Skip to content

Sum of Numeric Values in a DF Column

6 messages · Burhan ul haq, Richard M. Heiberger, David Winsemius +1 more

Burhan ul haq
# 
Hi,

I request help with the following:

INPUT: A data frame where column "Lower" is a character containing numeric
values (different count or occurrences of numeric values in each row,
mostly 2)
structure(list(State = c("Alabama", "Alaska", "Arizona", "Arkansas",
"California"), Lower = c("R 72?33", "R/Coalition 27(23 R, 4 D)?12 D, 1
Ind.",
"R 36?24", "R 64?35, 1 Ind.", "D 52?28"), Upper = c("R 26?8, 1 Ind.",
"R/Coalition 15(14 R, 1 D)?5 D", "R 18?12", "R 24?11", "D 26?14"
)), .Names = c("State", "Lower", "Upper"), row.names = c(NA,
5L), class = "data.frame")

PROBLEM: Need to extract all numeric values and sum them. There are few
exceptions like row2. But these can be ignored and will be fixed manually

SOLUTION SO FAR:
str_extract_all(dd[[2]],"[[:digit:]]+"), returns a list of numbers as
character. I am unable to unlist it, because it mixes them all together, ...

And if I may add, is there a "dplyr" way of doing it ...


Thanks
Richard M. Heiberger
# 
## Continuing with your data

AA <- stringr::str_extract_all(dd[[2]],"[[:digit:]]+")
BB <- lapply(AA, as.numeric)
## I think you are looking for one of the following two expressions
sum(unlist(BB))
sapply(BB, sum)
On Mon, Apr 18, 2016 at 12:48 PM, Burhan ul haq <ulhaqz at gmail.com> wrote:
David Winsemius
# 
I don't understand what is mean by "it mixes them all together". This runs without error and appears to deliver what was requested in your natural language description:
[1] 412
David Winsemius
Alameda, CA, USA
Bert Gunter
# 
... and here is a non-dplyr rsolution:
[1] 105  67  60 100  80


Cheers,
Bert
Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Mon, Apr 18, 2016 at 10:07 AM, Richard M. Heiberger <rmh at temple.edu> wrote:
Bert Gunter
# 
... and a slightly more efficient non-dplyr 1-liner:
function(x)sum(as.numeric(x), na.rm=TRUE))

[1] 105  67  60 100  80

Cheers,
Bert
Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Mon, Apr 18, 2016 at 10:43 AM, Bert Gunter <bgunter.4567 at gmail.com> wrote:
Burhan ul haq
# 
Dear Gunter /  Heiberger,

Thanks for the help. This is what I was looking for:
And that would explain, why one could not use "unlist" as a grand sum total
was not desired, but rather sum for each of the rows.


Br /

On Mon, Apr 18, 2016 at 10:57 PM, Bert Gunter <bgunter.4567 at gmail.com>
wrote: