Adding plike

DESCRIPTION

@@ -66,7 +66,8 @@ Authors@R: c(
   person("Ofek","Shilon",          role="ctb"),
   person("Vadim","Khotilovich",    role="ctb"),
   person("Hadley","Wickham",       role="ctb"),
-  person("Bennet","Becker",        role="ctb"))
+  person("Bennet","Becker",        role="ctb"),
+  person("Kyle","Haynes",          role="ctb"))
 Depends: R (>= 3.1.0)
 Imports: methods
 Suggests: bit64 (>= 4.0.0), bit (>= 4.0.4), curl, R.utils, xts, nanotime, zoo (>= 1.8-1), yaml, knitr, rmarkdown, markdown

NAMESPACE

@@ -8,7 +8,7 @@ exportClasses(data.table, IDate, ITime)
 export(data.table, tables, setkey, setkeyv, key, "key<-", haskey, CJ, SJ, copy)
 export(setindex, setindexv, indices)
 export(as.data.table,is.data.table,test.data.table)
-export(last,first,like,"%like%","%ilike%","%flike%",between,"%between%",inrange,"%inrange%")
+export(last,first,like,"%like%","%ilike%","%flike%","%plike%",between,"%between%",inrange,"%inrange%")
 export(timetaken)
 export(truelength, setalloccol, alloc.col, ":=")
 export(setattr, setnames, setcolorder, set, setDT, setDF)

NEWS.md

@@ -95,6 +95,8 @@
 
 14. `.datatable.aware` is now recognized in the calling environment in addition to the namespace of the calling package, [dtplyr#184](https://github.com/tidyverse/dtplyr/issues/184). Thanks to Hadley Wickham for the idea and PR.
 
+15. New convenience function `%plike%` maps to `like(..., perl=TRUE)`, [#3702](https://github.com/Rdatatable/data.table/issues/3702). `%plike%` uses Perl-compatible regular expressions (PCRE) which extend TRE, and may be more efficient in some cases. Thanks @KyleHaynes for the suggestion and PR.
+
 ## BUG FIXES
 
 1. `by=.EACHI` when `i` is keyed but `on=` different columns than `i`'s key could create an invalidly keyed result, [#4603](https://github.com/Rdatatable/data.table/issues/4603) [#4911](https://github.com/Rdatatable/data.table/issues/4911). Thanks to @myoung3 and @adamaltmejd for reporting, and @ColeMiller1 for the PR. An invalid key is where a `data.table` is marked as sorted by the key columns but the data is not sorted by those columns, leading to incorrect results from subsequent queries.

R/like.R

@@ -1,15 +1,15 @@
 # Intended for use with a data.table 'where'
 # Don't use * or % like SQL's like.  Uses regexpr syntax - more powerful.
 # returns 'logical' so can be combined with other where clauses.
-like = function(vector, pattern, ignore.case = FALSE, fixed = FALSE) {
+like = function(vector, pattern, ignore.case = FALSE, fixed = FALSE, perl = FALSE) {
   if (is.factor(vector)) {
     # indexing by factors is equivalent to indexing by the numeric codes, see ?`[` #4748
-    ret = grepl(pattern, levels(vector), ignore.case = ignore.case, fixed = fixed)[vector]
+    ret = grepl(pattern, levels(vector), ignore.case = ignore.case, fixed = fixed, perl = perl)[vector]
     ret[is.na(ret)] = FALSE
     ret
   } else {
     # most usually character, but integer and numerics will be silently coerced by grepl
-    grepl(pattern, vector, ignore.case = ignore.case, fixed = fixed)
+    grepl(pattern, vector, ignore.case = ignore.case, fixed = fixed, perl = perl)
   }
 }
 
