titlecase: all non-letters are considered word-separators

This is to be consistent with `istitle`.
JuliaLang · Aug 22, 2017 · 7ff1b18 · 7ff1b18
1 parent 637c623
commit 7ff1b18
Show file tree

Hide file tree

Showing 3 changed files with 26 additions and 7 deletions.
diff --git a/base/deprecated.jl b/base/deprecated.jl
@@ -1709,7 +1709,7 @@ export hex2num
 @deprecate diagm(A::SparseMatrixCSC) spdiagm(sparsevec(A))
 
 # PR #23393
-@deprecate titlecase(s::AbstractString) titlecase(s, false)
+@deprecate titlecase(s::AbstractString) titlecase(s, false, true)
 
 # END 0.7 deprecations
 

diff --git a/base/strings/basic.jl b/base/strings/basic.jl
@@ -423,17 +423,23 @@ julia> titlecase("ISS - international space station", false)
 "ISS - International Space Station"
 ```
 """
-function titlecase(s::AbstractString, strict::Bool)
+function titlecase(s::AbstractString, strict::Bool, compat=false)
     startword = true
     b = IOBuffer()
     for c in s
-        if isspace(c)
+        if compat
+            if isspace(c)
+                print(b, c)
+                startword = true
+                continue
+            end
+        elseif !iscased(c)
             print(b, c)
             startword = true
-        else
-            print(b, startword ? titlecase(c) : strict ? lowercase(c) : c)
-            startword = false
+            continue
         end
+        print(b, startword ? titlecase(c) : strict ? lowercase(c) : c)
+        startword = false
     end
     return String(take!(b))
 end

diff --git a/base/strings/utf8proc.jl b/base/strings/utf8proc.jl
@@ -5,7 +5,7 @@ module UTF8proc
 
 import Base: show, ==, hash, string, Symbol, isless, length, eltype, start, next, done, convert, isvalid, lowercase, uppercase, titlecase
 
-export isgraphemebreak, category_code, category_abbrev, category_string
+export isgraphemebreak, category_code, category_abbrev, category_string, iscased
 
 # also exported by Base:
 export normalize_string, graphemes, is_assigned_char, charwidth, isvalid,
@@ -333,6 +333,19 @@ false
 """
 isalpha(c::Char)  = (UTF8PROC_CATEGORY_LU <= category_code(c) <= UTF8PROC_CATEGORY_LO)
 
+"""
+    iscased(c::Char) -> Bool
+
+Tests whether a character is cased, i.e. is lower-, upper- or title-cased.
+"""
+function iscased(c::Char)
+    ccode = category_code(c)
+    return ccode == UTF8PROC_CATEGORY_LU ||
+           ccode == UTF8PROC_CATEGORY_LT ||
+           ccode == UTF8PROC_CATEGORY_LL
+end
+
+
 """
     isnumber(c::Char) -> Bool