Merge pull request #25021 from JuliaLang/nl/unicode

Move Unicode-related functions to new Unicode stdlib package

CONTRIBUTING.md

@@ -174,7 +174,7 @@ The steps required to add a new docstring are listed below:
 Examples written within docstrings can be used as testcases known as "doctests" by annotating code blocks with `jldoctest`.
 
     ```jldoctest
-    julia> uppercase("Docstring test")
+    julia> Unicode.uppercase("Docstring test")
     "DOCSTRING TEST"
     ```
 

NEWS.md

@@ -738,6 +738,16 @@ Deprecated or removed
   * The `sum_kbn` and `cumsum_kbn` functions have been moved to the
     [KahanSummation](https://github.com/JuliaMath/KahanSummation.jl) package ([#24869]).
 
+  * Unicode-related string functions have been moved to the new `Unicode` standard
+    library module ([#25021]). This applies to `normalize_string`, `graphemes`,
+    `is_assigned_char`, `textwidth`, `isascii`, `islower`, `isupper`, `isalpha`,
+    `isdigit`, `isxdigit`, `isnumber`, `isalnum`, `iscntrl`, `ispunct`, `isspace`,
+    `isprint`, `isgraph`, `lowercase`, `uppercase`, `titlecase`, `lcfirst` and `ucfirst`.
+
+  * `isnumber` has been deprecated in favor of `isnumeric`, `is_assigned_char`
+    in favor of `isassigned` and `normalize_string` in favor of `normalize`, all three
+    in the new `Unicode` standard library module ([#25021]).
+
 Command-line option changes
 ---------------------------
 
@@ -1711,3 +1721,4 @@ Command-line option changes
 [#24413]: https://github.com/JuliaLang/julia/issues/24413
 [#24653]: https://github.com/JuliaLang/julia/issues/24653
 [#24869]: https://github.com/JuliaLang/julia/issues/24869
+[#25021]: https://github.com/JuliaLang/julia/issues/25021

base/arrayshow.jl

@@ -166,7 +166,7 @@ function print_matrix(io::IO, X::AbstractVecOrMat,
     screenwidth -= length(pre) + length(post)
     presp = repeat(" ", length(pre))  # indent each row to match pre string
     postsp = ""
-    @assert textwidth(hdots) == textwidth(ddots)
+    @assert Unicode.textwidth(hdots) == Unicode.textwidth(ddots)
     sepsize = length(sep)
     rowsA, colsA = indices(X,1), indices(X,2)
     m, n = length(rowsA), length(colsA)

base/char.jl

@@ -64,7 +64,7 @@ function show(io::IO, c::Char)
             return
         end
     end
-    if isprint(c)
+    if Unicode.isprint(c)
         write(io, 0x27, c, 0x27)
     else
         u = UInt32(c)
@@ -81,6 +81,6 @@ end
 function show(io::IO, ::MIME"text/plain", c::Char)
     show(io, c)
     u = UInt32(c)
-    print(io, ": ", isascii(c) ? "ASCII/" : "", "Unicode U+", hex(u, u > 0xffff ? 6 : 4))
-    print(io, " (category ", UTF8proc.category_abbrev(c), ": ", UTF8proc.category_string(c), ")")
+    print(io, ": ", Unicode.isascii(c) ? "ASCII/" : "", "Unicode U+", hex(u, u > 0xffff ? 6 : 4))
+    print(io, " (category ", Unicode.category_abbrev(c), ": ", Unicode.category_string(c), ")")
 end

base/client.jl

@@ -359,7 +359,7 @@ function load_machine_file(path::AbstractString)
         s = split(line, '*'; keep = false)
         map!(strip, s, s)
         if length(s) > 1
-            cnt = isnumber(s[1]) ? parse(Int,s[1]) : Symbol(s[1])
+            cnt = all(isdigit, s[1]) ? parse(Int,s[1]) : Symbol(s[1])
             push!(machines,(s[2], cnt))
         else
             push!(machines,line)

base/deprecated.jl

@@ -1072,13 +1072,6 @@ function Matrix()
     return Matrix(uninitialized, 0, 0)
 end
 
-for name in ("alnum", "alpha", "cntrl", "digit", "number", "graph",
-             "lower", "print", "punct", "space", "upper", "xdigit")
-    f = Symbol("is",name)
-    @eval import .UTF8proc: $f
-    @eval @deprecate ($f)(s::AbstractString) all($f, s)
-end
-
 # TODO: remove warning for using `_` in parse_input_line in base/client.jl
 