@@ -19,3 +19,5 @@ like = function(vector, pattern, ignore.case = FALSE, fixed = FALSE) {
 # as grep -F or fgrep -- grep against a fixed pattern (no regex)
 #   (more efficient where applicable)
 "%flike%" = function(vector, pattern) like(vector, pattern, fixed = TRUE)
+# Perl-compatible regex
+"%plike%" = function(vector, pattern) like(vector, pattern, perl = TRUE)

inst/tests/tests.Rraw

@@ -7366,7 +7366,7 @@ test(1530.2, which.first(x), which(x)[1L])
 test(1530.3, which.last(1:5), error = "x not boolean")
 test(1530.4, which.last(x), tail(which(x), 1L))
 
-# test for like, %like%, %ilike%, %flike%
+# test for like, %like%, %ilike%, %flike%, %plike%
 set.seed(2L)
 x = apply(matrix(sample(letters, 12), nrow=2), 1, paste, collapse="")
 y = factor(sample(c(letters[1:5], x), 20, TRUE))
@@ -7382,10 +7382,11 @@ test(1532.06, like(x, '()'), c(TRUE, TRUE, TRUE))
 test(1532.07, like(x, '()', fixed = TRUE), c(FALSE, FALSE, TRUE))
 test(1532.08, x %ilike% 'hey', c(TRUE, TRUE, FALSE))
 test(1532.09, x %flike% '()', c(FALSE, FALSE, TRUE))
-## %like% test for ordered factor with NA
-x = c("A", "B", "C", NA_character_)
+test(1532.10, like(x, "(?=h)(?=.*y)", perl = TRUE), c(FALSE, TRUE, FALSE))
+test(1532.11, x %plike% "(?=h)(?=.*y)", c(FALSE, TRUE, FALSE)) #3702
+x = c("A", "B", "C", NA_character_)  # ordered factor with NA
 x = ordered(x, levels = rev(x)[-1L])
-test(1532.10, x %like% "A", c(TRUE, FALSE, FALSE, FALSE))
+test(1532.12, x %like% "A", c(TRUE, FALSE, FALSE, FALSE))
 
 # coverage for setkey() to 100%
 dt1 = data.table(x=sample(5), y=1:5, key="y")

man/like.Rd

@@ -3,23 +3,26 @@
 \alias{\%like\%}
 \alias{\%ilike\%}
 \alias{\%flike\%}
+\alias{\%plike\%}
 \title{ Convenience function for calling grep. }
 \description{
   Intended for use in \code{i} in \code{\link[=data.table]{[.data.table}}, i.e., for subsetting/filtering.
 
   Syntax should be familiar to SQL users, with interpretation as regex.
 }
 \usage{
-like(vector, pattern, ignore.case = FALSE, fixed = FALSE)
+like(vector, pattern, ignore.case = FALSE, fixed = FALSE, perl = FALSE)
 vector \%like\% pattern
 vector \%ilike\% pattern
 vector \%flike\% pattern
+vector \%plike\% pattern
 }
 \arguments{
    \item{vector}{ Either a \code{character} or a \code{factor} vector. }
    \item{pattern}{ Pattern to be matched }
    \item{ignore.case}{ \code{logical}; is \code{pattern} case-sensitive? }
    \item{fixed}{ \code{logical}; should \code{pattern} be interpreted as a literal string (i.e., ignoring regular expressions)? }
+   \item{perl}{ \code{logical}; is \code{pattern} Perl-compatible regular expression? }
 }
 \details{
   Internally, \code{like} is essentially a wrapper around \code{\link[base:grep]{base::grepl}}, except that it is smarter about handling \code{factor} input (\code{base::grep} uses slow \code{as.character} conversion).
@@ -34,5 +37,6 @@ DT = data.table(Name=c("Mary","George","Martha"), Salary=c(2,3,4))
 DT[Name \%like\% "^Mar"]
 DT[Name \%ilike\% "mar"]
 DT[Name \%flike\% "Mar"]
+DT[Name \%plike\% "(?=Ma)(?=.*y)"]
 }
 \keyword{ data }