add --dry-run: New option

This is intended for users who want to see what it would output in order to
eg, check if a file would be added to git or the annex. It is not intended
as a way for scripts to get information.

Sponsored-by: Dartmouth College's Datalad project

CHANGELOG

@@ -10,6 +10,7 @@ git-annex (10.20220725) UNRELEASED; urgency=medium
   * Added new matching options --want-get-by and --want-drop-by.
   * Allow find --branch to be used in a bare repository, the same as
     the deprecated findref can be.
+  * add --dry-run: New option. 
 
  -- Joey Hess <id@joeyh.name>  Mon, 25 Jul 2022 15:35:45 -0400
 

CmdLine/GitAnnex/Options.hs

@@ -128,6 +128,12 @@ parseUUIDOption :: String -> DeferredParse UUID
 parseUUIDOption = DeferredParse
 	. (Remote.nameToUUID)
 
+parseDryRunOption :: Parser DryRun
+parseDryRunOption = DryRun <$> switch
+	( long "dry-run"
+	<> help "don't make changes, but show what would be done"
+	)
+
 -- | From or To a remote.
 data FromToOptions
 	= FromRemote (DeferredParse Remote)

Command.hs

@@ -106,6 +106,12 @@ stop = return Nothing
 stopUnless :: Annex Bool -> Annex (Maybe a) -> Annex (Maybe a)
 stopUnless c a = ifM c ( a , stop )
 
+{- When doing a dry run, avoid actually performing the action, but pretend
+ - that it succeeded. -}
+skipWhenDryRun :: DryRun -> CommandPerform -> CommandPerform
+skipWhenDryRun (DryRun False) a = a
+skipWhenDryRun (DryRun True) _ = next $ return True
+
 {- When acting on a failed transfer, stops unless it was in the specified
  - direction. -}
 checkFailedTransferDirection :: ActionItem -> Direction -> Annex (Maybe a) -> Annex (Maybe a)

Command/Add.hs

@@ -52,6 +52,7 @@ data AddOptions = AddOptions
 	, updateOnly :: Bool
 	, largeFilesOverride :: Maybe Bool
 	, checkGitIgnoreOption :: CheckGitIgnore
+	, dryRunOption :: DryRun
 	}
 
 optParser :: CmdParamsDesc -> Parser AddOptions
@@ -65,6 +66,7 @@ optParser desc = AddOptions
 		)
 	<*> (parseforcelarge <|> parseforcesmall)
 	<*> checkGitIgnoreSwitch
+	<*> parseDryRunOption
   where
 	parseforcelarge = flag Nothing (Just True)
 		( long "force-large"
@@ -91,16 +93,16 @@ seek o = startConcurrency commandStages $ do
 			ifM (pure (annexdotfiles || not (dotfile file))
 				<&&> (checkFileMatcher largematcher file 
 				<||> Annex.getRead Annex.force))
-				( start si file addunlockedmatcher
+				( start dr si file addunlockedmatcher
 				, if includingsmall
 					then ifM (annexAddSmallFiles <$> Annex.getGitConfig)
-						( startSmall si file s
+						( startSmall dr si file s
 						, stop
 						)
 					else stop
 				)
-		Just True -> start si file addunlockedmatcher
-		Just False -> startSmallOverridden si file
+		Just True -> start dr si file addunlockedmatcher
+		Just False -> startSmallOverridden dr si file
 	case batchOption o of
 		Batch fmt
 			| updateOnly o ->
@@ -126,25 +128,26 @@ seek o = startConcurrency commandStages $ do
 			-- same as a modified unlocked file would get
 			-- locked when added.
 			go False withUnmodifiedUnlockedPointers
+  where
+	dr = dryRunOption o
 
 {- Pass file off to git-add. -}
-startSmall :: SeekInput -> RawFilePath -> FileStatus -> CommandStart
-startSmall si file s =
+startSmall :: DryRun -> SeekInput -> RawFilePath -> FileStatus -> CommandStart
+startSmall dr si file s =
 	starting "add" (ActionItemTreeFile file) si $
-		next $ addSmall file s
+		addSmall dr file s
 
-addSmall :: RawFilePath -> FileStatus -> Annex Bool
-addSmall file s = do
+addSmall :: DryRun -> RawFilePath -> FileStatus -> CommandPerform
+addSmall dr file s = do
 	showNote "non-large file; adding content to git repository"
-	addFile Small file s
+	skipWhenDryRun dr $ next $ addFile Small file s
 
-startSmallOverridden :: SeekInput -> RawFilePath -> CommandStart
-startSmallOverridden si file = 
+startSmallOverridden :: DryRun -> SeekInput -> RawFilePath -> CommandStart
+startSmallOverridden dr si file = 
 	liftIO (catchMaybeIO $ R.getSymbolicLinkStatus file) >>= \case
-		Just s -> starting "add" (ActionItemTreeFile file) si $ next $ do
-			
+		Just s -> starting "add" (ActionItemTreeFile file) si $ do
 			showNote "adding content to git repository"
-			addFile Small file s
+			skipWhenDryRun dr $ next $ addFile Small file s
 		Nothing -> stop
 
 data SmallOrLarge = Small | Large
@@ -188,8 +191,8 @@ addFile smallorlarge file s = do
 		isRegularFile a /= isRegularFile b ||
 		isSymbolicLink a /= isSymbolicLink b
 