 # Special functions have been moved to a package
@@ -1512,7 +1505,7 @@ export hex2num
 @deprecate convert(::Type{Symbol}, s::AbstractString)         Symbol(s)
 @deprecate convert(::Type{String}, s::Symbol)                 String(s)
 @deprecate convert(::Type{String}, v::Vector{UInt8})          String(v)
-@deprecate convert(::Type{S}, g::UTF8proc.GraphemeIterator) where {S<:AbstractString}  convert(S, g.s)
+@deprecate convert(::Type{S}, g::Unicode.GraphemeIterator) where {S<:AbstractString}  convert(S, g.s)
 
 # Issue #19923
 @deprecate ror                  circshift
@@ -2972,6 +2965,28 @@ end
 @deprecate_moved sum_kbn "KahanSummation"
 @deprecate_moved cumsum_kbn "KahanSummation"
 
+# PR #25021
+@deprecate_moved normalize_string "Unicode" true true
+@deprecate_moved graphemes "Unicode" true true
+@deprecate_moved is_assigned_char "Unicode" true true
+@deprecate_moved textwidth "Unicode" true true
+@deprecate_moved islower "Unicode" true true
+@deprecate_moved isupper "Unicode" true true
+@deprecate_moved isalpha "Unicode" true true
+@deprecate_moved isdigit "Unicode" true true
+@deprecate_moved isnumber "Unicode" true true
+@deprecate_moved isalnum "Unicode" true true
+@deprecate_moved iscntrl "Unicode" true true
+@deprecate_moved ispunct "Unicode" true true
+@deprecate_moved isspace "Unicode" true true
+@deprecate_moved isprint "Unicode" true true
+@deprecate_moved isgraph "Unicode" true true
+@deprecate_moved lowercase "Unicode" true true
+@deprecate_moved uppercase "Unicode" true true
+@deprecate_moved titlecase "Unicode" true true
+@deprecate_moved lcfirst "Unicode" true true
+@deprecate_moved ucfirst "Unicode" true true
+
 # END 0.7 deprecations
 
 # BEGIN 1.0 deprecations

base/dict.jl

@@ -1,15 +1,15 @@
 # This file is a part of Julia. License is MIT: https://julialang.org/license
 
 function _truncate_at_width_or_chars(str, width, chars="", truncmark="…")
-    truncwidth = textwidth(truncmark)
+    truncwidth = Unicode.textwidth(truncmark)
     (width <= 0 || width < truncwidth) && return ""
 
     wid = truncidx = lastidx = 0
     idx = start(str)
     while !done(str, idx)
         lastidx = idx
         c, idx = next(str, idx)
-        wid += textwidth(c)
+        wid += Unicode.textwidth(c)
         wid >= width - truncwidth && truncidx == 0 && (truncidx = lastidx)
         (wid >= width || c in chars) && break
     end

base/distributed/Distributed.jl

@@ -15,6 +15,7 @@ using Base: Process, Semaphore, JLOptions, AnyDict, buffer_writes, wait_connecte
             binding_module, notify_error, atexit, julia_exename, julia_cmd,
             AsyncGenerator, display_error, acquire, release, invokelatest, warn_once,
             shell_escape_posixly, uv_error
+using Base.Unicode: isascii, isdigit, isnumeric
 
 # NOTE: clusterserialize.jl imports additional symbols from Base.Serializer for use
 

base/docs/utils.jl

@@ -3,6 +3,7 @@
 # Text / HTML objects
 
 import Base: print, show, ==, hash
+using Base.Unicode
 
 export HTML, @html_str
 
@@ -231,7 +232,8 @@ function matchinds(needle, haystack; acronym = false)
     for (i, char) in enumerate(haystack)
         isempty(chars) && break
         while chars[1] == ' ' shift!(chars) end # skip spaces
-        if lowercase(char) == lowercase(chars[1]) && (!acronym || !isalpha(lastc))
+        if Unicode.lowercase(char) == Unicode.lowercase(chars[1]) &&
+           (!acronym || !Unicode.isalpha(lastc))
             push!(is, i)
             shift!(chars)
         end

base/exports.jl

@@ -725,40 +725,22 @@ export
     eachmatch,
     endswith,
     escape_string,
-    graphemes,
     hex,
     hex2bytes,
     hex2bytes!,
     ind2chr,
     info,
-    is_assigned_char,
-    isalnum,
-    isalpha,
-    isascii,
-    iscntrl,
-    isdigit,
-    isgraph,
-    islower,
     ismatch,
-    isnumber,
-    isprint,
-    ispunct,
-    isspace,
-    isupper,
     isvalid,
-    isxdigit,
     join,
-    lcfirst,
     logging,
-    lowercase,
     lpad,
     lstrip,
     match,
     matchall,
     ncodeunits,
     ndigits,
     nextind,
-    normalize_string,
     oct,
     prevind,
     print,
@@ -785,13 +767,9 @@ export
     string,
     strip,
     summary,
-    textwidth,
     thisind,
-    titlecase,
     transcode,
-    ucfirst,
     unescape_string,
-    uppercase,
     warn,
 
