Update to Unicode 18.0.0

.github/workflows/haskell.yml

@@ -58,75 +58,76 @@ jobs:
           - 9.4.8+macOS
           - 9.6.5
           - 9.8.4
-          - 9.10.2+ucd2haskell
-          - 9.12.2
-          - 9.14.1-alpha
+          - 9.10.3+ucd2haskell
+          - 9.12.4
+          - 9.14.1
+          # - 9.16.1-alpha
           - hlint
         include:
 
           - name: 8.2.2
             use_haskell_actions: true
             ghc_version: 8.2.2
             runner: ubuntu-latest
-            cabal_version: 3.2.0.0
+            cabal_version: 3.10.3.0
             pack_options: DISABLE_TEST=y
             ignore_error: false
 
           - name: 8.4.4
             ghc_version: 8.4.4
             runner: ubuntu-latest
-            cabal_version: 3.2.0.0
+            cabal_version: 3.10.3.0
             pack_options: DISABLE_TEST=y
             ignore_error: false
 
           - name: 8.6.5
             ghc_version: 8.6.5
             runner: ubuntu-latest
-            cabal_version: 3.2.0.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 8.8.4
             ghc_version: 8.8.4
             runner: ubuntu-latest
-            cabal_version: 3.2.0.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 8.10.7
             ghc_version: 8.10.7
             runner: ubuntu-latest
-            cabal_version: 3.2.0.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 9.0.2
             ghc_version: 9.0.2
             runner: ubuntu-latest
-            cabal_version: 3.2.0.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 9.2.8+ucd2haskell
             ghc_version: 9.2.8
             # [NOTE] Oldest version supported by ucd2haskell
             pack_options: CABAL_BUILD_OPTIONS="-f ucd2haskell"
             runner: ubuntu-latest
-            cabal_version: 3.6.2.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 9.4.8
             ghc_version: 9.4.8
             runner: ubuntu-latest
-            cabal_version: 3.8.1.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 9.4.8+macOS
             ghc_version: 9.4.8
             runner: macos-latest
-            cabal_version: 3.10.1.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 9.6.5
             ghc_version: 9.6.5
             runner: ubuntu-latest
-            cabal_version: 3.10.1.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
           - name: 9.8.4
@@ -135,32 +136,39 @@ jobs:
             runner: ubuntu-latest
             ignore_error: false
 
-          - name: 9.10.2+ucd2haskell
-            ghc_version: 9.10.2
+          - name: 9.10.3+ucd2haskell
+            ghc_version: 9.10.3
             cabal_version: 3.12.1.0
             pack_options: CABAL_BUILD_OPTIONS="-f ucd2haskell"
             runner: ubuntu-latest
             ignore_error: false
 
-          - name: 9.12.2
-            ghc_version: 9.12.2
+          - name: 9.12.4
+            ghc_version: 9.12.4
             cabal_version: 3.14.2.0
             runner: ubuntu-latest
             ignore_error: false
 
-          - name: 9.14.1-alpha
-            ghc_version: head
-            cabal_version: 3.16.0.0
-            pack_options: >
-                GHCUP_GHC_OPTIONS="-u https://downloads.haskell.org/ghc/9.14.1-alpha1/ghc-9.14.0.20250819-x86_64-deb12-linux.tar.xz"
-                DISABLE_TEST=y
-                CABAL_BUILD_OPTIONS="--allow-newer=base,template-haskell"
+          - name: 9.14.1
+            ghc_version: 9.14.1
+            cabal_version: 3.16.1.0
             runner: ubuntu-latest
             ignore_error: false
 
