[#issue 3490] Support for specifying file size in read-data-subset

changelog/unreleased/issue-3490

@@ -0,0 +1,9 @@
+Enhancement: Support for specifying file size in `check --read-data-subset`
+
+To check a subset of repository files, the `check --read-data-subset` command
+used to support two ways to select a subset - A specific range of pack files, 
+or random percentage of pack files. We have added a third method to select pack 
+files - By specifying file size. This new option is available with the 'restic check' command.
+
+https://github.com/restic/restic/issues/3490
+https://github.com/restic/restic/pull/3548

cmd/restic/cmd_check.go

@@ -65,7 +65,7 @@ func checkFlags(opts CheckOptions) error {
 	}
 	if opts.ReadDataSubset != "" {
 		dataSubset, err := stringToIntSlice(opts.ReadDataSubset)
-		argumentError := errors.Fatal("check flag --read-data-subset must have two positive integer values or a percentage, e.g. --read-data-subset=1/2 or --read-data-subset=2.5%%")
+		argumentError := errors.Fatal("check flag --read-data-subset must have two positive integer values or a percentage or a file size, e.g. --read-data-subset=1/2 or --read-data-subset=2.5%% or --read-data-subset=10G")
 		if err == nil {
 			if len(dataSubset) != 2 {
 				return argumentError
@@ -76,7 +76,7 @@ func checkFlags(opts CheckOptions) error {
 			if dataSubset[1] > totalBucketsMax {
 				return errors.Fatalf("check flag --read-data-subset=n/t t must be at most %d", totalBucketsMax)
 			}
-		} else {
+		} else if strings.HasSuffix(opts.ReadDataSubset, "%") {
 			percentage, err := parsePercentage(opts.ReadDataSubset)
 			if err != nil {
 				return argumentError
@@ -86,6 +86,17 @@ func checkFlags(opts CheckOptions) error {
 				return errors.Fatal(
 					"check flag --read-data-subset=n% n must be above 0.0% and at most 100.0%")
 			}
+
+		} else {
+			fileSize, err := parseSizeStr(opts.ReadDataSubset)
+			if err != nil {
+				return argumentError
+			}
+			if fileSize <= 0.0 {
+				return errors.Fatal(
+					"check flag --read-data-subset=n n must be above 0.0")
+			}
+
 		}
 	}
 
@@ -294,10 +305,27 @@ func runCheck(opts CheckOptions, gopts GlobalOptions, args []string) error {
 			packs = selectPacksByBucket(chkr.GetPacks(), bucket, totalBuckets)
 			packCount := uint64(len(packs))
 			Verbosef("read group #%d of %d data packs (out of total %d packs in %d groups)\n", bucket, packCount, chkr.CountPacks(), totalBuckets)
+		} else if strings.HasSuffix(opts.ReadDataSubset, "%") {
+			percentage, err := parsePercentage(opts.ReadDataSubset)
+			if err == nil {
+				packs = selectRandomPacksByPercentage(chkr.GetPacks(), percentage)
+				Verbosef("read %.1f%% of data packs\n", percentage)
+			}
 		} else {
-			percentage, _ := parsePercentage(opts.ReadDataSubset)
-			packs = selectRandomPacksByPercentage(chkr.GetPacks(), percentage)
-			Verbosef("read %.1f%% of data packs\n", percentage)
+			repoSize := int64(0)
+			allPacks := chkr.GetPacks()
+			for _, size := range allPacks {
+				repoSize += size
+			}
+			if repoSize == 0 {
+				return errors.Fatal("Cannot read from a repository having size 0")
+			}
+			subsetSize, _ := parseSizeStr(opts.ReadDataSubset)
+			if subsetSize > repoSize {
+				subsetSize = repoSize
+			}
+			packs = selectRandomPacksByFileSize(chkr.GetPacks(), subsetSize, repoSize)
+			Verbosef("read %d bytes of data packs\n", subsetSize)
 		}
 		if packs == nil {
 			return errors.Fatal("internal error: failed to select packs to check")
@@ -349,6 +377,11 @@ func selectRandomPacksByPercentage(allPacks map[restic.ID]int64, percentage floa
 		id := keys[idx[i]]
 		packs[id] = allPacks[id]
 	}
-
+	return packs
+}
+
+func selectRandomPacksByFileSize(allPacks map[restic.ID]int64, subsetSize int64, repoSize int64) map[restic.ID]int64 {
+	subsetPercentage := (float64(subsetSize) / float64(repoSize)) * 100.0
+	packs := selectRandomPacksByPercentage(allPacks, subsetPercentage)
 	return packs
 }

