info: Added "annex sizes of repositories" table to the overall display

Thanks to previous work in 11cc9f19339b5fa80f238e42407eb9ed87cf8855,
this is almost entirely free, it only needs to do some additional map
lookups and math.

The strictness annotations keep the memory use from blowing up.

Sponsored-by: unqueued on Patreon

diff --git a/CHANGELOG b/CHANGELOG
index c990a6b85eacd02b3da8535ccfa754130a20ed41..bb874441e561b27c454c6bdbf090fc37ce5ee32e 100644 (file)
--- a/CHANGELOG
+++ b/CHANGELOG
@@ -1,3 +1,9 @@
+git-annex (10.20231228) UNRELEASED; urgency=medium
+
+  * info: Added "annex sizes of repositories" table to the overall display.
+
+ -- Joey Hess <id@joeyh.name>  Fri, 29 Dec 2023 11:52:06 -0400
+
 git-annex (10.20231227) upstream; urgency=medium
 
   * migrate: Support distributed migrations by recording each migration,

diff --git a/Command/Info.hs b/Command/Info.hs
index 864ad0257913a314e8c53d25fe70ac9f71a5a9b1..6ec76f5b844837113e3dde43e1b23885f1bc49db 100644 (file)
--- a/Command/Info.hs
+++ b/Command/Info.hs
@@ -283,6 +283,7 @@ global_slow_stats =
        , known_annex_files True
        , known_annex_size True
        , total_annex_size
+       , reposizes_stats_global
        , backend_usage
        , bloom_info
        ]
@@ -298,7 +299,7 @@ tree_fast_stats isworktree =
 tree_slow_stats :: [FilePath -> Stat]
 tree_slow_stats =
        [ const numcopies_stats
-       , const reposizes_stats
+       , const reposizes_stats_tree
        , const reposizes_total
        ]
 
@@ -372,6 +373,10 @@ countRepoList :: Int -> String -> String
 countRepoList _ [] = "0"
 countRepoList n s = show n ++ "\n" ++ beginning s
 
+dispRepoList :: String -> String
+dispRepoList [] = ""
+dispRepoList s = "\n" ++ beginning s
+
 dir_name :: FilePath -> Stat
 dir_name dir = simpleStat "directory" $ pure dir
 
@@ -441,7 +446,8 @@ known_annex_size isworktree =
 total_annex_size :: Stat
 total_annex_size = 
        simpleStat "combined annex size of all repositories" $
-               showSizeKeys =<< cachedAllRepoData
+               showSizeKeys . fromMaybe mempty . allRepoData
+               =<< cachedAllRepoData
   
 treeDesc :: Bool -> String
 treeDesc True = "working tree"
@@ -545,21 +551,29 @@ numcopies_stats = stat "numcopies stats" $ json fmt $
                . map (\(variance, count) -> "numcopies " ++ variance ++ ": " ++ show count)
                . V.toList
 
-reposizes_stats :: Stat
-reposizes_stats = stat desc $ nojson $ do
+reposizes_stats_tree :: Stat
+reposizes_stats_tree = reposizes_stats True "repositories containing these files"
+       =<< cachedRepoData
+
+reposizes_stats_global :: Stat
+reposizes_stats_global = reposizes_stats False "annex sizes of repositories" 
+       . repoData =<< cachedAllRepoData
+
+reposizes_stats :: Bool -> String -> M.Map UUID KeyInfo -> Stat
+reposizes_stats count desc m = stat desc $ nojson $ do
        sizer <- mkSizer
-       l <- map (\(u, kd) -> (u, sizer storageUnits True (sizeKeys kd)))
-               . sortBy (flip (comparing (sizeKeys . snd)))
-               . M.toList
-               <$> cachedRepoData
+       let l = map (\(u, kd) -> (u, sizer storageUnits True (sizeKeys kd))) $
+               sortBy (flip (comparing (sizeKeys . snd))) $
+               M.toList m
        let maxlen = maximum (map (length . snd) l)
        descm <- lift Remote.uuidDescriptions
        -- This also handles json display.
        s <- lift $ Remote.prettyPrintUUIDsWith (Just "size") desc descm (Just . show) $
                map (\(u, sz) -> (u, Just $ mkdisp sz maxlen)) l
-       return $ countRepoList (length l) s
+       return $ if count
+               then countRepoList (length l) s
+               else dispRepoList s
   where
-       desc = "repositories containing these files"
        mkdisp sz maxlen = DualDisp
                { dispNormal = lpad maxlen sz
                , dispJson = sz
@@ -619,29 +633,34 @@ cachedReferencedData = do
                        put s { referencedData = Just v }
                        return v
 
