rename CharString -> UTF32String (fix #4943)

NEWS.md

@@ -36,6 +36,8 @@ Library improvements
 
   * New string type, `UTF16String` ([#4930]).
 
+  * `CharString` is renamed to `UTF32String` ([#4943]).
+
 Deprecated or removed
 ---------------------
 

base/char.jl

@@ -54,4 +54,4 @@ sizeof(::Type{Char}) = 4
 ## printing & showing characters ##
 
 print(io::IO, c::Char) = (write(io,c); nothing)
-show(io::IO, c::Char) = (print(io,'\''); print_escaped(io,CharString(c),"'"); print(io,'\''))
+show(io::IO, c::Char) = (print(io,'\''); print_escaped(io,UTF32String(c),"'"); print(io,'\''))

base/deprecated.jl

@@ -356,5 +356,8 @@ export mmread
 export Stat
 const Stat = StatStruct
 
+export CharString
+const CharString = UTF32String
+
 # 0.3 discontinued functions
 

base/exports.jl

@@ -25,7 +25,6 @@ export
     BitMatrix,
     BitVector,
     BunchKaufman,
-    CharString,
     Cholesky,
     CholeskyPivoted,
     Cmd,
@@ -105,6 +104,7 @@ export
     Triangular,
     Tridiagonal,
     UTF16String,
+    UTF32String,
     VecOrMat,
     Vector,
     VersionNumber,
@@ -819,6 +819,7 @@ export
     uppercase,
     utf8,
     utf16,
+    utf32,
     warn,
     xdump,
 

base/string.jl

@@ -565,25 +565,28 @@ next(s::GenericString, i::Int) = next(s.string, i)
 
 ## plain old character arrays ##
 
-immutable CharString <: DirectIndexString
+immutable UTF32String <: DirectIndexString
     chars::Array{Char,1}
 
-    CharString(a::Array{Char,1}) = new(a)
-    CharString(c::Char...) = new([ c[i] for i=1:length(c) ])
+    UTF32String(a::Array{Char,1}) = new(a)
+    UTF32String(c::Char...) = new([ c[i] for i=1:length(c) ])
 end
-CharString(x...) = CharString(map(char,x)...)
+UTF32String(x...) = UTF32String(map(char,x)...)
 
-next(s::CharString, i::Int) = (s.chars[i], i+1)
-endof(s::CharString) = length(s.chars)
-length(s::CharString) = length(s.chars)
+next(s::UTF32String, i::Int) = (s.chars[i], i+1)
+endof(s::UTF32String) = length(s.chars)
+length(s::UTF32String) = length(s.chars)
 
-convert(::Type{CharString}, s::String) = CharString(Char[c for c in s])
-convert{T<:String}(::Type{T}, v::Vector{Char}) = convert(T, CharString(v))
+utf32(x) = convert(UTF32String, x)
+convert(::Type{UTF32String}, s::String) = UTF32String(Char[c for c in s])
+convert{T<:String}(::Type{T}, v::Vector{Char}) = convert(T, UTF32String(v))
+convert(::Type{Array{Char,1}}, s::UTF32String) = s.chars
+convert(::Type{Array{Char}}, s::UTF32String) = s.chars
 
-reverse(s::CharString) = CharString(reverse(s.chars))
+reverse(s::UTF32String) = UTF32String(reverse(s.chars))
 
-sizeof(s::CharString) = sizeof(s.chars)
-convert{T<:Union(Int32,Uint32)}(::Type{Ptr{T}}, s::CharString) =
+sizeof(s::UTF32String) = sizeof(s.chars)
+convert{T<:Union(Int32,Uint32,Char)}(::Type{Ptr{T}}, s::UTF32String) =
     convert(Ptr{T}, s.chars)
 
 ## substrings reference original strings ##
@@ -813,8 +816,11 @@ end
 ## string promotion rules ##
 
 promote_rule(::Type{UTF8String} , ::Type{ASCIIString}) = UTF8String
-promote_rule(::Type{UTF8String} , ::Type{CharString} ) = UTF8String
-promote_rule(::Type{ASCIIString}, ::Type{CharString} ) = UTF8String
+promote_rule(::Type{UTF8String} , ::Type{UTF16String} ) = UTF8String
+promote_rule(::Type{ASCIIString}, ::Type{UTF16String} ) = UTF8String
+promote_rule(::Type{UTF32String} , ::Type{UTF16String} ) = UTF8String
+promote_rule(::Type{UTF8String} , ::Type{UTF32String} ) = UTF8String
+promote_rule(::Type{ASCIIString}, ::Type{UTF32String} ) = UTF8String
 promote_rule{T<:String}(::Type{RepString}, ::Type{T}) = RepString
 
 ## printing literal quoted string data ##

base/utf16.jl

@@ -42,6 +42,8 @@ convert(::Type{UTF16String}, s::UTF16String) = s
 convert(::Type{UTF16String}, s::String) = encode16(s)
 convert(::Type{UTF8String}, s::UTF16String) =
     sprint(length(s.data), io->for c in s; write(io,c::Char); end)
+convert(::Type{Array{Uint16,1}}, s::UTF16String) = s.data
+convert(::Type{Array{Uint16}}, s::UTF16String) = s.data
 
 sizeof(s::UTF16String) = sizeof(s.data)
 convert{T<:Union(Int16,Uint16)}(::Type{Ptr{T}}, s::UTF16String) =

test/strings.jl

@@ -799,7 +799,7 @@ bin_val = hex2bytes("07bf")
 @test sizeof(RopeString("abc","def")) == 6
 
 # issue #3597
-@test string(CharString(['T', 'e', 's', 't'])[1:1], "X") == "TX"
+@test string(UTF32String(['T', 'e', 's', 't'])[1:1], "X") == "TX"
 
 # issue #3710
 @test prevind(SubString("{var}",2,4),4) == 3

test/unicode.jl

@@ -20,7 +20,7 @@ for encoding in ["UTF-32LE", "UTF-16BE", "UTF-16LE", "UTF-8"]
 end
 
 f=open(joinpath(unicodedir,"UTF-32LE.unicode"))
-str1 = CharString(reinterpret(Char, read(f, Uint32, 1112065)[2:]))
+str1 = UTF32String(reinterpret(Char, read(f, Uint32, 1112065)[2:]))
 close(f)
 
 f=open(joinpath(unicodedir,"UTF-8.unicode"))
@@ -29,7 +29,7 @@ close(f)
 @test str1 == str2
 
 str1 = "∀ ε > 0, ∃ δ > 0: |x-y| < δ ⇒ |f(x)-f(y)| < ε"
-str2 = CharString(
+str2 = UTF32String(
     8704, 32, 949, 32, 62, 32, 48, 44, 32, 8707, 32,
     948, 32, 62, 32, 48, 58, 32, 124, 120, 45, 121, 124,
     32, 60, 32, 948, 32, 8658, 32, 124, 102, 40, 120,