jgaunt/rclone
1
0
Fork 0
mirror of https://github.com/rclone/rclone.git synced 2026-01-06 10:33:34 +00:00
Code Issues Projects Releases Wiki Activity

bisync: full support for comparing checksum, size, modtime - fixes #5679 fixes #5683 fixes #5684 fixes #5675

Browse Source 
Before this change, bisync could only detect changes based on modtime, and
would refuse to run if either path lacked modtime support. This made bisync
unavailable for many of rclone's backends. Additionally, bisync did not account
for the Fs's precision when comparing modtimes, meaning that they could only be
reliably compared within the same side -- not against the opposite side. Size
and checksum (even when available) were ignored completely for deltas.

After this change, bisync now fully supports comparing based on any combination
of size, modtime, and checksum, lifting the prior restriction on backends
without modtime support. The comparison logic considers the backend's
precision, hash types, and other features as appropriate.

The comparison features optionally use a new --compare flag (which takes any
combination of size,modtime,checksum) and even supports some combinations not
otherwise supported in `sync` (like comparing all three at the same time.) By
default (without the --compare flag), bisync inherits the same comparison
options as `sync` (that is: size and modtime by default, unless modified with
flags such as --checksum or --size-only.) If the --compare flag is set, it will
override these defaults.

If --compare includes checksum and both remotes support checksums but have no
hash types in common with each other, checksums will be considered only for
comparisons within the same side (to determine what has changed since the prior
sync), but not for comparisons against the opposite side. If one side supports
checksums and the other does not, checksums will only be considered on the side
that supports them. When comparing with checksum and/or size without modtime,
bisync cannot determine whether a file is newer or older -- only whether it is
changed or unchanged. (If it is changed on both sides, bisync still does the
standard equality-check to avoid declaring a sync conflict unless it absolutely
has to.)

Also included are some new flags to customize the checksum comparison behavior
on backends where hashes are slow or unavailable. --no-slow-hash and
--slow-hash-sync-only allow selectively ignoring checksums on backends such as
local where they are slow. --download-hash allows computing them by downloading
when (and only when) they're otherwise not available. Of course, this option
probably won't be practical with large files, but may be a good option for
syncing small-but-important files with maximum accuracy (for example, a source
code repo on a crypt remote.) An additional advantage over methods like
cryptcheck is that the original file is not required for comparison (for
example, --download-hash can be used to bisync two different crypt remotes with
different passwords.)

Additionally, all of the above are now considered during the final --check-sync
for much-improved accuracy (before this change, it only compared filenames!)

Many other details are explained in the included docs.
This commit is contained in:
nielash
2023-11-30 19:44:38 -05:00
parent d8e07bfd8e
commit b4216648e4
308 changed files with 5469 additions and 3243 deletions
Download Patch File Download Diff File Expand all files Collapse all files