+          # TODO: update to 9.16.1-alpha when available
+          # - name: 9.16.1-alpha
+          #   ghc_version: head
+          #   cabal_version: 3.18.1.0
+          #   pack_options: >
+          #       GHCUP_GHC_OPTIONS="-u https://downloads.haskell.org/ghc/9.16.1-alpha1/ghc-9.16.0.XXXXX-x86_64-deb12-linux.tar.xz"
+          #       DISABLE_TEST=y
+          #       CABAL_BUILD_OPTIONS="--allow-newer=base,template-haskell"
+          #   runner: ubuntu-latest
+          #   ignore_error: false
+
           # Template for release candidates
-          # - name: 9.10.0
-          #   ghc_version: 9.8.0.XXXXX
+          # - name: 9.16.0
+          #   ghc_version: 9.16.0.XXXXX
           #   ghcup_release_channel: "https://raw.githubusercontent.com/haskell/ghcup-metadata/master/ghcup-prereleases-0.0.7.yaml"
           #   runner: ubuntu-latest
           #   cabal_version: latest
@@ -169,7 +177,7 @@ jobs:
           - name: hlint
             pack_options: HLINT_OPTIONS="lint" HLINT_TARGETS="lib exe"
             runner: ubuntu-latest
-            cabal_version: 3.2.0.0
+            cabal_version: 3.10.3.0
             ignore_error: false
 
     steps:

README.md

@@ -12,7 +12,7 @@ This repository provides packages to use the
 
 The Haskell data structures are generated programmatically from the UCD files.
 The latest Unicode version supported by these libraries is
-[`17.0.0`](https://www.unicode.org/versions/Unicode17.0.0/).
+[`18.0.0`](https://www.unicode.org/versions/Unicode18.0.0/).
 
 ### `unicode-data`
 

experimental/icu/cbits/icu.c

@@ -1,3 +1,4 @@
+#include <stdbool.h>
 #include <unicode/utypes.h>
 #include <unicode/uchar.h>
 #include <unicode/uscript.h>

experimental/icu/icu.cabal

@@ -14,6 +14,9 @@ copyright:           2023 Composewell Technologies and Contributors
 category:            Data,Text,Unicode
 stability:           Experimental
 build-type:          Simple
+extra-source-files:
+  cbits/icu.c
+  cbits/icu.h
 
 source-repository head
   type: git
@@ -48,9 +51,6 @@ common compile-options
                -fwarn-tabs
   default-language: Haskell2010
 
-extra-source-files:
-  cbits/icu.h
-
 library
   import: default-extensions, compile-options
   if flag(has-icu)
@@ -68,7 +68,8 @@ library
     , text             >= 2.0   && < 2.2
   include-dirs: cbits
   c-sources: cbits/icu.c
-  cc-options: -Wall -Wextra -pedantic
+  -- FIXME: -pedantic
+  cc-options: -Wall -Wextra
   extra-libraries: icuuc
   pkgconfig-depends:
     icu-uc >= 72.1

experimental/icu/lib/ICU/Char.chs

@@ -35,7 +35,7 @@ import Data.Int (Int8)
 import Data.Version (Version, makeVersion)
 import Data.Word (Word32)
 import Foreign (Ptr)
-import Foreign.C (CInt(..))
+import Foreign.C (CBool(..), CInt(..))
 import Foreign.Marshal.Array (allocaArray, peekArray)
 import System.IO.Unsafe (unsafePerformIO)
 
@@ -153,11 +153,13 @@ toGeneralCategory = \case
     deriving (Bounded, Eq, Ord, Show) #}
 
 foreign import ccall safe "icu.h __hs_u_hasBinaryProperty" u_hasBinaryProperty
-    :: UChar32 -> CInt -> Bool
+    :: UChar32 -> CInt -> CBool
 
 hasBinaryProperty :: Char -> UProperty -> Bool
 hasBinaryProperty c
-    = u_hasBinaryProperty (fromIntegral (ord c))
+    = toEnum
+    . fromIntegral
+    . u_hasBinaryProperty (fromIntegral (ord c))
     . fromIntegral
     . fromEnum
 

experimental/icu/lib/ICU/Scripts.hsc

@@ -294,4 +294,6 @@ data Script
     | Tayo                        -- ^ USCRIPT_TAI_YO                        = 210
     | Tols                        -- ^ USCRIPT_TOLONG_SIKI                   = 211
     | Hntl                        -- ^ USCRIPT_TRADITIONAL_HAN_WITH_LATIN    = 212
+    | Pcun                        -- ^ USCRIPT_PROTO_CUNEIFORM               = 213
+    | Seal                        -- ^ USCRIPT_SEAL                          = 214
     deriving (Bounded, Enum, Eq, Ord, Show)