cmd/restic/cmd_check_test.go

@@ -129,3 +129,37 @@ func TestSelectNoRandomPacksByPercentage(t *testing.T) {
 	selectedPacks := selectRandomPacksByPercentage(testPacks, 10.0)
 	rtest.Assert(t, len(selectedPacks) == 0, "Expected 0 selected packs")
 }
+
+func TestSelectRandomPacksByFileSize(t *testing.T) {
+	var testPacks = make(map[restic.ID]int64)
+	for i := 1; i <= 10; i++ {
+		id := restic.NewRandomID()
+		// ensure unique ids
+		id[0] = byte(i)
+		testPacks[id] = 0
+	}
+
+	selectedPacks := selectRandomPacksByFileSize(testPacks, 10, 500)
+	rtest.Assert(t, len(selectedPacks) == 1, "Expected 1 selected packs")
+
+	selectedPacks = selectRandomPacksByFileSize(testPacks, 10240, 51200)
+	rtest.Assert(t, len(selectedPacks) == 2, "Expected 2 selected packs")
+	for pack := range selectedPacks {
+		_, ok := testPacks[pack]
+		rtest.Assert(t, ok, "Unexpected selection")
+	}
+
+	selectedPacks = selectRandomPacksByFileSize(testPacks, 500, 500)
+	rtest.Assert(t, len(selectedPacks) == 10, "Expected 10 selected packs")
+	for pack := range selectedPacks {
+		_, ok := testPacks[pack]
+		rtest.Assert(t, ok, "Unexpected item in selection")
+	}
+}
+
+func TestSelectNoRandomPacksByFileSize(t *testing.T) {
+	// that the a repository without pack files works
+	var testPacks = make(map[restic.ID]int64)
+	selectedPacks := selectRandomPacksByFileSize(testPacks, 10, 500)
+	rtest.Assert(t, len(selectedPacks) == 0, "Expected 0 selected packs")
+}

doc/045_working_with_repos.rst

@@ -249,9 +249,9 @@ integrity of the pack files in the repository, use the ``--read-data`` flag:
     and also that it takes more time than the default ``check``.
 
 Alternatively, use the ``--read-data-subset`` parameter to check only a
-subset of the repository pack files at a time. It supports two ways to select a
-subset. One selects a specific range of pack files, the other selects a random
-percentage of pack files.
+subset of the repository pack files at a time. It supports three ways to select a
+subset. One selects a specific range of pack files, the second selects a random
+percentage of pack files, and the third selects pack files of the specified size.
 
 Use ``--read-data-subset=n/t`` to check only a subset of the repository pack
 files at a time. The parameter takes two values, ``n`` and ``t``. When the check
@@ -285,3 +285,16 @@ integer:
 .. code-block:: console
 
     $ restic -r /srv/restic-repo check --read-data-subset=10%
+
+Use ``--read-data-subset=NS`` to check a randomly chosen subset of the repository pack files. 
+It takes one parameter, ``NS``, where 'N' is a whole number representing file size and 'S' is the unit 
+of file size (B/K/M/G/T) of pack files to check. Behind the scenes, the specified size will be converted 
+to percentage of the total repository size. The behaviour of the check command following this conversion 
+will be the same as the percentage option above. For a file size value the following command may be used:
+
+.. code-block:: console
+
+    $ restic -r /srv/restic-repo check --read-data-subset=50M
+    $ restic -r /srv/restic-repo check --read-data-subset=10G
+
+