switch to utf8proc's portable, up-to-date, upper/lowercase functions (f…

…ixes JuliaLang#11471)

base/string.jl

@@ -793,10 +793,6 @@ write(io::IO, s::RopeString) = (write(io, s.head); write(io, s.tail))
 sizeof(s::RopeString) = sizeof(s.head) + sizeof(s.tail)
 
 ## uppercase and lowercase transformations ##
-
-uppercase(c::Char) = convert(Char, ccall(:towupper, Cwchar_t, (Cwchar_t,), c))
-lowercase(c::Char) = convert(Char, ccall(:towlower, Cwchar_t, (Cwchar_t,), c))
-
 uppercase(s::AbstractString) = map(uppercase, s)
 lowercase(s::AbstractString) = map(lowercase, s)
 

base/utf8proc.jl

@@ -3,7 +3,7 @@
 # Various Unicode functionality from the utf8proc library
 module UTF8proc
 
-import Base: show, showcompact, ==, hash, string, symbol, isless, length, eltype, start, next, done, convert, isvalid
+import Base: show, showcompact, ==, hash, string, symbol, isless, length, eltype, start, next, done, convert, isvalid, lowercase, uppercase
 
 export isgraphemebreak
 
@@ -121,6 +121,12 @@ end
 
 charwidth(c::Char) = Int(ccall(:utf8proc_charwidth, Cint, (UInt32,), c))
 
+# faster x+y that does no overflow checking
+fastplus(x::Char, y::UInt32) = reinterpret(Char, reinterpret(UInt32, x) + y)
+
+lowercase(c::Char) = isascii(c) ? ('A' <= c <= 'Z' ? fastplus(c,0x00000020) : c) : ccall(:utf8proc_tolower, Char, (UInt32,), c)
+uppercase(c::Char) = isascii(c) ? ('a' <= c <= 'z' ? fastplus(c,0xffffffe0) : c) : ccall(:utf8proc_toupper, Char, (UInt32,), c)
+
 ############################################################################
 
 # returns UTF8PROC_CATEGORY code in 0:30 giving Unicode category

deps/Makefile

@@ -1436,7 +1436,7 @@ UTF8PROC_OBJ_HEADER = $(build_includedir)/utf8proc.h
 UTF8PROC_OBJ_TARGET = $(UTF8PROC_OBJ_LIB) $(UTF8PROC_OBJ_HEADER)
 
 $(UTF8PROC_SRC_TARGET): $(UTF8PROC_SRC_DIR)/Makefile
-	$(MAKE) -C $(UTF8PROC_SRC_DIR) cc="$(CC) -O2 -std=c99 $(fPIC) -DUTF8PROC_EXPORTS $(DEPS_CFLAGS)" AR="$(AR)" libutf8proc.a
+	$(MAKE) -C $(UTF8PROC_SRC_DIR) CC="$(CC)" CFLAGS="-O2 -std=c99 $(fPIC) -DUTF8PROC_EXPORTS $(DEPS_CFLAGS)" AR="$(AR)" libutf8proc.a
 	touch -c $@
 $(UTF8PROC_SRC_DIR)/checked: $(UTF8PROC_SRC_TARGET)
 ifeq ($(OS),$(BUILD_OS))

deps/checksums/utf8proc-1.3-dev1.tar.gz/md5

@@ -0,0 +1 @@
+e61de478d2a858586671dfd56477281e

deps/checksums/utf8proc-1.3-dev1.tar.gz/sha512

@@ -0,0 +1 @@
+167d3a7779335d7ac0103431c4e7b1d9dc4f31836a7704524ea6c1c29d447012ec49f122247f9bf5d445500d7ddd7af5403ca260da05ffac9dba114e32afac07

deps/utf8proc.version

@@ -1,2 +1,2 @@
-UTF8PROC_BRANCH=v1.2
-UTF8PROC_SHA1=e1fdad0ca9dc518b429439b6f4eac546a1bdd0de
+UTF8PROC_BRANCH=1.3-dev1
+UTF8PROC_SHA1=f7219d516efe7496737d041f22b0ba567174fb16

test/strings.jl

@@ -673,6 +673,10 @@ end
 @test lowercase("AbC") == "abc"
 @test lowercase('A') == 'a'
 @test lowercase('a') == 'a'
+@test uppercase('α') == '\u0391'
+@test lowercase('Δ') == 'δ'
+@test lowercase('\U118bf') == '\U118df'
+@test uppercase('\U1044d') == '\U10425'
 @test ucfirst("Abc") == "Abc"
 @test ucfirst("abc") == "Abc"
 @test lcfirst("ABC") == "aBC"