R : Using AND in grep on different columns

Asked Jun 09 '16 at 19:04

Active Jun 09 '16 at 19:44

Viewed 131 times

Consider below as data frame

test1 test2 test5
1a     dsa    xx
1c     ad     xxxx
1e     as     cccxx
2f     ssa    xxxxx
2a     bs     xxxx
2s     bs     xxzzx

I want to create a Subset which has "1" in test1 and "a" in test2, like the below one

test1 test2 test5
1a     dsa    xx
1c     ad     xxxx
1e     as     cccxx

I tried the following method

test4 = (test3[ grep("1", test3$test1),] & test3[ grep("a", test3$test2),])

but does not work

edited Jun 09 '16 at 19:44

asked Jun 09 '16 at 19:04

Praveen DA

1

how about `dplyr::filter(df, test1 == 1, test2 == 'a')`? – Psidom Jun 09 '16 at 19:13
3

or `df[df$test1 == 1 & df$test2 == "a", ]` – J_F Jun 09 '16 at 19:15
@ Psidom and @J_F both the methods work perfect – Praveen DA Jun 09 '16 at 19:25
Do we have any way to use grep in there as my original data set has multiple characters on both test1 and test2. or some thing like "like" in SQL – Praveen DA Jun 09 '16 at 19:38
1

You can use `%like%` from `data.table`. `setDT(df);df[test1 %like% "^1" & test2 %like% "^a"]` – ytk Jun 09 '16 at 20:11
@ytk. It worked. Thanks – Praveen DA Jun 09 '16 at 20:26

0 Answers0