Implement titlecase function. Fix JuliaLang#19465. (JuliaLang#19469)

Clarify docstring. Single-pass version without call to lowercase. Test subsequent spaces. NEWS.md entry and RST docs. Add NEWS.md links.
inkydragon · Dec 6, 2016 · 6f8e94e · 6f8e94e
1 parent dedd3cd
commit 6f8e94e
Show file tree

Hide file tree

Showing 6 changed files with 50 additions and 2 deletions.
diff --git a/NEWS.md b/NEWS.md
@@ -62,6 +62,8 @@ Library improvements
 
  * New `accumulate` and `accumulate!` functions, which generalize `cumsum` and `cumprod`. Also known as a [scan](https://en.wikipedia.org/wiki/Prefix_sum) operation ([#18931]).
 
+ * New `titlecase` function, which capitalizes the first character of each word within a string ([#19469]).
+
 Compiler/Runtime improvements
 -----------------------------
 
@@ -706,8 +708,10 @@ Language tooling improvements
 [#18442]: https://github.com/JuliaLang/julia/issues/18442
 [#18473]: https://github.com/JuliaLang/julia/issues/18473
 [#18839]: https://github.com/JuliaLang/julia/issues/18839
+[#18931]: https://github.com/JuliaLang/julia/issues/18931
 [#18977]: https://github.com/JuliaLang/julia/issues/18977
 [#19018]: https://github.com/JuliaLang/julia/issues/19018
 [#19233]: https://github.com/JuliaLang/julia/issues/19233
 [#19288]: https://github.com/JuliaLang/julia/issues/19288
 [#19305]: https://github.com/JuliaLang/julia/issues/19305
+[#19469]: https://github.com/JuliaLang/julia/issues/19469
diff --git a/base/exports.jl b/base/exports.jl
@@ -873,6 +873,7 @@ export
  strip,
  strwidth,
  summary,
+ titlecase,
  transcode,
  ucfirst,
  unescape_string,

diff --git a/base/strings/basic.jl b/base/strings/basic.jl
@@ -376,7 +376,7 @@ byte_string_classify(s::String) = byte_string_classify(s.data)
 isvalid(::Type{String}, s::Union{Vector{UInt8},String}) = byte_string_classify(s) != 0
 isvalid(s::String) = isvalid(String, s)
 
-## uppercase and lowercase transformations ##
+## uppercase, lowercase, and titlecase transformations ##
 
 """
  uppercase(s::AbstractString)
@@ -402,6 +402,31 @@ julia> lowercase("STRINGS AND THINGS")
 """
 lowercase(s::AbstractString) = map(lowercase, s)
 
+"""
+ titlecase(s::AbstractString)
+
+Capitalizes the first character of each word in `s`.
+
+```jldoctest
+julia> titlecase("the julia programming language")
+"The Julia Programming Language"
+```
+"""
+function titlecase(s::AbstractString)
+ startword = true
+ b = IOBuffer()
+ for c in s
+ if isspace(c)
+ print(b, c)
+ startword = true
+ else
+ print(b, startword ? titlecase(c) : c)
+ startword = false
+ end
+ end
+ return String(take!(b))
+end
+
 """
  ucfirst(s::AbstractString)
 

diff --git a/base/strings/utf8proc.jl b/base/strings/utf8proc.jl
@@ -3,7 +3,7 @@
 # Various Unicode functionality from the utf8proc library
 module UTF8proc
 
-import Base: show, ==, hash, string, Symbol, isless, length, eltype, start, next, done, convert, isvalid, lowercase, uppercase
+import Base: show, ==, hash, string, Symbol, isless, length, eltype, start, next, done, convert, isvalid, lowercase, uppercase, titlecase
 
 export isgraphemebreak
 
@@ -159,6 +159,7 @@ charwidth(c::Char) = Int(ccall(:utf8proc_charwidth, Cint, (UInt32,), c))
 
 lowercase(c::Char) = isascii(c) ? ('A' <= c <= 'Z' ? c + 0x20 : c) : Char(ccall(:utf8proc_tolower, UInt32, (UInt32,), c))
 uppercase(c::Char) = isascii(c) ? ('a' <= c <= 'z' ? c - 0x20 : c) : Char(ccall(:utf8proc_toupper, UInt32, (UInt32,), c))
+titlecase(c::Char) = isascii(c) ? ('a' <= c <= 'z' ? c - 0x20 : c) : Char(ccall(:utf8proc_totitle, UInt32, (UInt32,), c))
 
 ############################################################################
 

diff --git a/doc/stdlib/strings.rst b/doc/stdlib/strings.rst
@@ -407,6 +407,17 @@
  julia> lowercase("STRINGS AND THINGS")
  "strings and things"
 
+.. function:: titlecase(s::AbstractString)
+
+ .. Docstring generated from Julia source
+
+ Capitalizes the first character of each word in ``s``\ .
+
+ .. doctest::
+
+ julia> titlecase("the julia programming language")
+ "The Julia Programming Language"
+
 .. function:: ucfirst(s::AbstractString)
 
  .. Docstring generated from Julia source

diff --git a/test/strings/basic.jl b/test/strings/basic.jl
@@ -392,6 +392,12 @@ end
 @test lcfirst(GenericString("a")) == "a"
 @test ucfirst(GenericString("A")) == "A"
 
+# titlecase
+@test titlecase('ǉ') == 'ǈ'
+@test titlecase("ǉubljana") == "ǈubljana"
+@test titlecase("aBc ABC") == "ABc ABC"
+@test titlecase("abcD EFG\n\thij") == "AbcD EFG\n\tHij"
+
 # issue # 11464: uppercase/lowercase of GenericString becomes a String
 str = "abcdef\uff\uffff\u10ffffABCDEF"
 @test typeof(uppercase("abcdef")) == String