 # random numbers

base/interactiveutil.jl

@@ -51,7 +51,7 @@ function edit(path::AbstractString, line::Integer=0)
         cmd = line != 0 ? `$command $path -l $line` : `$command $path`
     elseif startswith(name, "subl") || startswith(name, "atom")
         cmd = line != 0 ? `$command $path:$line` : `$command $path`
-    elseif name == "code" || (Sys.iswindows() && uppercase(name) == "CODE.EXE")
+    elseif name == "code" || (Sys.iswindows() && Unicode.uppercase(name) == "CODE.EXE")
         cmd = line != 0 ? `$command -g $path:$line` : `$command -g $path`
     elseif startswith(name, "notepad++")
         cmd = line != 0 ? `$command $path -n$line` : `$command $path`

base/io.jl

@@ -923,6 +923,8 @@ characters from that character until the start of the next line are ignored.
 julia> buf = IOBuffer("    text")
 IOBuffer(data=UInt8[...], readable=true, writable=false, seekable=true, append=false, size=8, maxsize=Inf, ptr=1, mark=-1)
 
+julia> using Unicode
+
 julia> skipchars(buf, isspace)
 IOBuffer(data=UInt8[...], readable=true, writable=false, seekable=true, append=false, size=8, maxsize=Inf, ptr=5, mark=-1)
 
@@ -967,7 +969,7 @@ julia> countlines(io, '.')
 ```
 """
 function countlines(io::IO, eol::Char='\n')
-    isascii(eol) || throw(ArgumentError("only ASCII line terminators are supported"))
+    Unicode.isascii(eol) || throw(ArgumentError("only ASCII line terminators are supported"))
     aeol = UInt8(eol)
     a = Vector{UInt8}(uninitialized, 8192)
     nl = 0

base/libuv.jl

@@ -20,10 +20,10 @@ function uv_sizeof_req(req)
 end
 
 for h in uv_handle_types
-@eval const $(Symbol("_sizeof_",lowercase(string(h)))) = uv_sizeof_handle($h)
+@eval const $(Symbol("_sizeof_",Unicode.lowercase(string(h)))) = uv_sizeof_handle($h)
 end
 for r in uv_req_types
-@eval const $(Symbol("_sizeof_",lowercase(string(r)))) = uv_sizeof_req($r)
+@eval const $(Symbol("_sizeof_",Unicode.lowercase(string(r)))) = uv_sizeof_req($r)
 end
 
 uv_handle_data(handle) = ccall(:jl_uv_handle_data,Ptr{Void},(Ptr{Void},),handle)

base/loading.jl

@@ -68,8 +68,8 @@ elseif Sys.isapple()
 
         # If there is no match, it's possible that the file does exist but HFS+
         # performed unicode normalization. See  https://developer.apple.com/library/mac/qa/qa1235/_index.html.
-        isascii(path_basename) && return false
-        Vector{UInt8}(normalize_string(path_basename, :NFD)) == casepreserved_basename
+        Unicode.isascii(path_basename) && return false
+        Vector{UInt8}(Unicode.normalize(path_basename, :NFD)) == casepreserved_basename
     end
 else
     # Generic fallback that performs a slow directory listing.

base/markdown/Markdown.jl

@@ -7,6 +7,7 @@ module Markdown
 
 import Base: show, ==
 import Core: @doc_str
+using Base.Unicode: lowercase, ucfirst, isspace
 
 include(joinpath("parse", "config.jl"))
 include(joinpath("parse", "util.jl"))

base/mpfr.jl

@@ -122,7 +122,7 @@ convert(::Type{BigFloat}, x::Union{Float16,Float32}) = BigFloat(Float64(x))
 convert(::Type{BigFloat}, x::Rational) = BigFloat(numerator(x)) / BigFloat(denominator(x))
 
 function tryparse(::Type{BigFloat}, s::AbstractString, base::Int=0)
-    !isempty(s) && isspace(s[end]) && return tryparse(BigFloat, rstrip(s), base)
+    !isempty(s) && Base.Unicode.isspace(s[end]) && return tryparse(BigFloat, rstrip(s), base)
     z = BigFloat()
     err = ccall((:mpfr_set_str, :libmpfr), Int32, (Ref{BigFloat}, Cstring, Int32, Int32), z, s, base, ROUNDING_MODE[])
     err == 0 ? Nullable(z) : Nullable{BigFloat}()

base/operators.jl

@@ -806,6 +806,8 @@ entered in the Julia REPL (and most editors, appropriately configured) by typing
 
 # Examples
 ```jldoctest
+julia> using Unicode
+
 julia> map(uppercase∘hex, 250:255)
 6-element Array{String,1}:
  "FA"

base/parse.jl

@@ -53,7 +53,7 @@ end
 function parseint_preamble(signed::Bool, base::Int, s::AbstractString, startpos::Int, endpos::Int)
     c, i, j = parseint_next(s, startpos, endpos)
 
