gsub with fixed=TRUE (fix tidyverse#181)

vspinu · Apr 29, 2013 · 76bfdb5 · 76bfdb5
1 parent 043147b
commit 76bfdb5
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 6 deletions.
diff --git a/R/guess.r b/R/guess.r
@@ -338,7 +338,7 @@ guess_formats <- function(x, orders, locale = Sys.getlocale("LC_TIME"),
     p <- p[nzchar(p)]
     alpha["p"] <-
       if ( length(p) == 0L ) ""
-    else sprintf("(?<p>%s)(?![[:alpha:]])", paste(p, collapse = "|"))
+      else sprintf("(?<p>%s)(?![[:alpha:]])", paste(p, collapse = "|"))
 
     alpha <- unlist(alpha)
 
@@ -405,10 +405,10 @@ guess_formats <- function(x, orders, locale = Sys.getlocale("LC_TIME"),
 
     num_exact <- num_flex <- num
     num_flex[] <- sprintf("%s(?!\\d)", num) 
-
-    num_exact[] <- gsub("(?<!\\()\\?(?!<)", "", perl = T, # remove ?
-                        gsub("+", "*",  fixed = T,  
-                             gsub(">", "_e>", num))) # append _e to avoid duplicates    
+    ## it is crittcal tu use fixed = TRUE because of the multibyte bug #181
+    num_exact[] <- gsub("(?<!\\()\\?(?!<)", "", perl = TRUE, # remove ?
+                        gsub("+", "*",  fixed = TRUE,
+                             gsub(">", "_e>", num, fixed = TRUE))) # append _e to avoid duplicates
 
     num_flex["m"] <- sprintf("((?<m>1[0-2]|0?[1-9](?!\\d))|(%s))", gsub("_[bB]", "\\1_m", alpha[["b"]])) 
     num_exact["m"] <- sprintf("((?<m_e>1[0-2]|0[1-9])|(%s))", gsub("_[bB]", "\\1_m_e>", alpha[["b"]]))

diff --git a/inst/tests/test-parsers.R b/inst/tests/test-parsers.R
@@ -535,6 +535,8 @@ test_that("ymd_hms parses OO and Oo formats correctly", {
 ## x_RU2 <- "Вт. 2012 авг. 14 11:52:57 "
 ## ymd_hms(c(x_RU, x_RU2), locale = "ru_RU.utf8")
 
+## Sys.setlocale("LC_TIME", "ja_JP.utf8")
+## format(Sys.time(), format = "%A %Y %B %d %I:%M:%S %p")
 
 ## gsub("([].|(){^$*+?[])", "\\\\\\1", "sdfs.")
 
@@ -561,4 +563,4 @@ test_that("ymd_hms parses OO and Oo formats correctly", {
 ## guess_formats("101010202020", "ymdT")
 ## guess_formats("001020", "ymd")
 
-## guess_formats("10 am", "r")
+## guess_formats("10 am", "r")