experimental/unicode-data-text/unicode-data-text.cabal

@@ -26,8 +26,9 @@ tested-with:         GHC==8.2.2
                    , GHC==9.4.8
                    , GHC==9.6.5
                    , GHC==9.8.4
-                   , GHC==9.10.2
-                   , GHC==9.12.2
+                   , GHC==9.10.3
+                   , GHC==9.12.4
+                   , GHC==9.14.1
 
 extra-doc-files:
     Changelog.md
@@ -68,7 +69,7 @@ library
   build-depends:
     base             >= 4.7   && < 4.23,
     text             >= 1.2.4 && < 2.2,
-    unicode-data     >= 0.8   && < 0.9
+    unicode-data     >= 0.9   && < 0.10
 
 test-suite test
   import: default-extensions, compile-options
@@ -82,7 +83,7 @@ test-suite test
     base             >= 4.7   && < 4.23,
     hspec            >= 2.0   && < 2.12,
     text             >= 1.2.4 && < 2.2,
-    unicode-data     >= 0.8   && < 0.9,
+    unicode-data     >= 0.9   && < 0.10,
     unicode-data-text
   build-tool-depends:
     hspec-discover:hspec-discover >= 2.0 && < 2.12
@@ -98,7 +99,7 @@ benchmark bench
     tasty-bench      >= 0.2.5 && < 0.6,
     tasty            >= 1.4.1 && < 1.6,
     text             >= 1.2.4 && < 2.2,
-    unicode-data     >= 0.8   && < 0.9,
+    unicode-data     >= 0.9   && < 0.10,
     unicode-data-text
   -- [NOTE] Recommendation of tasty-bench to reduce garbage collection noisiness
   ghc-options: -O2 -fdicts-strict -rtsopts -with-rtsopts=-A32m

ucd.sh

@@ -5,7 +5,7 @@
 # we used to generate them earlier are exactly the same as the ones we are
 # downloading. To ensure that verfication of the checksum is necessary.
 
