Trim strings ending with various characters

[Peter_Griffin]

I have a vector of strings like below:

x <- c("AA.3", "BEAM.2", "BRK.B", "BF.B", "CB.1", "DD.2",  
       "IGT.1", "LSI.1",  "NSM.2",  "PLL.1",  "SUN.1",  "DTV.2", 
       "ALTR.1", "NVLS.1", "AGN.2",  "CPWR.1", "LIFE.3", "FTI.1", 
       "SE.7",  "MMI.3", "ABC", "AS.")

I need to trim these strings. For the strings ending with a dot and number (.1 for example), remove the dot and number. For strings ending with a dot, remove the dot. For the strings ending with a dot and letter (.B for example), do not change them. For other strings, do not change them.

I am looking at the stri_trim_right function from the package stringi, but could not figure out how to do it.

[sergiouribe]

what I would try is to first separate it using the dot and then put it back together according to the rules you indicate

[andresrcs]

This is a job for regular expressions

library(stringr)

x <- c("AA.3", "BEAM.2", "BRK.B", "BF.B", "CB.1", "DD.2",  
       "IGT.1", "LSI.1",  "NSM.2",  "PLL.1",  "SUN.1",  "DTV.2", 
       "ALTR.1", "NVLS.1", "AGN.2",  "CPWR.1", "LIFE.3", "FTI.1", 
       "SE.7",  "MMI.3", "ABC", "AS.")

str_remove(x, "\\.\\d?$")
#>  [1] "AA"    "BEAM"  "BRK.B" "BF.B"  "CB"    "DD"    "IGT"   "LSI"   "NSM"  
#> [10] "PLL"   "SUN"   "DTV"   "ALTR"  "NVLS"  "AGN"   "CPWR"  "LIFE"  "FTI"  
#> [19] "SE"    "MMI"   "ABC"   "AS"

^{Created on 2021-01-01 by the reprex package (v0.3.0.9001)}

[Peter_Griffin]

Thanks so much. My data is large and I prefer to use the package stringi, which is much faster than stringr. Do you know how to do it in stringi? Thanks.

[andresrcs]

stringr uses stringi as backend so it shouldn't make much of a difference but you can use this function

library(stringi)

x <- c("AA.3", "BEAM.2", "BRK.B", "BF.B", "CB.1", "DD.2",  
       "IGT.1", "LSI.1",  "NSM.2",  "PLL.1",  "SUN.1",  "DTV.2", 
       "ALTR.1", "NVLS.1", "AGN.2",  "CPWR.1", "LIFE.3", "FTI.1", 
       "SE.7",  "MMI.3", "ABC", "AS.")

stri_replace(x, "", regex = "\\.\\d?$")
#>  [1] "AA"    "BEAM"  "BRK.B" "BF.B"  "CB"    "DD"    "IGT"   "LSI"   "NSM"  
#> [10] "PLL"   "SUN"   "DTV"   "ALTR"  "NVLS"  "AGN"   "CPWR"  "LIFE"  "FTI"  
#> [19] "SE"    "MMI"   "ABC"   "AS"

^{Created on 2021-01-01 by the reprex package (v0.3.0.9001)}

[Peter_Griffin]

Thanks so much! Are there any learning materials that can help me understand what these regex means?

[andresrcs]

A general overview

Regular expressions

And a more complete resource

Mastering Regular Expressions, 3rd Edition

[Peter_Griffin]

Thanks so much! I will read the first link.