Add more unit tests for casefolding

Author: ScottPJones

Project.toml

@@ -4,12 +4,9 @@ authors  = ["ScottPJones <[email protected]>"]
 keywords = ["Strings"]
 license  = "MIT"
 uuid     = "e79e7a6a-7bb1-5a4d-9d64-da657b06f53a"
-version  = "0.1.10"
+version  = "0.1.11"
 
 [deps]
-Pkg     = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
-Test    = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
-Random  = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
 Unicode = "4ec0a83e-493e-50e2-b9ac-8f72acf5a8f5"
 
 ModuleInterfaceTools = "5cb8414e-7aab-5a03-a681-351269c074bf"
@@ -18,3 +15,10 @@ MurmurHash3 = "b10b62ed-fbae-5ea5-b934-abaf0477b71d"
 
 StrAPI = "69e7dfc3-c4d0-5e14-8d95-d6042a05b383"
 ChrBase = "c13fa7b1-fb91-5a40-8b3c-3aad7fd30002"
+
+[extras]
+Test = "8dfed614-e22c-5e08-85e1-65c5234f0b40"
+Random  = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
+
+[targets]
+test = ["Test", "Random"]

src/StrBase.jl

@@ -32,7 +32,7 @@ using ModuleInterfaceTools
              _memcmp, _memcpy, _memset, _fwd_memchr, _rev_memchr,
              empty_string, _calcpnt, _mask_bytes, _allocate,
              MS_UTF8, MS_UTF16, MS_UTF32, MS_SubUTF32, MS_Latin, MS_ByteStr, MS_RawUTF8,
-             _wrap_substr, _empty_sub, 
+             _wrap_substr, _empty_sub,
              AccessType, UInt16_U, UInt32_U, UInt16_S, UInt32_S, UInt16_US, UInt32_US,
              alignedtype, swappedtype, checkkeep, splitarr, __split, __rsplit, __replace
 

test/basic.jl

@@ -34,6 +34,29 @@ end
     end
 end
 
+@testset "casefold string" begin
+    for ST in (ASCIIStr, LatinStr, UCS2Str, UTF32Str, UTF8Str, UTF16Str)
+        C = eltype(ST)
+        tm = typemax(C)
+        for c = 0:Int(tm)
+            # Skip surrogates
+            0xd800 <= c < 0xe000 && continue
+            ch = C(c)
+            # Check to make sure this character would still fit uppercased
+            cu = uppercase(ch)
+            cu > tm && continue
+            for str in ("$ch test Beg", "test End $ch", "test $ch Mid", "$ch")
+                cvtstr = convert(ST, str)
+                # Don't do this for LatinStr until ChrBase bug fixed
+                ST !== LatinStr && @test uppercase(str) == uppercase(cvtstr)
+                @test lowercase(str) == lowercase(cvtstr)
+                #@test titlecase(str) == titlecase(cvtstr)
+                #@test uppercase_first(str) == uppercase_first(cvtstr)
+            end
+        end
+    end
+end
+
 @testset "{starts,ends}_with" begin
     for (ST, type_list) in compat_types, CT in type_list, str in test_strings_base[CT]
         cvtstr = convert(ST, str)