-cachedAllRepoData :: StatState KeyInfo
+cachedAllRepoData :: StatState StatInfo
 cachedAllRepoData = do
        s <- get
        case allRepoData s of
-               Just v -> return v
+               Just _ -> return s
                Nothing -> do
                        matcher <- lift getKeyOnlyMatcher
-                       !v <- lift $ overLocationLogs emptyKeyInfo $ \k locs d -> do
+                       !(d, rd) <- lift $ overLocationLogs (emptyKeyInfo, mempty) $ \k locs (d, rd) -> do
                                ifM (matchOnKey matcher k)
                                        ( do
-                                               numcopies <- genericLength . snd
+                                               alivelocs <- snd
                                                        <$> trustPartition DeadTrusted locs
-                                               return (addKeyCopies numcopies k d)
-                                       , return d
+                                               let !d' = addKeyCopies (genericLength alivelocs) k d
+                                               let !rd' = foldl' (flip (accumrepodata k)) rd alivelocs
+                                               return (d', rd')
+                                       , return (d, rd)
                                        )
-                       put s { allRepoData = Just v }
-                       return v
+                       let s' = s { allRepoData = Just d, repoData = rd }
+                       put s'
+                       return s'
+  where
+       accumrepodata k = M.alter (Just . addKey k . fromMaybe emptyKeyInfo)
 
--- currently only available for directory info
+-- only available for directory info, populated by earlier getDirStatInfo
 cachedNumCopiesStats :: StatState (Maybe NumCopiesStats)
 cachedNumCopiesStats = numCopiesStats <$> get
 
--- currently only available for directory info
+-- only available for directory info, populated by earlier getDirStatInfo
 cachedRepoData :: StatState (M.Map UUID KeyInfo)
 cachedRepoData = repoData <$> get
 

diff --git a/doc/git-annex-info.mdwn b/doc/git-annex-info.mdwn
index bdbcf1415e6abefe526c442b33f41a2e60c6d188..9b0e9715dfe21d4bb8f5a0b82a12898ed2915669 100644 (file)
--- a/doc/git-annex-info.mdwn
+++ b/doc/git-annex-info.mdwn
@@ -12,8 +12,9 @@ Displays statistics and other information for the specified item.
 
 When no item is specified, displays overall information. This includes a
 list of all known repositories, how much annexed data is present in the
-local repository, and the total size of all annexed data in the working
-tree.
+local repository, the total size of all annexed data in the working
+tree, the combined size of annexed data in all repositories, and the annex
+sizes of each repository.
 
 When a directory is specified, displays information
 about the annexed files in that directory (and subdirectories).

diff --git a/doc/todo/info_show_total_annex_sizes_of_repositories.mdwn b/doc/todo/info_show_total_annex_sizes_of_repositories.mdwn
index ce48edc537bac2ff6179d1e0d2ffbcbd82e5a18f..d96e7e9b5170ef5c8b0c5b99fdbe7896f39e460e 100644 (file)
--- a/doc/todo/info_show_total_annex_sizes_of_repositories.mdwn
+++ b/doc/todo/info_show_total_annex_sizes_of_repositories.mdwn
@@ -6,3 +6,5 @@ I think that this recently became possible to inplement cheaply.
 cachedAllRepoData is collected now, used for "combined annex size of all repositories".
 For this, it also needs to accumulate the size of each separate repository.
 --[[Joey]]
+
+> [[done]] --[[Joey]]