CODEWITHSUNDEEP
rregexstring
user7729135
user7729135

Reputation: 409

Searching for specific string pattern

I want to be able to search for a specific string that has this format:

"q4-2015"
"q2-2013"
"q3-2011"

from a long list of file names and break them down into two variables: quarter and year.

so example a long list of names can include:

"aaaaa-ttttt-eeee-q4-2015-file"
"aaaaaa-fffff-3333-q2-2012-file"

and the code should loop through the file names and then throw out the specific variables, in the first case,

year = 2015, quarter = q4

and in second case:

year = 2012, quarter = q2

etc

Upvotes: 0

Views: 121

Answers (3)

G. Grothendieck
G. Grothendieck

Reputation: 270248

1) strcapture Using the test input shown reproducibly in the Note at the end can invoke strcapture from base R:

pat <- "(q\\d)-(\\d{4})"
strcapture(pat, x, list(quarter = "", year = 0))

giving:

  quarter year
1      q4 2015
2      q2 2012

An alternative might be to have a numeric quarter column. In that case we would use pat <- "(\\d)-(\\d{4})" and list(quarter = 0, year = 0).

2) read.pattern read.pattern in the gsubfn package could be used with the same pattern.

library(gsubfn)

read.pattern(text = x, pattern = pat, col.names = c("quarter", "year"),
  as.is = TRUE)

giving:

  quarter year
1      q4 2015
2      q2 2012

2a) Another approach is to use gsubfn's strapply to produce a yearqtr class object and then we could readily extract the quarter and year or we could just leave it as a yearmon object:

library(gsubfn)
library(zoo)

ym  <- do.call("c", 
  strapply(x, pat, q + y ~ as.yearqtr(paste(y, q, sep = "-"))))

ym
## [1] "2015 Q4" "2012 Q2"

data.frame(quarter = paste0("q", cycle(ym), year = as.integer(ym),
  stringsAsFactors = FALSE)
##   quarter year
## 1      q4 2015
## 2      q2 2012

Note

# test input
x <- c("aaaaa-ttttt-eeee-q4-2015-file",
  "aaaaaa-fffff-3333-q2-2012-file")

Upvotes: 0

Ronak Shah
Ronak Shah

Reputation: 389265

We can try with this pattern

captured_words <- sub(".*\\b(q\\d)-(\\d+)\\b.*", "\\1-\\2", x)
captured_words
#[1] "q4-2015" "q2-2012"

Here, we capture two terms:

1) q followed by a single digit number and 2) the numbers following that.

We can separate them and read them in a dataframe using read.table

read.table(text = paste0(captured_words, collapse = "\n"), sep = "-")

#  V1   V2
#1 q4 2015
#2 q2 2012

data

x <- c("aaaaa-ttttt-eeee-q4-2015-file","aaaaaa-fffff-3333-q2-2012-file")

Upvotes: 1

Tim Biegeleisen
Tim Biegeleisen

Reputation: 522719

We can try using sub here:

quarters <- sapply(input, function(x) {
    sub(".*\\b(q\\d+)-\\d{4}\\b.*", "\\1", x)
})

years <- sapply(input, function(x) {
    sub(".*\\bq\\d+-(\\d{4})\\b.*", "\\1", x)
})

df <- data.frame(quarters, years)
df

        quarters years
q4-2015       q4  2015
q2-2013       q2  2013
q3-2011       q3  2011

Upvotes: 3

Related Questions