Separated unreachable objects by type in the output.

github · ScottArbeit · Apr 1, 2025 · Apr 1, 2025 · Apr 2, 2025 · Apr 2, 2025
commit 99ec1bfa62e68c023026ea02b3ad2c37f277061c
@@ -89,66 +89,71 @@ Is your Git repository bursting at the seams?
 
 ## Usage
 
-By default, `git-sizer` outputs its results in tabular format. For example, let's use it to analyze [the Linux repository](https://github.com/torvalds/linux), using the `--verbose` option so that all statistics are output:
+By default, `git-sizer` outputs its results in tabular format. For example, let's use it to analyze [the Linux repository](https://github.com/torvalds/linux) (as of April, 2025), using the `--verbose` option so that all statistics are output:
 
 ```
 $ git-sizer --verbose
-Processing blobs: 1652370
-Processing trees: 3396199
-Processing commits: 722647
-Matching commits to trees: 722647
-Processing annotated tags: 534
-Processing references: 539
+Processing blobs: 2928490
+Processing trees: 6510174
+Processing commits: 1351500
+Matching commits to trees: 1351500
+Processing annotated tags: 877
+Processing references: 883
+
 | Name                         | Value     | Level of concern               |
 | ---------------------------- | --------- | ------------------------------ |
-| Overall repository size      |           |                                |
+| Repository statistics        |           |                                |
 | * Commits                    |           |                                |
-|   * Count                    |   723 k   | *                              |
-|   * Total size               |   525 MiB | **                             |
+|   * Count                    |  1.35 M   | **                             |
+|   * Total size               |  1.11 GiB | ****                           |
 | * Trees                      |           |                                |
-|   * Count                    |  3.40 M   | **                             |
-|   * Total size               |  9.00 GiB | ****                           |
-|   * Total tree entries       |   264 M   | *****                          |
+|   * Count                    |  6.51 M   | ****                           |
+|   * Total size               |  19.0 GiB | **********                     |
+|   * Total tree entries       |   547 M   | **********                     |
 | * Blobs                      |           |                                |
-|   * Count                    |  1.65 M   | *                              |
-|   * Total size               |  55.8 GiB | *****                          |
+|   * Count                    |  2.93 M   | *                              |
+|   * Uncompressed total size  |   115 GiB | ************                   |
+| * On-disk size               |           |                                |
+|   * Compressed total size    |  5.68 GiB | ******                         |
 | * Annotated tags             |           |                                |
-|   * Count                    |   534     |                                |
+|   * Count                    |   877     |                                |
 | * References                 |           |                                |
-|   * Count                    |   539     |                                |
+|   * Count                    |   883     |                                |
+|     * Branches               |     1     |                                |
+|     * Tags                   |   880     |                                |
+|     * Remote-tracking refs   |     2     |                                |
 |                              |           |                                |
 | Biggest objects              |           |                                |
 | * Commits                    |           |                                |
 |   * Maximum size         [1] |  72.7 KiB | *                              |
 |   * Maximum parents      [2] |    66     | ******                         |
 | * Trees                      |           |                                |
-|   * Maximum entries      [3] |  1.68 k   | *                              |
+|   * Maximum entries      [3] |  2.60 k   | **                             |
 | * Blobs                      |           |                                |
-|   * Maximum size         [4] |  13.5 MiB | *                              |
+|   * Maximum size         [4] |  22.8 MiB | **                             |
 |                              |           |                                |
 | History structure            |           |                                |
-| * Maximum history depth      |   136 k   |                                |
+| * Maximum history depth      |   198 k   |                                |
 | * Maximum tag depth      [5] |     1     |                                |
 |                              |           |                                |
 | Biggest checkouts            |           |                                |
-| * Number of directories  [6] |  4.38 k   | **                             |
-| * Maximum path depth     [7] |    13     | *                              |
-| * Maximum path length    [8] |   134 B   | *                              |
-| * Number of files        [9] |  62.3 k   | *                              |
-| * Total size of files    [9] |   747 MiB |                                |
-| * Number of symlinks    [10] |    40     |                                |
+| * Number of directories  [6] |  5.89 k   | **                             |
+| * Maximum path depth     [6] |    14     | *                              |
+| * Maximum path length    [7] |   134 B   | *                              |
+| * Number of files        [8] |  88.7 k   | *                              |
+| * Total size of files    [9] |  1.41 GiB | *                              |
+| * Number of symlinks     [6] |    78     |                                |
 | * Number of submodules       |     0     |                                |
 
 [1]  91cc53b0c78596a73fa708cceb7313e7168bb146
 [2]  2cde51fbd0f310c8a2c5f977e665c0ac3945b46d
-[3]  4f86eed5893207aca2c2da86b35b38f2e1ec1fc8 (refs/heads/master:arch/arm/boot/dts)
-[4]  a02b6794337286bc12c907c33d5d75537c240bd0 (refs/heads/master:drivers/gpu/drm/amd/include/asic_reg/vega10/NBIO/nbio_6_1_sh_mask.h)
+[3]  ac1d84c335bcbd5fc5d82b8e985d8a9cc4c67d79 (6a1d798feb65d2a67e6e2cafb0b0e4f430603226:arch/arm/boot/dts)
+[4]  c20bf730dc553e5ae44ad9e769b1f8dface9fa9e (refs/heads/master:drivers/gpu/drm/amd/include/asic_reg/dcn/dcn_3_2_0_sh_mask.h)
 [5]  5dc01c595e6c6ec9ccda4f6f69c131c0dd945f8c (refs/tags/v2.6.11)
-[6]  1459754b9d9acc2ffac8525bed6691e15913c6e2 (589b754df3f37ca0a1f96fccde7f91c59266f38a^{tree})
-[7]  78a269635e76ed927e17d7883f2d90313570fdbc (dae09011115133666e47c35673c0564b0a702db7^{tree})
-[8]  ce5f2e31d3bdc1186041fdfd27a5ac96e728f2c5 (refs/heads/master^{tree})
-[9]  532bdadc08402b7a72a4b45a2e02e5c710b7d626 (e9ef1fe312b533592e39cddc1327463c30b0ed8d^{tree})
-[10] f29a5ea76884ac37e1197bef1941f62fda3f7b99 (f5308d1b83eba20e69df5e0926ba7257c8dd9074^{tree})
+[6]  549fc717f82345cf115dfa586ce076a8d1f296a6 (refs/heads/master^{tree})
+[7]  b0da5ce619daec8138cf92dfcf00e7a51ce856a9 (d8763340d2cb6262fb86424315a1f92cabc0e23c^{tree})
+[8]  fd94fec4e9c4e08df8e919e57fcc974c52c88c3c (3491aa04787f4d7e00da98d94b1b10001c398b5a^{tree})
+[9]  80e16948c5baba02ea2eeda7aa4b2478b68bbaf0 (524c03585fda36584cc7ada49a1827666d37eb4e^{tree})
 ```
 
