From 2aa9154b1f850ac14cd4291357594da0d59df8d9 Mon Sep 17 00:00:00 2001
From: Joey Hess <joeyh@joeyh.name>
Date: Tue, 23 Jul 2024 12:30:27 -0400
Subject: [PATCH] require a valid uuid at the end of an annex+http url

---
 P2P/Http/Url.hs                        | 34 ++++++++++++++++++++------
 doc/design/p2p_protocol_over_http.mdwn |  6 +++--
 doc/git-annex-p2phttp.mdwn             |  4 +--
 3 files changed, 32 insertions(+), 12 deletions(-)

diff --git a/P2P/Http/Url.hs b/P2P/Http/Url.hs
index 5d85b54353..b80ae33350 100644
--- a/P2P/Http/Url.hs
+++ b/P2P/Http/Url.hs
@@ -9,8 +9,14 @@
 
 module P2P.Http.Url where
 
+import Types.UUID
+import Utility.FileSystemEncoding
+import Utility.PartialPrelude
+
 import Data.List
 import Network.URI
+import System.FilePath.Posix as P
+import qualified Data.UUID as UUID
 #ifdef WITH_SERVANT
 import Servant.Client (BaseUrl(..), Scheme(..))
 import Text.Read
@@ -24,37 +30,49 @@ isP2PHttpProtocolUrl s =
 	"annex+http://" `isPrefixOf` s ||
 	"annex+https://" `isPrefixOf` s
 
-data P2PHttpUrl =
+data P2PHttpUrl = P2PHttpUrl
+	{ p2pHttpUrlString :: String
+	, p2pHttpUUID :: Maybe UUID
 #ifdef WITH_SERVANT
-	P2PHttpUrl String BaseUrl
-#else
-	P2PHttpUrl String
+	, p2pHttpBaseUrl :: BaseUrl
 #endif
+	}
+	deriving (Show)
 
 parseP2PHttpUrl :: String -> Maybe P2PHttpUrl
 parseP2PHttpUrl us
 	| isP2PHttpProtocolUrl us = case parseURI (drop prefixlen us) of
 		Nothing -> Nothing
+		Just u ->
 #ifdef WITH_SERVANT
-		Just u -> 
 			case uriScheme u of
 				"http:" -> mkbaseurl Http u
 				"https:" -> mkbaseurl Https u
 				_ -> Nothing
 #else
-		Just _ -> 
-			Just (P2PHttpUrl us)
+			Just $ P2PHttpUrl us (extractuuid u)
 #endif
 	| otherwise = Nothing
   where
 	prefixlen = length "annex+"
+
+	extractuuid u = do
+		p <- lastMaybe $ P.splitDirectories (uriPath u)
+		-- While git-annex generally allows a UUID that is not
+		-- well formed, here it's important to make sure that the
+		-- url a user provided really ends with a UUID, so check
+		-- that it's well formed.
+		case UUID.fromString p of
+			Nothing -> Nothing
+			Just _ -> return (UUID (encodeBS p)) 
+
 #ifdef WITH_SERVANT
 	mkbaseurl s u = do
 		auth <- uriAuthority u
 		port <- if null (uriPort auth)
 			then Just defaultP2PHttpProtocolPort
 			else readMaybe (dropWhile (== ':') (uriPort auth))
-		return $ P2PHttpUrl us $ BaseUrl
+		return $ P2PHttpUrl us (extractuuid u) $ BaseUrl
 			{ baseUrlScheme = s
 			, baseUrlHost = uriRegName auth
 			, baseUrlPath = uriPath u
diff --git a/doc/design/p2p_protocol_over_http.mdwn b/doc/design/p2p_protocol_over_http.mdwn
index af3c3702ca..f24d3a6038 100644
--- a/doc/design/p2p_protocol_over_http.mdwn
+++ b/doc/design/p2p_protocol_over_http.mdwn
@@ -9,8 +9,10 @@ connection (mostly). This is a translation of that protocol to HTTP.
 
 To indicate that an url uses this protocol, use 
 `annex+http` or `annex+https` as the url scheme. Such an url uses
-port 9417 by default, although another port can be specified. For example,
-`annex+http://example.com/` or `annex+http://example.com:80/`
+port 9417 by default, although another port can be specified. 
+The last part of the path of such an url is always the repository uuid.
+For example, `annex+http://example.com/git-annex/$uuid` or 
+`annex+http://example.com:80/git-annex/$uuid`
 
 ## base64 encoding of keys, uuids, and filenames
 
diff --git a/doc/git-annex-p2phttp.mdwn b/doc/git-annex-p2phttp.mdwn
index dc2a02f236..9102af2d4a 100644
--- a/doc/git-annex-p2phttp.mdwn
+++ b/doc/git-annex-p2phttp.mdwn
@@ -17,8 +17,8 @@ API.
 
 Typically a remote will have `remote.name.url` set to a http url
 as usual, and `remote.name.annexUrl` set to an annex+http url such as
-`annex+http://example.com/`. The annex+http url is served by this
-server, and uses port 9417 by default.
+`annex+http://example.com/git-annex/$uuid`. The annex+http url is
+served by this server, and uses port 9417 by default.
 
 As well as serving the git-annex HTTP API, this server provides a
 convenient way to download the content of any key, by using the path
-- 
2.30.2