-start :: SeekInput -> RawFilePath -> AddUnlockedMatcher -> CommandStart
-start si file addunlockedmatcher = 
+start :: DryRun -> SeekInput -> RawFilePath -> AddUnlockedMatcher -> CommandStart
+start dr si file addunlockedmatcher = 
 	liftIO (catchMaybeIO $ R.getSymbolicLinkStatus file) >>= \case
 		Nothing -> stop
 		Just s
@@ -200,6 +203,7 @@ start si file addunlockedmatcher =
   where
 	go s = ifAnnexed file (addpresent s) (add s)
 	add s = starting "add" (ActionItemTreeFile file) si $
+		skipWhenDryRun dr $
 			if isSymbolicLink s
 				then next $ addFile Small file s
 				else perform file addunlockedmatcher
@@ -209,7 +213,7 @@ start si file addunlockedmatcher =
 	fixuplink key = 
 		starting "add" (ActionItemTreeFile file) si $
 			addingExistingLink file key $
-				withOtherTmp $ \tmp -> do
+				skipWhenDryRun dr $ withOtherTmp $ \tmp -> do
 					let tmpf = tmp P.</> P.takeFileName file
 					liftIO $ moveFile file tmpf
 					ifM (isSymbolicLink <$> liftIO (R.getSymbolicLinkStatus tmpf))
@@ -223,7 +227,8 @@ start si file addunlockedmatcher =
 						)
 	fixuppointer s key =
 		starting "add" (ActionItemTreeFile file) si $
-			addingExistingLink file key $ do
+			addingExistingLink file key $
+				skipWhenDryRun dr $ do
 					Database.Keys.addAssociatedFile key =<< inRepo (toTopFilePath file)
 					next $ addFile Large file s
 

Command/AddUrl.hs

@@ -476,7 +476,7 @@ addWorkTree _ addunlockedmatcher u url file key mtmp = case mtmp of
 					(fromRawFilePath file)
 					(fromRawFilePath tmp)
 				go
-			else void $ Command.Add.addSmall file s
+			else void $ Command.Add.addSmall (DryRun False) file s
   where
 	go = do
 		maybeShowJSON $ JSONChunk [("key", serializeKey key)]

Command/Import.hs

@@ -248,7 +248,7 @@ startLocal o addunlockedmatcher largematcher mode (srcfile, destfile) =
 				>>= maybe
 					stop
 					(\addedk -> next $ Command.Add.cleanup addedk True)
-			, next $ Command.Add.addSmall destfile s
+			, Command.Add.addSmall (DryRun False) destfile s
 			)
 	notoverwriting why = do
 		warning $ "not overwriting existing " ++ fromRawFilePath destfile ++ " " ++ why

Types/Command.hs

@@ -133,3 +133,6 @@ descSection SectionUtility = "Utility commands"
 descSection SectionPlumbing = "Plumbing commands"
 descSection SectionTesting = "Testing commands"
 descSection SectionAddOn = "Addon commands"
+
+newtype DryRun = DryRun Bool
+	deriving (Show)

doc/git-annex-add.mdwn

@@ -77,6 +77,10 @@ annexed content, and other symlinks.
   Like `git add --update`, this does not add new files, but any updates
   to tracked files will be added to the index.
 
+* `--dry-run`
+
+  Output what would be done for each file, but avoid making any changes.
+
 * `--json`
 
   Enable JSON output. This is intended to be parsed by programs that use

doc/todo/add_--dry-run.mdwn

@@ -22,3 +22,4 @@ well -- unless there was a version staged already you don't want to loose etc.
 [[!meta author=yoh]]
 [[!tag projects/datalad]]
 
+> [[done]] joey

doc/todo/add_--dry-run/comment_2_312127fc2780fd894332d7d8b7f7a209._comment

@@ -0,0 +1,25 @@
+[[!comment format=mdwn
+ username="joey"
+ subject="""comment 2"""
+ date="2022-08-03T14:53:35Z"
+ content="""
+Implemented `git-annex add --dry-run`.
+
+As noted it does not let you tell if a file would be added locked or unlocked
+since `git-annex add` output is the same either way.
+
+Also when JSON output is enabled, `git-annex add` usually outputs the key,
+but with `--dry-run`, it does not. Generating the key would involve locking
+down the file, and by that point the command is making changes to the
+filesystem. Use `git-annex calckey` instead.
+
+Implementation was not as simple as making CommandPerform a no-op. Some parts
+of CommandPerform actions needed to be run when doing a dry-run, in order for
+addingExistingLink to display its warning, and for showNote to display the
+very information that the user is interested in seeing from --dry-run.
+
+A cleaner implementation would probably involve moving those actions out of
+CommandPerform, perhaps by splitting CommandPerform into two parts.
+Which would be a lot of work and still failure prone. That is why I don't
+think it's a good idea to add --dry-run to very many commands.
+"""]]