 The output is a table showing the thing that was measured, its numerical value, and a rough indication of which values might be a cause for concern. In all cases, only objects that are reachable from references are included (i.e., not unreachable objects, nor objects that are reachable only from the reflogs).

@@ -47,7 +47,7 @@ const usage = `usage: git-sizer [OPTS] [ROOT...]
                                gitconfig: 'sizer.jsonVersion'.
       --[no-]progress          report (don't report) progress to stderr. Can
                                be set via gitconfig: 'sizer.progress'.
-      --include-unreachable    include unreachable objects
+      --include-unreachable    include unreachable objects in the analysis
       --version                only report the git-sizer version number
 
  Object selection:
@@ -353,8 +353,15 @@ func mainImplementation(ctx context.Context, stdout, stderr io.Writer, args []st
 		historySize.ShowUnreachable = true
 		unreachableStats, err := repo.GetUnreachableStats()
 		if err == nil {
-			historySize.UnreachableObjectCount = counts.Count64(unreachableStats.Count)
-			historySize.UnreachableObjectSize = counts.Count64(unreachableStats.Size)
+			// Store per-type unreachable stats for output
+			historySize.UnreachableBlobsCount = counts.Count64(unreachableStats.Blobs.Count)
+			historySize.UnreachableBlobsSize = counts.Count64(unreachableStats.Blobs.Size)
+			historySize.UnreachableTreesCount = counts.Count64(unreachableStats.Trees.Count)
+			historySize.UnreachableTreesSize = counts.Count64(unreachableStats.Trees.Size)
+			historySize.UnreachableCommitsCount = counts.Count64(unreachableStats.Commits.Count)
+			historySize.UnreachableCommitsSize = counts.Count64(unreachableStats.Commits.Size)
+			historySize.UnreachableTagsCount = counts.Count64(unreachableStats.Tags.Count)
+			historySize.UnreachableTagsSize = counts.Count64(unreachableStats.Tags.Size)
 		}
 	}
 
