| is.DBCS {lazy.utf8} | R Documentation |
Checks if the Characters in a String are Multi Byte Characters.
is.DBCS(x) is.MBCS(x)
x |
String vector |
Locale of your Windows machine can be obtained by the following:
shell("ver")
shell("PowerShell Get-WinSystemLocale")
shell("chcp > chcp.txt") # chcp.txt will be created somewhere.
strsplit(readLines("chcp.txt")," ")[[1]][3]
Code 923 means Shift-JIS.
Version and locale of R can be obtained as follows;
version Sys.getlocale()
T check the character encoding, use: Encoding(char).
The result will be UTF-8 or unknown meaning system encoding.
About rawToChar and charToRaw functions:
To get the raw expression of character, use
char <- paste(c(0:9," ",letters,LETTERS),collapse="")
raw <- charToRaw(char)
To get the numeric equivalent of raw, use
num <- as.numeric(raw) .
To recover the original characters, use
char <- rawToChar( as.raw(num) ) .
Hex numbers used to express raw expression can be bypassed by using
num <- char2num(char)
char <- num2char(num)
A logical vector or list of logical vectors indicating if each character in the input string vector is a multi byte chractor or not.
# A string consisting of single and multi byte characters
# The 3rd, 6th and 7th characters are uppercase Zenkaku Romans.
{if( regexpr("utf8",Sys.getlocale (category = "LC_CTYPE")) > 0 )
nza=c(97, 98, 239, 188, 163, 100, 101, 239, 188, 166, 239, 188, 167, 104)
else
nza <- c(97,98,130,98,100,101,130,101,130,102,104)}
za <- rawToChar(as.raw(nza))
# za <- num2char(nza)
# location of the multi byte characters in a string
is.DBCS(za)
# Pick up the single byte characters.
strsplit(za,"")[[1]][!is.DBCS(za)]
# Pick up the multi byte characters.
strsplit(za,"")[[1]][is.DBCS(za)]
# strsplit changes the encoding from Shift-JIS (double byte) to UTF-8.
Encoding(strsplit(za,"")[[1]])
Encoding(substring(za,1:nchar(za),1:nchar(za)))
unlist(nbyte(strsplit(za,"")[[1]]))
unlist(nbyte(substring(za,1:nchar(za),1:nchar(za))))