One of the most headache problems in R (maybe in all programming) is the Encoding issue, especially on Windows. Even the veterans in the R-Core team can’t survive.
In the new version of R 3.3.0, there’s an upgrade related to match(x, table) ( #PR16491). However, it leads to unconsistent behaviors when x and table contains non-ascii characters. So I filed my first bug report to the bugs.r-project.org.
What a wonderful world it would be, had all the characters been encoded in UTF-8.