170
cmd/bisync/deltas.go
View File

@@ -13,6 +13,7 @@ import (
"github.com/rclone/rclone/fs"
"github.com/rclone/rclone/fs/filter"
"github.com/rclone/rclone/fs/operations"
"github.com/rclone/rclone/lib/terminal"
"golang.org/x/text/unicode/norm"
)
@@ -24,14 +25,17 @@ const (
deltaNew delta = 1 << iota
deltaNewer
deltaOlder
deltaSize
deltaLarger
deltaSmaller
deltaHash
deltaDeleted
)
const (
deltaModified delta = deltaNewer | deltaOlder | deltaSize | deltaHash | deltaDeleted
deltaOther delta = deltaNew | deltaNewer | deltaOlder
deltaSize delta = deltaLarger | deltaSmaller
deltaTime delta = deltaNewer | deltaOlder
deltaModified delta = deltaTime | deltaSize | deltaHash
deltaOther delta = deltaNew | deltaTime | deltaSize | deltaHash
)
func (d delta) is(cond delta) bool {
@@ -41,6 +45,8 @@ func (d delta) is(cond delta) bool {
// deltaSet
type deltaSet struct {
deltas map[string]delta
size map[string]int64
hash map[string]string
opt *Options
fs fs.Fs // base filesystem
msg string // filesystem name for logging
@@ -72,25 +78,71 @@ func (ds *deltaSet) printStats() {
}
nAll := len(ds.deltas)
nNew := 0
nMod := 0
nTime := 0
nNewer := 0
nOlder := 0
nSize := 0
nLarger := 0
nSmaller := 0
nHash := 0
nDeleted := 0
for _, d := range ds.deltas {
if d.is(deltaNew) {
nNew++
}
if d.is(deltaModified) {
nMod++
}
if d.is(deltaTime) {
nTime++
}
if d.is(deltaNewer) {
nNewer++
}
if d.is(deltaOlder) {
nOlder++
}
if d.is(deltaSize) {
nSize++
}
if d.is(deltaLarger) {
nLarger++
}
if d.is(deltaSmaller) {
nSmaller++
}
if d.is(deltaHash) {
nHash++
}
if d.is(deltaDeleted) {
nDeleted++
}
}
fs.Infof(nil, "%s: %4d changes: %4d new, %4d newer, %4d older, %4d deleted",
ds.msg, nAll, nNew, nNewer, nOlder, nDeleted)
if nAll != nNew+nMod+nDeleted {
fs.Errorf(nil, "something doesn't add up! %4d != %4d + %4d + %4d", nAll, nNew, nMod, nDeleted)
}
fs.Infof(nil, "%s: %4d changes: "+Color(terminal.GreenFg, "%4d new")+", "+Color(terminal.YellowFg, "%4d modified")+", "+Color(terminal.RedFg, "%4d deleted"),
ds.msg, nAll, nNew, nMod, nDeleted)
if nMod > 0 {
details := []string{}
if nTime > 0 {
details = append(details, fmt.Sprintf(Color(terminal.CyanFg, "%4d newer"), nNewer))
details = append(details, fmt.Sprintf(Color(terminal.BlueFg, "%4d older"), nOlder))
}
if nSize > 0 {
details = append(details, fmt.Sprintf(Color(terminal.CyanFg, "%4d larger"), nLarger))
details = append(details, fmt.Sprintf(Color(terminal.BlueFg, "%4d smaller"), nSmaller))
}
if nHash > 0 {
details = append(details, fmt.Sprintf(Color(terminal.CyanFg, "%4d hash differs"), nHash))
}
if (nNewer+nOlder != nTime) || (nLarger+nSmaller != nSize) || (nMod > nTime+nSize+nHash) {
fs.Errorf(nil, "something doesn't add up!")
}
fs.Infof(nil, "(%s: %s)", Color(terminal.YellowFg, "Modified"), strings.Join(details, ", "))
}
}
// findDeltas
@@ -117,6 +169,8 @@ func (b *bisyncRun) findDeltas(fctx context.Context, f fs.Fs, oldListing string,
ds = &deltaSet{
deltas: map[string]delta{},
size: map[string]int64{},
hash: map[string]string{},
fs: f,
msg: msg,
oldCount: len(old.list),
@@ -125,31 +179,70 @@ func (b *bisyncRun) findDeltas(fctx context.Context, f fs.Fs, oldListing string,
}
for _, file := range old.list {
// REMEMBER: this section is only concerned with comparing listings from the same side (not different sides)
d := deltaZero
s := int64(0)
h := ""
if !now.has(file) {
b.indent(msg, file, "File was deleted")
b.indent(msg, file, Color(terminal.RedFg, "File was deleted"))
ds.deleted++
d |= deltaDeleted
} else {
// skip dirs here, as we only care if they are new/deleted, not newer/older
if !now.isDir(file) {
if old.getTime(file) != now.getTime(file) {
if old.beforeOther(now, file) {
fs.Debugf(file, "(old: %v current: %v", old.getTime(file), now.getTime(file))
b.indent(msg, file, "File is newer")
d |= deltaNewer
} else { // Current version is older than prior sync.
fs.Debugf(file, "(old: %v current: %v", old.getTime(file), now.getTime(file))
b.indent(msg, file, "File is OLDER")
d |= deltaOlder
whatchanged := []string{}
if b.opt.Compare.Size {
if sizeDiffers(old.getSize(file), now.getSize(file)) {
fs.Debugf(file, "(old: %v current: %v)", old.getSize(file), now.getSize(file))
if now.getSize(file) > old.getSize(file) {