@@ -374,6 +381,8 @@ func mainImplementation(ctx context.Context, stdout, stderr io.Writer, args []st
 		}
 		fmt.Fprintf(stdout, "%s\n", j)
 	} else {
+		// Print a blank line between progress output and the table
+		fmt.Fprintln(stdout)
 		if _, err := io.WriteString(
 			stdout, historySize.TableString(rg.Groups(), threshold, nameStyle),
 		); err != nil {

@@ -177,45 +177,91 @@ func (repo *Repository) GitPath(relPath string) (string, error) {
 	return string(bytes.TrimSpace(out)), nil
 }
 
-// UnreachableStats holds the count and size of unreachable objects.
+// UnreachableStats holds the count and size of unreachable objects, broken out by type.
 type UnreachableStats struct {
-	Count int64
-	Size  int64
+	Blobs struct {
+		Count int64
+		Size  int64
+	}
+	Trees struct {
+		Count int64
+		Size  int64
+	}
+	Commits struct {
+		Count int64
+		Size  int64
+	}
+	Tags struct {
+		Count int64
+		Size  int64
+	}
 }
 
-// GetUnreachableStats runs 'git fsck --unreachable --no-reflogs --full'
-// and returns the count and total size of unreachable objects.
-// This implementation collects all OIDs from fsck output and then uses
-// batch mode to efficiently retrieve their sizes.
+// GetUnreachableStats runs 'git fsck --unreachable --no-reflogs'
+// and returns the count and total size of unreachable objects, broken out by type.
 func (repo *Repository) GetUnreachableStats() (UnreachableStats, error) {
-	// Run git fsck. Using CombinedOutput captures both stdout and stderr.
-	cmd := repo.GitCommand("fsck", "--unreachable", "--no-reflogs", "--full")
+	cmd := repo.GitCommand("fsck", "--unreachable", "--no-reflogs")
 	output, err := cmd.Output()
 	if err != nil {
-		return UnreachableStats{Count: 0, Size: 0}, fmt.Errorf(
-			"running 'git fsck --unreachable --no-reflogs --full': %w", err,
+		return UnreachableStats{}, fmt.Errorf(
+			"running 'git fsck --unreachable --no-reflogs': %w", err,
 		)
 	}
 
-	var oids []string
-	count := int64(0)
+	// Collect OIDs by type
+
+	oidsByType := map[string][]string{
+		"blob":   {},
+		"tree":   {},
+		"commit": {},
+		"tag":    {},
+	}
+	countsByType := map[string]int64{
+		"blob":   0,
+		"tree":   0,
+		"commit": 0,
+		"tag":    0,
+	}
+
 	for _, line := range bytes.Split(output, []byte{'\n'}) {
 		fields := bytes.Fields(line)
 		// Expected line format: "unreachable <type> <oid> ..."
 		if len(fields) >= 3 && string(fields[0]) == "unreachable" {
-			count++
-			oid := string(fields[2])
-			oids = append(oids, oid)
+			typeStr := string(fields[1])
+			if _, ok := oidsByType[typeStr]; ok {
+				oid := string(fields[2])
+				oidsByType[typeStr] = append(oidsByType[typeStr], oid)
+				countsByType[typeStr]++
+			}
 		}
 	}
 
-	// Retrieve the total size using batch mode.
-	totalSize, err := repo.getTotalSizeFromOids(oids)
-	if err != nil {
-		return UnreachableStats{}, fmt.Errorf("failed to get sizes via batch mode: %w", err)
+	var stats UnreachableStats
+	var errBlob, errTree, errCommit, errTag error
+	stats.Blobs.Count = countsByType["blob"]
+	stats.Trees.Count = countsByType["tree"]
+	stats.Commits.Count = countsByType["commit"]
+	stats.Tags.Count = countsByType["tag"]
+
+	stats.Blobs.Size, errBlob = repo.getTotalSizeFromOids(oidsByType["blob"])
+	stats.Trees.Size, errTree = repo.getTotalSizeFromOids(oidsByType["tree"])
+	stats.Commits.Size, errCommit = repo.getTotalSizeFromOids(oidsByType["commit"])
+	stats.Tags.Size, errTag = repo.getTotalSizeFromOids(oidsByType["tag"])
+
+	if errBlob != nil {
+		return stats, fmt.Errorf("failed to get blob sizes: %w", errBlob)
+	}
+	if errTree != nil {
+		return stats, fmt.Errorf("failed to get tree sizes: %w", errTree)
+	}
+	if errCommit != nil {
+		return stats, fmt.Errorf("failed to get commit sizes: %w", errCommit)
+	}
+	if errTag != nil {
+		return stats, fmt.Errorf("failed to get tag sizes: %w", errTag)
 	}
 
