Last active
August 29, 2015 14:11
-
-
Save jeroen/9edf97f873f17a4ce5d3 to your computer and use it in GitHub Desktop.
Test encoding/recoding of strings in Rcpp
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# UTF8 tests on Windows | |
# Using R 3.1.2 for win64 with LC_CTYPE=English_United States.1252 | |
library(Rcpp) | |
# Part 1: simply echo a string | |
echo1src <- "#include <Rcpp.h> | |
// [[Rcpp::export]] | |
std::string echo1(std::string src){ | |
return src; | |
}" | |
echo2src <- "#include <Rcpp.h> | |
// [[Rcpp::export]] | |
std::wstring echo2(std::wstring src){ | |
return src; | |
}" | |
echo3src <- "#include <Rcpp.h> | |
// [[Rcpp::export]] | |
Rcpp::String echo3(Rcpp::String src){ | |
return src; | |
}" | |
sourceCpp(code = echo1src) | |
sourceCpp(code = echo2src) | |
sourceCpp(code = echo3src) | |
x1 <- enc2native("Zürich") | |
x2 <- enc2utf8("Zürich") | |
x3 <- "東京" | |
echo1(x1) #OK | |
echo1(x2) #FAIL | |
echo1(x3) #FAIL | |
echo2(x1) #FAIL | |
echo2(x2) #FAIL | |
echo2(x3) #FAIL | |
echo3(x1) #OK | |
echo3(x2) #OK | |
echo3(x3) #ok | |
# Part 2: actually manipulating the string | |
replacesrc <- "#include <Rcpp.h> | |
// [[Rcpp::export]] | |
Rcpp::String replace_first( Rcpp::String z, Rcpp::String x, Rcpp::String y){ | |
z.replace_first( x, y ) ; | |
return z ; | |
}" | |
sourceCpp(code = replacesrc) | |
replace_first(paste("I like", x1), x1, x3) #FAIL | |
replace_first(paste("I like", x2), x2, x3) #FAIL | |
replace_first(paste("I like", x3), x3, x2) #FAIL | |
replace_first(paste("I like", x3), x3, x1) #OK | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment