S3: Support signature=anonymous to access a S3 bucket anonymously

This can be used, for example, with importtree=yes to import from a public
bucket.

This needs a patch that has not yet landed in the aws library, and will
need to be adjusted to support compiling with old versions of the library,
so is not yet suitable for merging.
See https://github.com/aristidb/aws/pull/281

The stack.yaml changes are provided to show how to build against the aws
fork and will need to be reverted as well.

Sponsored-by: Dartmouth College's DANDI project

CHANGELOG

@@ -6,6 +6,9 @@ git-annex (10.20221004) UNRELEASED; urgency=medium
     do not operate on a repository that has an empty name.
   * move: Fix openFile crash with -J
     (Fixes a reversion in 8.20201103)
+  * S3: Support signature=anonymous to access a S3 bucket anonymously.
+    This can be used, for example, with importtree=yes to import from
+    a public bucket.
 
  -- Joey Hess <id@joeyh.name>  Mon, 03 Oct 2022 13:36:42 -0400
 

Remote/S3.hs

@@ -1,6 +1,6 @@
 {- S3 remotes
  -
- - Copyright 2011-2020 Joey Hess <id@joeyh.name>
+ - Copyright 2011-2022 Joey Hess <id@joeyh.name>
  -
  - Licensed under the GNU AGPL version 3 or higher.
  -}
@@ -156,7 +156,9 @@ requeststyleField = Accepted "requeststyle"
 signatureField :: RemoteConfigField
 signatureField = Accepted "signature"
 
-newtype SignatureVersion = SignatureVersion Int
+data SignatureVersion 
+	= SignatureVersion Int
+	| Anonymous
 
 signatureVersionParser :: RemoteConfigField -> FieldDesc -> RemoteConfigFieldParser
 signatureVersionParser f fd =
@@ -165,10 +167,17 @@ signatureVersionParser f fd =
   where
 	go "v2" = Just (SignatureVersion 2)
 	go "v4" = Just (SignatureVersion 4)
+	go "anonymous" = Just Anonymous
 	go _ = Nothing
 
 	defver = SignatureVersion 2
 
+isAnonymous :: ParsedRemoteConfig -> Bool
+isAnonymous c = 
+	case getRemoteConfigValue signatureField c of
+		Just Anonymous -> True
+		_ -> False
+
 portField :: RemoteConfigField
 portField = Accepted "port"
 
@@ -272,7 +281,9 @@ s3Setup'  ss u mcreds c gc
 		(c', encsetup) <- encryptionSetup (c `M.union` defaults) gc
 		pc <- either giveup return . parseRemoteConfig c'
 			=<< configParser remote c'
-		c'' <- setRemoteCredPair ss encsetup pc gc (AWS.creds u) mcreds
+		c'' <- if isAnonymous pc
+			then pure c'
+			else setRemoteCredPair ss encsetup pc gc (AWS.creds u) mcreds
 		pc' <- either giveup return . parseRemoteConfig c''
 			=<< configParser remote c''
 		info <- extractS3Info pc'
@@ -286,7 +297,9 @@ s3Setup'  ss u mcreds c gc
 		showNote "Internet Archive mode"
 		pc <- either giveup return . parseRemoteConfig c
 			=<< configParser remote c
-		c' <- setRemoteCredPair ss noEncryptionUsed pc gc (AWS.creds u) mcreds
+		c' <- if isAnonymous pc
+			then pure c
+			else setRemoteCredPair ss noEncryptionUsed pc gc (AWS.creds u) mcreds
 		-- Ensure user enters a valid bucket name, since
 		-- this determines the name of the archive.org item.
 		let validbucket = replace " " "-" $ map toLower $
@@ -841,17 +854,20 @@ type S3HandleVar = TVar (Either (Annex (Maybe S3Handle)) (Maybe S3Handle))
 {- Prepares a S3Handle for later use. Does not connect to S3 or do anything
  - else expensive. -}
 mkS3HandleVar :: ParsedRemoteConfig -> RemoteGitConfig -> UUID -> Annex S3HandleVar
-mkS3HandleVar c gc u = liftIO $ newTVarIO $ Left $ do
+mkS3HandleVar c gc u = liftIO $ newTVarIO $ Left $
-	mcreds <- getRemoteCredPair c gc (AWS.creds u)
+	if isAnonymous c
-	case mcreds of
+		then go =<< liftIO AWS.anonymousCredentials
-		Just creds -> do
+		else do
-			awscreds <- liftIO $ genCredentials creds
+			mcreds <- getRemoteCredPair c gc (AWS.creds u)
-			let awscfg = AWS.Configuration AWS.Timestamp awscreds debugMapper Nothing
+			case mcreds of
-			ou <- getUrlOptions
+				Just creds -> go =<< liftIO (genCredentials creds)
-			return $ Just $ S3Handle (httpManager ou) awscfg s3cfg
+				Nothing -> return Nothing
-		Nothing -> return Nothing
   where
 	s3cfg = s3Configuration c
+	go awscreds = do
+		let awscfg = AWS.Configuration AWS.Timestamp awscreds debugMapper Nothing
+		ou <- getUrlOptions
+		return $ Just $ S3Handle (httpManager ou) awscfg s3cfg
 
 withS3Handle :: S3HandleVar -> (Maybe S3Handle -> Annex a) -> Annex a
 withS3Handle hv a = liftIO (readTVarIO hv) >>= \case

doc/special_remotes/S3.mdwn

@@ -82,6 +82,7 @@ the S3 remote.
 * `signature` - This controls the S3 signature version to use.
   "v2" is currently the default, "v4" is needed to use some S3 services.
   If you get some kind of authentication error, try "v4".
+  To access a S3 bucket anonymously, use "anonymous".
 
 * `bucket` - S3 requires that buckets have a globally unique name, 
   so by default, a bucket name is chosen based on the remote name

stack.yaml

@@ -11,10 +11,10 @@ flags:
     gitlfs: true
 packages:
 - '.'
-resolver: lts-18.13
+resolver: lts-19.16
 extra-deps:
 - IfElse-0.85
-- aws-0.22
+- /home/joey/tmp/aws
 - bloomfilter-2.0.1.0
 - git-lfs-1.2.0
 - http-client-restricted-0.0.4
@@ -24,4 +24,3 @@ extra-deps:
 - base16-bytestring-0.1.1.7
 - base64-bytestring-1.0.0.3
 - bencode-0.6.1.1
-- http-client-0.7.9