whatchanged = append(whatchanged, Color(terminal.MagentaFg, "size (larger)"))
d |= deltaLarger
} else {
whatchanged = append(whatchanged, Color(terminal.MagentaFg, "size (smaller)"))
d |= deltaSmaller
}
s = now.getSize(file)
}
}
// TODO Compare sizes and hashes
if b.opt.Compare.Modtime {
if timeDiffers(fctx, old.getTime(file), now.getTime(file), f, f) {
if old.beforeOther(now, file) {
fs.Debugf(file, "(old: %v current: %v)", old.getTime(file), now.getTime(file))
whatchanged = append(whatchanged, Color(terminal.MagentaFg, "time (newer)"))
d |= deltaNewer
} else { // Current version is older than prior sync.
fs.Debugf(file, "(old: %v current: %v)", old.getTime(file), now.getTime(file))
whatchanged = append(whatchanged, Color(terminal.MagentaFg, "time (older)"))
d |= deltaOlder
}
}
}
if b.opt.Compare.Checksum {
if hashDiffers(old.getHash(file), now.getHash(file), old.hash, now.hash, old.getSize(file), now.getSize(file)) {
fs.Debugf(file, "(old: %v current: %v)", old.getHash(file), now.getHash(file))
whatchanged = append(whatchanged, Color(terminal.MagentaFg, "hash"))
d |= deltaHash
h = now.getHash(file)
}
}
// concat changes and print log
if d.is(deltaModified) {
summary := fmt.Sprintf(Color(terminal.YellowFg, "File changed: %s"), strings.Join(whatchanged, ", "))
b.indent(msg, file, summary)
}
}
}
if d.is(deltaModified) {
ds.deltas[file] = d
if b.opt.Compare.Size {
ds.size[file] = s
}
if b.opt.Compare.Checksum {
ds.hash[file] = h
}
} else if d.is(deltaDeleted) {
ds.deltas[file] = d
} else {
// Once we've found at least one unchanged file,
// we know that not everything has changed,
@@ -160,8 +253,14 @@ func (b *bisyncRun) findDeltas(fctx context.Context, f fs.Fs, oldListing string,
for _, file := range now.list {
if !old.has(file) {
b.indent(msg, file, "File is new")
b.indent(msg, file, Color(terminal.GreenFg, "File is new"))
ds.deltas[file] = deltaNew
if b.opt.Compare.Size {
ds.size[file] = now.getSize(file)
}
if b.opt.Compare.Checksum {
ds.hash[file] = now.getHash(file)
}
}
}
@@ -234,20 +333,28 @@ func (b *bisyncRun) applyDeltas(ctx context.Context, ds1, ds2 *deltaSet) (change
d1 := ds1.deltas[file]
if d1.is(deltaOther) {
d2, in2 := ds2.deltas[file]
file2 := file
if !in2 && file != alias {
d2 = ds2.deltas[alias]
file2 = alias
}
if d2.is(deltaOther) {
checkit := func(filename string) {
if err := filterCheck.AddFile(filename); err != nil {
fs.Debugf(nil, "Non-critical error adding file to list of potential conflicts to check: %s", err)
} else {
fs.Debugf(nil, "Added file to list of potential conflicts to check: %s", filename)
// if size or hash differ, skip this, as we already know they're not equal
if (b.opt.Compare.Size && sizeDiffers(ds1.size[file], ds2.size[file2])) ||
(b.opt.Compare.Checksum && hashDiffers(ds1.hash[file], ds2.hash[file2], b.opt.Compare.HashType1, b.opt.Compare.HashType2, ds1.size[file], ds2.size[file2])) {
fs.Debugf(file, "skipping equality check as size/hash definitely differ")
} else {
checkit := func(filename string) {
if err := filterCheck.AddFile(filename); err != nil {
fs.Debugf(nil, "Non-critical error adding file to list of potential conflicts to check: %s", err)
} else {
fs.Debugf(nil, "Added file to list of potential conflicts to check: %s", filename)
}
}
checkit(file)
if file != alias {
checkit(alias)
}
}
checkit(file)
if file != alias {
checkit(alias)
}
}
}
@@ -294,6 +401,17 @@ func (b *bisyncRun) applyDeltas(ctx context.Context, ds1, ds2 *deltaSet) (change
// the Path1 version is deemed "correct" in this scenario
fs.Infof(alias, "Files are equal but will copy anyway to fix case to %s", file)
copy1to2.Add(file)
} else if b.opt.Compare.Modtime && timeDiffers(ctx, ls1.getTime(ls1.getTryAlias(file, alias)), ls2.getTime(ls2.getTryAlias(file, alias)), b.fs1, b.fs2) {
fs.Infof(file, "Files are equal but will copy anyway to update modtime (will not rename)")
if ls1.getTime(ls1.getTryAlias(file, alias)).Before(ls2.getTime(ls2.getTryAlias(file, alias))) {
// Path2 is newer
b.indent("Path2", p1, "Queue copy to Path1")
copy2to1.Add(ls2.getTryAlias(file, alias))
} else {
// Path1 is newer
b.indent("Path1", p2, "Queue copy to Path2")
copy1to2.Add(ls1.getTryAlias(file, alias))
}
} else {
fs.Infof(nil, "Files are equal! Skipping: %s", file)
renameSkipped.Add(file)