba5582617a9f3d6dfd8ca    123 vk.com  vk.com  None    None
16c28c057720ab9fbbb5ee53357eadb7    16  facebook.com    facebook.com    None    None
0120bc30e78ba5582617a9f3d6dfd8ca    16  megarand.ru megarand.ru None    None
0120bc30e78ba5582617a9f3d6dfd8ca    6   vk.com  vk.com  None    None

What sould I use to do it?

Why weren't the last two lines joined together?

unutbu
– unutbu

2016-07-20 10:09:24 +00:00
Commented Jul 20, 2016 at 10:09 — unutbu
– unutbu, Commented Jul 20, 2016 at 10:09
@unutbu because domain[i] != domain[i-1]

NineWasps
– NineWasps

2016-07-20 10:13:04 +00:00
Commented Jul 20, 2016 at 10:13 — NineWasps
– NineWasps, Commented Jul 20, 2016 at 10:13

piRSquared · Accepted Answer · 2016-07-20 10:29:40Z

2

This get's real close. Not sure if getting that order correct is important to you.

Also, I made an assumption that I should groupby ID. This means that if the same ID spans across another ID and still in the same subdomain, I'll aggregate the active_seconds.

def proc_id(df):
    cond = df.subdomain != df.subdomain.shift()
    part = cond.cumsum()
    df_ = df.groupby(part).first()
    df_.active_seconds = df.groupby(part).active_seconds.sum()
    return df_

df.groupby('ID').apply(proc_id).reset_index(drop=True)

answered Jul 20, 2016 at 10:29

piRSquared

296k68 gold badges509 silver badges654 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

NineWasps Over a year ago

Can you say, if I have a list: lst_domain = ['vkontakte.ru', 'yandex.ru', 'vk.com', 'moscow.vk.com', 'city-link.ru'] and I need if domain is equal some of this list, I should use that, but if for example domain in df is equal msk.city-link.ru I should write, how it write in lst.domain. I mean in list I have city-link.ru but in df msk.city-link.ru. And I should rewrite it like in a list

piRSquared Over a year ago

This seems to be a different question? You should ask a new question asking about this specifically.

shivsn Over a year ago

@piRSquared Is there something like a complex one line solution.

Collectives™ on Stack Overflow

Pandas: Union strings in dataframe

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related