Compare commits

..
Author SHA1 Message Date
sneak 9e6e3b9287 Say in --help where the database is and what each report prints (closes #177)
check / check (push) Successful in 3m4s
The help of sfdupes and of scan names the database, its default path
and SFDUPES_DATABASE. The help of report and trees lists the columns,
the order of the rows and the four escapes. The help of info lists its
lines. The help of dwim says that it scans /, should run as root and
keeps the scan in the database.

The README's opening drops its sentence on copies deleted after the
scan, an edge case that its report and trees sections already cover.

Model: opus-5-5
2026-11-11 05:57:25 +00:00
4 changed files with 14 additions and 110 deletions

No files matched your search

+1 -2
View File
@@ -849,8 +849,7 @@ The summary line on stderr gives:
- the records read; - the records read;
- the duplicate-tree groups; - the duplicate-tree groups;
- the dupe trees, which is the number of rows; - the dupe trees, which is the number of rows;
- the reclaimable bytes, the sum of `size` over all rows, leaving out a row - the reclaimable bytes, the sum of `size` over all rows.
whose `dupe` lies inside another row's `dupe`.
``` ```
trees: 14 records read, 1 duplicate tree groups, 1 dupe trees, 3.0 KiB reclaimable trees: 14 records read, 1 duplicate tree groups, 1 dupe trees, 3.0 KiB reclaimable
-4
View File
@@ -29,10 +29,6 @@
Newest first. An entry ends with its date and its issue or branch, where those Newest first. An entry ends with its date and its issue or branch, where those
are known. are known.
- The reclaimable total of `trees` is the sum of `size` over all rows, leaving
out a row whose `dupe` lies inside another row's `dupe`. Before, the bytes of
such a row were counted twice (2026-11-11,
https://git.eeqj.de/sneak/sfdupes/issues/168)
- The help of `sfdupes` and of `scan` names the database, its default path and - The help of `sfdupes` and of `scan` names the database, its default path and
`SFDUPES_DATABASE`. The help of `report` and `trees` lists the columns, the `SFDUPES_DATABASE`. The help of `report` and `trees` lists the columns, the
order of the rows and the escapes. The help of `info` lists its lines. The order of the rows and the escapes. The help of `info` lists its lines. The
+12 -42
View File
@@ -101,7 +101,7 @@ func runTrees(ctx context.Context, stdout io.Writer) error {
label: treesPhase("write"), unit: "groups", total: int64(len(dupes)), label: treesPhase("write"), unit: "groups", total: int64(len(dupes)),
stalledOn: "out", stalledOn: "out",
}) })
dupeTrees, err := writeTreeGroups(stdout, dupes, prog) dupeTrees, reclaimable, err := writeTreeGroups(stdout, dupes, prog)
prog.finish() prog.finish()
@@ -112,7 +112,7 @@ func runTrees(ctx context.Context, stdout io.Writer) error {
fmt.Fprintf(os.Stderr, fmt.Fprintf(os.Stderr,
"trees: %d records read, %d duplicate tree groups, %d dupe trees, "+ "trees: %d records read, %d duplicate tree groups, %d dupe trees, "+
"%s reclaimable\n", "%s reclaimable\n",
records, len(dupes), dupeTrees, humanBytes(reclaimableBytes(dupes))) records, len(dupes), dupeTrees, humanBytes(reclaimable))
return nil return nil
} }
@@ -146,19 +146,22 @@ func readTree(ctx context.Context, db *sql.DB,
// a row for every member of each group but its first, counting each // a row for every member of each group but its first, counting each
// group on prog once its rows are written. The disk_size column is the // group on prog once its rows are written. The disk_size column is the
// dupe's own. stdout may be the terminal prog is drawn on, so the lines // dupe's own. stdout may be the terminal prog is drawn on, so the lines
// go through prog's writer. It returns the number of rows (dupe trees). // go through prog's writer. It returns the number of rows (dupe trees)
// and the bytes they could reclaim.
func writeTreeGroups(stdout io.Writer, dupes [][]*treeNode, func writeTreeGroups(stdout io.Writer, dupes [][]*treeNode,
prog *progress, prog *progress,
) (int, error) { ) (int, int64, error) {
out := prog.writer(stdout) out := prog.writer(stdout)
_, err := fmt.Fprintln(out, "first\tdupe\tfiles\tsize\tdisk_size") _, err := fmt.Fprintln(out, "first\tdupe\tfiles\tsize\tdisk_size")
if err != nil { if err != nil {
return 0, fmt.Errorf("write stdout: %w", err) return 0, 0, fmt.Errorf("write stdout: %w", err)
} }
dupeTrees := 0 dupeTrees := 0
var reclaimable int64
for _, g := range dupes { for _, g := range dupes {
first := g[0] first := g[0]
