How to remove rows from a data.frame containing a symbol in a particular column

问题

I would like to remove all the rows from a dataframe named mydata where the column named first contains a "+". What is the correct command for this in R?

Here is what my dataframe looks like:

session                              first     last           city
1  9cf571c8faa67cad2aa9ff41f3a26e38     cat   biddix          fresno
2  e30f853d4e54604fd62858badb68113a caleb+joey amos                
3  2ad41134cc285bcc06892fd68a471cd7  daniel  folkers                
4  2ad41134cc285bcc06892fd68a471cd7  daniel  folkers                
5  63a5e839510a647c1ff3b8aed684c2a5 charles   pierce           flint
6  691df47f2df12f14f000f9a17d1cc40e j+henry    franz prescott+valley
7  691df47f2df12f14f000f9a17d1cc40e j+henry    franz prescott+valley
8  b3a1476aa37ae4b799495256324a8d3d  carrie mascorro            brea
9  bd9f1404b313415e7e7b8769376d2705    fred  morales       las+vegas
10 b50a610292803dc302f24ae507ea853a  aurora      lee                
11 fb74940e6feb0dc61a1b4d09fcbbcb37  andrew    price       yorkville

回答1:

grep!

mydata <- read.table(textConnection("session  first  last  city
9cf571c8faa67cad2aa9ff41f3a26e38  cat biddix  fresno
e30f853d4e54604fd62858badb68113a  caleb+joey  amos  blah
63a5e839510a647c1ff3b8aed684c2a5  me+you  amos  blah"), header=T, stringsAsFactors=FALSE)

grep("\\+",mydata$first)

which returns

[1] 2 3

Telling you in column 2, rows 2 and 3 have a '+' in them.

So you could run:

mydata <- mydata[-grep("\\+",mydata$first),]
mydata

And those entire rows would be deleted. Not sure if it's a typo in your question or not but you say you want to "remove the rows in the first column", do you mean the entries or the entire row?

来源：https://stackoverflow.com/questions/8424540/how-to-remove-rows-from-a-data-frame-containing-a-symbol-in-a-particular-column