Remove n number of characters in a string after a specific character

Question

I'm trying to figure out how to remove citation numbers from a string in the following way using R:

Original string:

"There were 100 people outside.231 They were sharing 10 hotdogs.42 Nice!"

Desired string:

"There were 100 people outside. They were sharing 10 hotdogs. Nice!"

but I'm admittedly very bad with Regex. Would anyone possibly have any ideas? Thanks!

@RomanLuštrik Yes! But only those which are following a period without a space. — honeymoow, Jan 16 '21 at 19:00

ThomasIsCoding · Accepted Answer · 2021-01-16T21:18:58.357

2

You can try (?<=\\.)\\d+ to match the digits after a period, e.g.,

> gsub("(?<=\\.)\\d+", "", s, perl = TRUE)
[1] "There were 100 people outside. They were sharing 10 hotdogs. Nice!"

A more efficient way (thank @JvdV's comment) might be

gsub("\\.\\d+", ".", s, perl = TRUE)

edited Jan 16 '21 at 21:18

answered Jan 16 '21 at 19:13

ThomasIsCoding

53,240
4
13
45

I think the lookbehind is creating a bit too much overhead than needed. Maybe just `gsub("\\.\\d+", ".", s, perl = TRUE)`? Where this would only take a few steps (7), the lookbehind takes [100+](https://regex101.com/r/GnYADL/2). – JvdV Jan 16 '21 at 19:33
@JvdV Thanks for the suggestion, I added yours into the answer – ThomasIsCoding Jan 16 '21 at 21:19

Roman Luštrik · Answer 2 · 2021-01-16T19:07:11.893

To remove a number after a period (you could make this more elaborate for any edge cases you might have) you can find a word, a period and then group capture the integers. Replace those integers with an empty string.

What I do here is I find two groups. First group is a word with a period and the second group are integers that follow it. I return just the word (first group denoted by \\1), discarding the integers.

> gsub("(\\w\\.)(\\d+)", replacement = "\\1", x = xy, perl = TRUE)
[1] "There were 100 people outside. They were eating hotdogs. Nice!"

Remove n number of characters in a string after a specific character

2 Answers2