-    while isspace(c)
+    while Unicode.isspace(c)
         c, i, j = parseint_next(s,i,endpos)
     end
     (j == 0) && (return 0, 0, 0)
@@ -66,7 +66,7 @@ function parseint_preamble(signed::Bool, base::Int, s::AbstractString, startpos:
         end
     end
 
-    while isspace(c)
+    while Unicode.isspace(c)
         c, i, j = parseint_next(s,i,endpos)
     end
     (j == 0) && (return 0, 0, 0)
@@ -125,10 +125,10 @@ function tryparse_internal(::Type{T}, s::AbstractString, startpos::Int, endpos::
             return Nullable{T}(n)
         end
         c, i = next(s,i)
-        isspace(c) && break
+        Unicode.isspace(c) && break
     end
     (T <: Signed) && (n *= sgn)
-    while !isspace(c)
+    while !Unicode.isspace(c)
         d::T = '0' <= c <= '9' ? c-'0'    :
         'A' <= c <= 'Z' ? c-'A'+10 :
             'a' <= c <= 'z' ? c-'a'+a  : base
@@ -149,7 +149,7 @@ function tryparse_internal(::Type{T}, s::AbstractString, startpos::Int, endpos::
     end
     while i <= endpos
         c, i = next(s,i)
-        if !isspace(c)
+        if !Unicode.isspace(c)
             raise && throw(ArgumentError("extra characters after whitespace in $(repr(SubString(s,startpos,endpos)))"))
             return _n
         end
@@ -168,10 +168,10 @@ function tryparse_internal(::Type{Bool}, sbuff::Union{String,SubString{String}},
     orig_end   = endpos
 
     # Ignore leading and trailing whitespace
-    while isspace(sbuff[startpos]) && startpos <= endpos
+    while Unicode.isspace(sbuff[startpos]) && startpos <= endpos
         startpos = nextind(sbuff, startpos)
     end
-    while isspace(sbuff[endpos]) && endpos >= startpos
+    while Unicode.isspace(sbuff[endpos]) && endpos >= startpos
         endpos = prevind(sbuff, endpos)
     end
 
@@ -186,7 +186,7 @@ function tryparse_internal(::Type{Bool}, sbuff::Union{String,SubString{String}},
 
     if raise
         substr = SubString(sbuff, orig_start, orig_end) # show input string in the error to avoid confusion
-        if all(isspace, substr)
+        if all(Unicode.isspace, substr)
             throw(ArgumentError("input string only contains whitespace"))
         else
             throw(ArgumentError("invalid Bool representation: $(repr(substr))"))
@@ -243,7 +243,7 @@ tryparse_internal(::Type{Float16}, s::AbstractString, startpos::Int, endpos::Int
 
 function tryparse_internal(::Type{Complex{T}}, s::Union{String,SubString{String}}, i::Int, e::Int, raise::Bool) where {T<:Real}
     # skip initial whitespace
-    while i ≤ e && isspace(s[i])
+    while i ≤ e && Unicode.isspace(s[i])
         i = nextind(s, i)
     end
     if i > e

base/pkg/entry.jl

@@ -89,7 +89,7 @@ function available()
     for (pkg, vers) in all_avail
         any(x->Types.satisfies("julia", VERSION, x[2].requires), vers) && push!(avail, pkg)
     end
-    sort!(avail, by=lowercase)
+    sort!(avail, by=Base.Unicode.lowercase)
 end
 
 function available(pkg::AbstractString)
@@ -572,7 +572,7 @@ end
 
 function warnbanner(msg...; label="[ WARNING ]", prefix="")
     cols = Base.displaysize(STDERR)[2]
-    str = rpad(lpad(label, div(cols+textwidth(label), 2), "="), cols, "=")
+    str = rpad(lpad(label, div(cols+Base.Unicode.textwidth(label), 2), "="), cols, "=")
     warn(prefix="", str)
     println(STDERR)
     warn(prefix=prefix, msg...)

base/pkg/reqs.jl

@@ -78,7 +78,7 @@ function write(io::IO, lines::Vector{Line})
     end
 end
 function write(io::IO, reqs::Requires)
-    for pkg in sort!(collect(keys(reqs)), by=lowercase)
+    for pkg in sort!(collect(keys(reqs)), by=Unicode.lowercase)
         println(io, Requirement(pkg, reqs[pkg]).content)
     end
 end