如何用正则表达式填补两个字符之间的差距

Question

I have a data set like below. I would like to replace all dots between two 1's with 1's, as shown in the desired.result. Can I do this with regex in base R ?

I tried:

regexpr("^1\\.1$", my.data$my.string, perl = TRUE)

Here is a solution in c#

Characters between two exact characters

Thank you for any suggestions.

my.data <- read.table(text='
     my.string                           state
     ................1...............1.    A
     ......1..........................1    A
     .............1.....2..............    B
     ......1.................1...2.....    B
     ....1....2........................    B
     1...2.............................    C
     ..........1....................1..    C
     .1............................1...    C
     .................1...........1....    C
     ........1....2....................    C
     ......1........................1..    C
     ....1....1...2....................    D
     ......1....................1......    D
     .................1...2............    D
', header = TRUE, na.strings = 'NA', stringsAsFactors = FALSE)

desired.result <- read.table(text='
     my.string                           state
     ................11111111111111111.    A
     ......1111111111111111111111111111    A
     .............1.....2..............    B
     ......1111111111111111111...2.....    B
     ....1....2........................    B
     1...2.............................    C
     ..........1111111111111111111111..    C
     .111111111111111111111111111111...    C
     .................1111111111111....    C
     ........1....2....................    C
     ......11111111111111111111111111..    C
     ....111111...2....................    D
     ......1111111111111111111111......    D
     .................1...2............    D
', header = TRUE, na.strings = 'NA', stringsAsFactors = FALSE)

Answer 1

Below is an option using gsub with the \\G feature and lookaround assertions.

> gsub('(?:1|\\G(?<!^))\\K\\.(?=\\.*1)', '1', my.data$my.string, perl = TRUE)
# [1] "................11111111111111111." "......1111111111111111111111111111"
# [3] ".............1.....2.............." "......1111111111111111111...2....."
# [5] "....1....2........................" "1...2............................."
# [7] "..........1111111111111111111111.." ".111111111111111111111111111111..."
# [9] ".................1111111111111...." "........1....2...................."
# [11] "......11111111111111111111111111.." "....111111...2...................."
# [13] "......1111111111111111111111......" ".................1...2............"

The \\G feature is an anchor that can match at one of two positions; the start of the string position or the position at the end of the last match. Since it seems you want to avoid the dots at the start of the string position we use a lookaround assertion \\G(?<!^) to exclude the start of the string.

The \\K escape sequence resets the starting point of the reported match and any previously consumed characters are no longer included.

You can find an overall breakdown that explains the regular expression here .

Answer 2

Using gsubfn , the first argument is a regular expression which matches the 1's and the characters between the 1's and captures the latter. The second argument is a function, expressed in formula notation, which uses gsub to replace each character in the captured string with 1:

library(gsubfn)
transform(my.data, my.string = gsubfn("1(.*)1", ~ gsub(".", 1, x), my.string))

If there can be multiple pairs of 1's in a string then use "1(.*?)1" as the regular expression instead.

Visualization The regular expression here is simple enough that it can be directly understood but here is a debuggex visualization anwyays:

1(.*)1

Debuggex Demo

Answer 3

Here is an option that uses a relatively simple regex and the standard combination of gregexpr() , regmatches() , and regmatches<-() to identify, extract, operate on, and then replace substrings matching that regex.

## Copy the character vector
x <- my.data$my.string
## Find sequences of "."s bracketed on either end by a "1"
m <- gregexpr("(?<=1)\\.+(?=1)", x, perl=TRUE)
## Standard template for operating on and replacing matched substrings
regmatches(x,m) <- sapply(regmatches(x,m), function(X) gsub(".", "1", X))

## Check that it worked
head(x)
# [1] "................11111111111111111." "......1111111111111111111111111111"
# [3] ".............1.....2.............." "......1111111111111111111...2....."
# [5] "....1....2........................" "1...2............................."