for _, n := range g[1:] { for _, n := range g[1:] {
@@ -166,10 +169,11 @@ func writeTreeGroups(stdout io.Writer, dupes [][]*treeNode,
escapePath(first.path), escapePath(n.path), escapePath(first.path), escapePath(n.path),
first.fileCount, first.totalSize, n.diskSize) first.fileCount, first.totalSize, n.diskSize)
if err != nil { if err != nil {
return 0, fmt.Errorf("write stdout: %w", err) return 0, 0, fmt.Errorf("write stdout: %w", err)
} }
dupeTrees++ dupeTrees++
reclaimable += first.totalSize
} }
prog.increment() prog.increment()
@@ -177,44 +181,10 @@ func writeTreeGroups(stdout io.Writer, dupes [][]*treeNode,
err = out.Flush() err = out.Flush()
if err != nil { if err != nil {
return 0, fmt.Errorf("write stdout: %w", err) return 0, 0, fmt.Errorf("write stdout: %w", err)
} }
return dupeTrees, nil return dupeTrees, reclaimable, nil
}
// reclaimableBytes returns the sum of the size column over the rows
// that writeTreeGroups writes for dupes, leaving out a row whose dupe
// lies inside another row's dupe.
func reclaimableBytes(dupes [][]*treeNode) int64 {
isDupe := make(map[*treeNode]bool)
for _, g := range dupes {
for _, n := range g[1:] {
isDupe[n] = true
}
}
var total int64
for n := range isDupe {
if !insideDupe(n, isDupe) {
total += n.totalSize
}
}
return total
}
// insideDupe reports whether a directory above n is in isDupe.
func insideDupe(n *treeNode, isDupe map[*treeNode]bool) bool {
for p := n.parent; p != nil; p = p.parent {
if isDupe[p] {
return true
}
}
return false
} }
// treeBuilder reconstructs the directory hierarchy from records added // treeBuilder reconstructs the directory hierarchy from records added
+1 -62
View File
@@ -313,7 +313,7 @@ func TestWriteTreeGroupsOnTerminal(t *testing.T) {
label: treesPhase("write"), unit: "groups", total: int64(len(dupes)), label: treesPhase("write"), unit: "groups", total: int64(len(dupes)),
}, terminalWidth, time.Hour, time.Now) }, terminalWidth, time.Hour, time.Now)
_, err := writeTreeGroups(os.Stderr, dupes, p) _, _, err := writeTreeGroups(os.Stderr, dupes, p)
p.finish() p.finish()
@@ -587,64 +587,3 @@ func TestCollectTreeGroupsKeepsGroupLeftWithTwoCopies(t *testing.T) {
t.Errorf("groups = %v, want %v", got, want) t.Errorf("groups = %v, want %v", got, want)
} }
} }
// TestReclaimableBytesCountsNestedDupeOnce checks that a dupe tree
// inside another dupe tree adds nothing to the reclaimable bytes. /B/x
// lies in /B, so the total adds the sizes of /B and /C/x, 1500 + 1000.
func TestReclaimableBytesCountsNestedDupeOnce(t *testing.T) {
t.Parallel()
fh, ft, fc := testHash("fh"), testHash("ft"), testHash("fc")
gh, gt, gc := testHash("gh"), testHash("gt"), testHash("gc")
recs := []scanRec{
{size: 1000, head: fh, tail: ft, content: fc, path: "/A/x/f"},
{size: 500, head: gh, tail: gt, content: gc, path: "/A/y/g"},
{size: 1000, head: fh, tail: ft, content: fc, path: "/B/x/f"},
{size: 500, head: gh, tail: gt, content: gc, path: "/B/y/g"},
{size: 1000, head: fh, tail: ft, content: fc, path: "/C/x/f"},
}
super, dirs := treeOf(t, recs)
dupes := collectTreeGroups(dirs, super, everyPathExists, nil)
wantGroups := [][]string{{"/A", "/B"}, {"/A/x", "/B/x", "/C/x"}}
if got := groupPaths(dupes); !slices.EqualFunc(got, wantGroups,
slices.Equal) {
t.Fatalf("groups = %v, want %v", got, wantGroups)
}
if got := reclaimableBytes(dupes); got != 2500 {
t.Errorf("reclaimableBytes = %d, want 2500", got)
}
}
// TestReclaimableBytesCountsDupeInsideGrandparentDupeOnce checks that a
// dupe tree adds nothing when a dupe tree above its parent holds it.
// /B/m/x lies in /B, and /B/m is no dupe, since the group of /A and /B
// implies that of /A/m and /B/m. The total adds the sizes of /B and
// /C/x, 1300 + 1000.
func TestReclaimableBytesCountsDupeInsideGrandparentDupeOnce(t *testing.T) {
t.Parallel()
f, w := testHash("f"), testHash("w")
recs := []scanRec{
{size: 1000, head: f, tail: f, content: f, path: "/A/m/x/f"},
{size: 300, head: w, tail: w, content: w, path: "/A/m/w"},
{size: 1000, head: f, tail: f, content: f, path: "/B/m/x/f"},
{size: 300, head: w, tail: w, content: w, path: "/B/m/w"},
{size: 1000, head: f, tail: f, content: f, path: "/C/x/f"},
}
super, dirs := treeOf(t, recs)
dupes := collectTreeGroups(dirs, super, everyPathExists, nil)
wantGroups := [][]string{{"/A", "/B"}, {"/A/m/x", "/B/m/x", "/C/x"}}
if got := groupPaths(dupes); !slices.EqualFunc(got, wantGroups,
slices.Equal) {
t.Fatalf("groups = %v, want %v", got, wantGroups)
}
if got := reclaimableBytes(dupes); got != 2300 {
t.Errorf("reclaimableBytes = %d, want 2300", got)
}
}