-VERSION=17.0.0
+VERSION=18.0.0
 
 # When downloading fresh new version comment this out
 VERIFY_CHECKSUM=y
@@ -16,31 +16,31 @@ UCD_URL="https://www.unicode.org/Public/$VERSION/ucd"
 # $ find data/$VERSION/ -type f -print0 | xargs -0 sha256sum
 # Format: filename:checksum
 UCD_FILES="\
-    Blocks.txt:c0edefaf1a19771e830a82735472716af6bf3c3975f6c2a23ffbe2580fbbcb15\
-    CaseFolding.txt:ff8d8fefbf123574205085d6714c36149eb946d717a0c585c27f0f4ef58c4183\
-    DerivedCoreProperties.txt:24c7fed1195c482faaefd5c1e7eb821c5ee1fb6de07ecdbaa64b56a99da22c08\
-    DerivedNormalizationProps.txt:71fd6a206a2c0cdd41feb6b7f656aa31091db45e9cedc926985d718397f9e488\
-    NameAliases.txt:793f6f1e4d15fd90f05ae66460191dc4d75d1fea90136a25f30dd6a4cb950eac\
-    PropertyValueAliases.txt:64e9a5f76f7a1e8b5a47d6a1f9a26522a251208f5276bdfa1559dac7cf2e827a\
-    PropList.txt:130dcddcaadaf071008bdfce1e7743e04fdfbc910886f017d9f9ac931d8c64dd\
-    Scripts.txt:9f5e50d3abaee7d6ce09480f325c706f485ae3240912527e651954d2d6b035bf\
-    ScriptExtensions.txt:ec2107e58825a1586acee8e0911ce18260394ac8b87e535ca325f1ccbeb06bc6\
-    SpecialCasing.txt:efc25faf19de21b92c1194c111c932e03d2a5eaf18194e33f1156e96de4c9588\
-    UnicodeData.txt:2e1efc1dcb59c575eedf5ccae60f95229f706ee6d031835247d843c11d96470c\
-    extracted/DerivedCombiningClass.txt:191463abfbd202703c6fd6776a92a23ac44ec65e0476a7f95aa91ca492cef29b\
-    extracted/DerivedName.txt:019758bbe6c756c40fca6d505187ea660c5e195533e2ff2c841963a212c9d369\
-    extracted/DerivedNumericValues.txt:139b976bdc288be01c80f018523da769cf2845109b5a7f0f8a432db64bfedcfa"
+    Blocks.txt:a58f8d322f3c5e254f9f97b1cbf76a454a7e02de6ac35619a5e4f27aaabfd553\
+    CaseFolding.txt:a004797658a457bec4dc11683e39f69249ea3b595b752dbea6721c4c9f587b0d\
+    DerivedCoreProperties.txt:09c928886a178fcafd93c29e4bd59073a058e5a100b716d425cb563ab50f68c9\
+    DerivedNormalizationProps.txt:98ac7f67d985fe781e317f6182e885e94cabb0c314769e6dd73e48b226931ccd\
+    NameAliases.txt:3d5cf5e468901b080cd99adf2230061b748083705fe633db43ac8b73ec7a13da\
+    PropertyValueAliases.txt:06c4c8eaf7b0bf34abe73b113da1215bd784ac254d4c223600b90267caa4bbbd\
+    PropList.txt:f438f532e8737bb8a2702126cdf9c4af5e357c58c7acf9d9eb2fc7c1a1d955d6\
+    Scripts.txt:0071fd81b6aeae25f6e8bce8efec3066a6476a91b49bdb2f52dc76e817862a6a\
+    ScriptExtensions.txt:5c9d34a922f687726f2a8bcf57d49f905987e51f1b21b58c95a00fbe255cec23\
+    SpecialCasing.txt:8538dea57c184f1ef3783885ea79677b10f6efa06423717157e63712f14d1ad2\
+    UnicodeData.txt:0736451de439ae7baf1425136617da495e09ee5afbe6e394374db7009ea08950\
+    extracted/DerivedCombiningClass.txt:ef6b2611cfb660dba3f6b458b9eb4b05f44ed2417302ee7749d7f0f348793121\
+    extracted/DerivedName.txt:ac6cf808ea4ee29323031d5ba9449de13da51398a3f5ce401546d8fceb976df7\
+    extracted/DerivedNumericValues.txt:c84f084f83ec6852e1db6e7ef15f340a3af2df8cf0c386ac0d076c2cebd189e6"
 
 # Security files:
 # - < 17.0.0: https://www.unicode.org/Public/security/$VERSION/$file)
 # - ≥ 17.0.0: https://www.unicode.org/Public/$VERSION/security/$file)
 SECURITY_URL="https://www.unicode.org/Public/$VERSION/security"
 # Format: filename:checksum
 SECURITY_FILES="\
-    IdentifierStatus.txt:617228a16da13850bf8af28b6cd08f5e9b6595d2eb60404fe6eee2c85b4e4a35\
-    IdentifierType.txt:924ac63faa97ed73420d6ac48d08279d90968c7da0502ab701e08bfbb9683c22\
-    confusables.txt:091c7f82fc39ef208faf8f94d29c244de99254675e09de163160c810d13ef22a\
-    intentional.txt:33738217c15c1a0df0b7a2cc0a0b50b27ebdca119ca11253440ec0102f05626b"
+    IdentifierStatus.txt:5863c7d99ca18f213c41c7318aa5528bebfb6d32ec0f1d5944e37192c119aebd\
+    IdentifierType.txt:fa24851acc669e58670e354e7b98a4ec8f52a809ec4f80524b6a60efdb868831\
+    confusables.txt:6ed3ee967c9dfdf6677d563c9985182fbc50a2efb7d6059cd57b2e2ce18f5b92\
+    intentional.txt:5b69cdfd7be6be45d51b9cf7ec799df91c1acc47c557d66c92a8d6623df78b0e"
 
 # Download the files