Merge pull request #5804

Fix corrupted config file header for non-ASCII package names

Cabal/ChangeLog.md

@@ -18,6 +18,9 @@
       supports fully static linking;
       [`glibc` has some issues](https://sourceware.org/glibc/wiki/FAQ#Even_statically_linked_programs_need_some_shared_libraries_which_is_not_acceptable_for_me.__What_can_I_do.3F)
       with fully static linking.
+  * Fix corrupted config file header for non-ASCII package names
+    ([2557](https://github.com/haskell/cabal/issues/2557)).
+  * Extend `Distribution.Simple.Utils.rewriteFileEx` from ASCII to UTF-8 encoding.
 
 ----
 

Cabal/Distribution/Simple/Configure.hs

@@ -276,7 +276,7 @@ parseHeader header = case BLC8.words header of
   ["Saved", "package", "config", "for", pkgId, "written", "by", cabalId,
    "using", compId] ->
       fromMaybe (throw ConfigStateFileBadHeader) $ do
-          _ <- simpleParsec (BLC8.unpack pkgId) :: Maybe PackageIdentifier
+          _ <- simpleParsec (fromUTF8LBS pkgId) :: Maybe PackageIdentifier
           cabalId' <- simpleParsec (BLC8.unpack cabalId)
           compId' <- simpleParsec (BLC8.unpack compId)
           return (cabalId', compId')
@@ -287,7 +287,7 @@ showHeader :: PackageIdentifier -- ^ The processed package.
             -> ByteString
 showHeader pkgId = BLC8.unwords
     [ "Saved", "package", "config", "for"
-    , BLC8.pack $ prettyShow pkgId
+    , toUTF8LBS $ prettyShow pkgId
     , "written", "by"
     , BLC8.pack $ prettyShow currentCabalId
     , "using"

Cabal/Distribution/Simple/Utils.hs

@@ -205,7 +205,7 @@ import Control.Concurrent.MVar
     ( newEmptyMVar, putMVar, takeMVar )
 import Data.Typeable
     ( cast )
-import qualified Data.ByteString.Lazy.Char8 as BS.Char8
+import qualified Data.ByteString.Lazy as BS
 
 import System.Directory
     ( Permissions(executable), getDirectoryContents, getPermissions
@@ -1401,19 +1401,22 @@ rewriteFile = rewriteFileEx normal
 -- the same as the existing content then leave the file as is so that we do not
 -- update the file's modification time.
 --
--- NB: the file is assumed to be ASCII-encoded.
+-- NB: Before Cabal-3.0 the file content was assumed to be
+--     ASCII-representable. Since Cabal-3.0 the file is assumed to be
+--     UTF-8 encoded.
 rewriteFileEx :: Verbosity -> FilePath -> String -> IO ()
 rewriteFileEx verbosity path newContent =
   flip catchIO mightNotExist $ do
-    existingContent <- annotateIO verbosity $ readFile path
-    _ <- evaluate (length existingContent)
-    unless (existingContent == newContent) $
+    existingContent <- annotateIO verbosity $ BS.readFile path
+    _ <- evaluate (BS.length existingContent)
+    unless (existingContent == newContent') $
       annotateIO verbosity $
-        writeFileAtomic path (BS.Char8.pack newContent)
+        writeFileAtomic path newContent'
   where
+    newContent' = toUTF8LBS newContent
+
     mightNotExist e | isDoesNotExistError e
-                    = annotateIO verbosity $ writeFileAtomic path
-                        (BS.Char8.pack newContent)
+                    = annotateIO verbosity $ writeFileAtomic path newContent'
                     | otherwise
                     = ioError e