-	return UnreachableStats{Count: count, Size: totalSize}, nil
+	return stats, nil
 }
 
 // getTotalSizeFromOids uses 'git cat-file --batch-check' to retrieve sizes for

@@ -613,12 +613,38 @@ func (s *HistorySize) contents(refGroups []RefGroup) tableContents {
 	if s.ShowUnreachable {
 		sections = append(sections, S(
 			"Unreachable objects",
-			I("unreachableObjectCount", "Count",
-				"The total number of unreachable objects in the repository",
-				nil, s.UnreachableObjectCount, metric, "", 1e7),
-			I("unreachableObjectSize", "Uncompressed total size",
-				"The total size of unreachable objects in the repository",
-				nil, s.UnreachableObjectSize, binary, "B", 1e9),
+			S("Blobs",
+				I("unreachableBlobsCount", "Count",
+					"The total number of unreachable blobs in the repository",
+					nil, s.UnreachableBlobsCount, metric, "", 1.5e6),
+				I("unreachableBlobsSize", "Uncompressed total size",
+					"The total size of unreachable blobs in the repository",
+					nil, s.UnreachableBlobsSize, binary, "B", 1e9),
+			),
+			S("Trees",
+				I("unreachableTreesCount", "Count",
+					"The total number of unreachable trees in the repository",
+					nil, s.UnreachableTreesCount, metric, "", 1.5e6),
+				I("unreachableTreesSize", "Total size",
+					"The total size of unreachable trees in the repository",
+					nil, s.UnreachableTreesSize, binary, "B", 2e9),
+			),
+			S("Commits",
+				I("unreachableCommitsCount", "Count",
+					"The total number of unreachable commits in the repository",
+					nil, s.UnreachableCommitsCount, metric, "", 500e3),
+				I("unreachableCommitsSize", "Total size",
+					"The total size of unreachable commits in the repository",
+					nil, s.UnreachableCommitsSize, binary, "B", 250e6),
+			),
+			S("Tags",
+				I("unreachableTagsCount", "Count",
+					"The total number of unreachable tags in the repository",
+					nil, s.UnreachableTagsCount, metric, "", 25e3),
+				I("unreachableTagsSize", "Total size",
+					"The total size of unreachable tags in the repository",
+					nil, s.UnreachableTagsSize, binary, "B", 250e6),
+			),
 		))
 	}
 

@@ -214,11 +214,29 @@ type HistorySize struct {
 	// The actual size of the .git directory on disk.
 	GitDirSize counts.Count64 `json:"git_dir_size"`
 
-	// The total number of unreachable objects in the repository.
-	UnreachableObjectCount counts.Count64 `json:"unreachable_object_count"`
+	// The total number of unreachable blobs in the repository.
+	UnreachableBlobsCount counts.Count64 `json:"unreachable_blobs_count"`
 
-	// The total size of unreachable objects in the repository.
-	UnreachableObjectSize counts.Count64 `json:"unreachable_object_size"`
+	// The total size of unreachable blobs in the repository.
+	UnreachableBlobsSize counts.Count64 `json:"unreachable_blobs_size"`
+
+	// The total number of unreachable trees in the repository.
+	UnreachableTreesCount counts.Count64 `json:"unreachable_trees_count"`
+
+	// The total size of unreachable trees in the repository.
+	UnreachableTreesSize counts.Count64 `json:"unreachable_trees_size"`
+
+	// The total number of unreachable commits in the repository.
+	UnreachableCommitsCount counts.Count64 `json:"unreachable_commits_count"`
+
+	// The total size of unreachable commits in the repository.
+	UnreachableCommitsSize counts.Count64 `json:"unreachable_commits_size"`
+
+	// The total number of unreachable tags in the repository.
+	UnreachableTagsCount counts.Count64 `json:"unreachable_tags_count"`
+
+	// The total size of unreachable tags in the repository.
+	UnreachableTagsSize counts.Count64 `json:"unreachable_tags_size"`
 
 	ShowUnreachable bool `json:"-"`
 }