Extracting repeated characters

Question

I am trying to extract artist and title names. However it is a bit complicated. Here is the list;

nlist <- c(
"Lil' SlimLil' Slim feat. PxMxWxPxMxWx Where Your Ward At!!",               
"I Like It (Mannie Fresh Style)I Like It (Mannie Fresh Style)Ms. Tee",
"Bella VistaBella Vista Mister Wong",
"Tom WareTom WareChina Town",                                        
"Race 'N RhythmRace 'N Rhythm Teenage Girls",                                    
"Ronald MarquisseRonald MarquisseElectro Link 7",
"PleasurePleasure Thoughts Of Old Flames",
"OM, OM, Dom Um RomaoDom Um Romao Chipero",
"HookfaceHookface4 07 181221"
)

Here is the pattern in the strings.

Description:

There are three different patterns (1, 2-7, 8).
RED: Artist (repeated),
BLUE: Title (non-repeated),
GREEN: Conjunction (non-rep&between artist names)

1 and 8 is very hard and I couldn't solve. But for 2 to 7 below codes solve my problem.

title = str_trim(gsub('(.+?)\1','', nlist))
artist = re.match('(.+?)\1', nlist)[,2]
data = cbind(title,artist);data

And here the outputs of the above codes.

     title                                     artist                          
[1,] "feat. PxMxWxPxMxWx Where Your Ward At!!" "Lil' Slim"                     
[2,] "Ms. Tee"                                 "I Like It (Mannie Fresh Style)"
[3,] "Mister Wong"                             "Bella Vista"                   
[4,] "China Town"                              "Tom Ware"                      
[5,] "Teenage Girls"                           "Race 'N Rhythm"                
[6,] "Electro Link 7"                          "Ronald Marquisse"              
[7,] "Thoughts Of Old Flames"                  "Pleasure"                      
[8,] "Chipero"                                 "OM, "  
[9,] "4 07 181221"                             "Hookeface"

Problem: When there is "feat." or "," in the string that cuts the repeated sequence of the string. Question: How can I extract truly the artist names like in below?

My expected result is here (Check 1 and 8);

     title                                     artist                          
[1,] "Where Your Ward At!!"                    "Lil' Slim feat. PxMxWx"                     
[2,] "Ms. Tee"                                 "I Like It (Mannie Fresh Style)"
[3,] "Mister Wong"                             "Bella Vista"                   
[4,] "China Town"                              "Tom Ware"                      
[5,] "Teenage Girls"                           "Race 'N Rhythm"                
[6,] "Electro Link 7"                          "Ronald Marquisse"              
[7,] "Thoughts Of Old Flames"                  "Pleasure"                      
[8,] "Chipero"                                 "OM, Dom Um Romao"                             
[9,] "4 07 181221"                             "Hookeface"

Thanks...

user_3832970 · Answer

You can use the following to get artists:

artist <- sub("^(?:.*?\b(.+?)\1(?=\b|\p{Lu}))*\s*(.*)", "\2", nlist, perl=TRUE)
## => [1] "Where Your Ward At!!"   "Ms. Tee"                "Mister Wong"           
## => [4] "China Town"             "Teenage Girls"          "Electro Link 7"        
## => [7] "Thoughts Of Old Flames" "Chipero"

You may use the following code to get the titles:

rx <- "^(?:.*?\b(.+?)\1(?=\b|\p{Lu}))*"
titles <- regmatches(nlist, regexpr(rx, nlist, perl=TRUE))
titles <- gsub("(.+?)\1", "\1", titles, perl=TRUE)
## => [1] "Lil' Slim feat. PxMxWx"        
## => [2] "I Like It (Mannie Fresh Style)"
## => [3] "Bella Vista"                   
## => [4] "Tom Ware"                      
## => [5] "Race 'N Rhythm"                
## => [6] "Ronald Marquisse"              
## => [7] "Pleasure"                      
## => [8] "OM, Dom Um Romao"

See the regex demo. Details:

^ - start of string
(?:.*?\b(.+?)\1(?=\b|\p{Lu}))* - zero or more occurrences of
- .*? - any zero or more chars other than line break chars, as few as possible
- \b - a word boundary
- (.+?) - Group 1: any one or more chars other than line break chars, as few as possible
- \1 - same value as in Group 1
- (?=\b|\p{Lu}) - next must be either a word boundary or an uppercase letter
\s* - zero or more whitespaces
(.*) - Group 2 (\2): the rest of the line.

See the R demo online.

user_10488504 · Answer

Maybe the following extracts what you want. I remove everything and the last repetition and store it in title. To get the artist I remove the length form the previously found title using substr and then remove the repetitions of the artist using gsub with (.{2,})\1, but this will also remove repetitions in the conjunction .

title <- sub(".*(.{2,})\1\s*", "", nlist)
artist <- trimws(gsub("(.{2,})\1", "\1"
              , substr(nlist, 1, nchar(nlist) - nchar(title)), perl=TRUE))
cbind(title,artist)
#      title                    artist                          
# [1,] "Where Your Ward At!!"   "Lil' Slim feat. PxMxWx"        
# [2,] "Ms. Tee"                "I Like It (Mannie Fresh Style)"
# [3,] "Mister Wong"            "Bella Vista"                   
# [4,] "China Town"             "Tom Ware"                      
# [5,] "Teenage Girls"          "Race 'N Rhythm"                
# [6,] "Electro Link 7"         "Ronald Marquisse"              
# [7,] "Thoughts Of Old Flames" "Pleasure"                      
# [8,] "Chipero"                "OM, Dom Um Romao"              
# [9,] "4 07 181221"            "Hookface"

Another way might be:

x <- sub("^(.*)\1\s*", "", nlist)     #Remove the first repetition of artist
title <- sub(".*?(.{2,})\1\s*", "", x) #Remove Conjunction and repetition of Artist if there is one
artist <- trimws(gsub("(.{2,})\1", "\1"
              , substr(nlist, 1, nchar(nlist) - nchar(title)), perl=TRUE))
cbind(title,artist)
#      title                    artist                          
# [1,] "Where Your Ward At!!"   "Lil' Slim feat. PxMxWx"        
# [2,] "Ms. Tee"                "I Like It (Mannie Fresh Style)"
# [3,] "Mister Wong"            "Bella Vista"                   
# [4,] "China Town"             "Tom Ware"                      
# [5,] "Teenage Girls"          "Race 'N Rhythm"                
# [6,] "Electro Link 7"         "Ronald Marquisse"              
# [7,] "Thoughts Of Old Flames" "Pleasure"                      
# [8,] "Chipero"                "OM, Dom Um Romao"              
# [9,] "4 07 181221"            "Hookface"

Extracting repeated characters

2 Answers