Support for hardlinks to symlinks

This is a specialized use-case, but it is indeed possible to do this
and can be used if xattrs are attached to the symlink.

src/duplicacy_backupmanager.go

@@ -705,7 +705,7 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 
 	type hardLinkEntry struct {
 		entry *Entry
-		willDownload bool
+		willExist bool
 	}
 	var hardLinkTable []hardLinkEntry
 	var hardLinks []*Entry
@@ -752,12 +752,16 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 		}
 
 		fullPath := joinPath(top, remoteEntry.Path)
+
 		if remoteEntry.IsLink() {
 			if stat, _ := os.Lstat(fullPath); stat != nil {
 				if stat.Mode()&os.ModeSymlink != 0 {
 					isRegular, link, err := Readlink(fullPath)
 					if err == nil && link == remoteEntry.Link && !isRegular {
 						remoteEntry.RestoreMetadata(fullPath, nil, setOwner)
+						if remoteEntry.IsHardlinkRoot() {
+							hardLinkTable[len(hardLinkTable)-1].willExist = true
+						}
 						continue
 					}
 				}
@@ -771,11 +775,33 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 				os.Remove(fullPath)
 			}
 
+			if remoteEntry.IsHardlinkRoot() {
+				hardLinkTable[len(hardLinkTable)-1].willExist = true
+			} else if remoteEntry.IsHardlinkedFrom() {
+				i, err := remoteEntry.GetHardlinkId()
+				if err !=  nil {
+					LOG_ERROR("RESTORE_HARDLINK", "Decode error for hardlinked entry %s, %v", remoteEntry.Path, err)
+					return 0
+				}
+				if !hardLinkTable[i].willExist {
+					hardLinkTable[i] = hardLinkEntry{remoteEntry, true}
+				} else {
+					sourcePath := joinPath(top, hardLinkTable[i].entry.Path)
+					LOG_INFO("RESTORE_HARDLINK", "Hard linking %s to %s", fullPath, sourcePath)
+					if err := MakeHardlink(sourcePath, fullPath); err != nil {
+						LOG_ERROR("RESTORE_HARDLINK", "Failed to create hard link %s to %s %v", fullPath, sourcePath, err)
+						return 0
+					}
+					continue
+				}
+			}
+
 			if err := os.Symlink(remoteEntry.Link, fullPath); err != nil {
 				LOG_ERROR("RESTORE_SYMLINK", "Can't create symlink %s: %v", remoteEntry.Path, err)
 				return 0
 			}
 			remoteEntry.RestoreMetadata(fullPath, nil, setOwner)
+
 			LOG_TRACE("DOWNLOAD_DONE", "Symlink %s updated", remoteEntry.Path)
 		} else if remoteEntry.IsDir() {
 
@@ -814,20 +840,21 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 			remoteEntry.RestoreMetadata(fullPath, nil, setOwner)
 		} else {
 			if remoteEntry.IsHardlinkRoot() {
-				hardLinkTable[len(hardLinkTable)-1] = hardLinkEntry{remoteEntry, true}
+				hardLinkTable[len(hardLinkTable)-1].willExist = true
 			} else if remoteEntry.IsHardlinkedFrom() {
-				i, err := strconv.ParseUint(remoteEntry.Link, 16, 64)
+				i, err := remoteEntry.GetHardlinkId()
 				if err !=  nil {
-					LOG_ERROR("RESTORE_HARDLINK", "Decode error in hardlink entry, expected hex int, got %s", remoteEntry.Link)
+					LOG_ERROR("RESTORE_HARDLINK", "Decode error for hardlinked entry %s, %v", remoteEntry.Path, err)
 					return 0
 				}
-				if !hardLinkTable[i].willDownload {
+				if !hardLinkTable[i].willExist {
 					hardLinkTable[i] = hardLinkEntry{remoteEntry, true}
 				} else {
 					hardLinks = append(hardLinks, remoteEntry)
 					continue
 				}
 			}
+
 			// We can't download files here since fileEntries needs to be sorted
 			fileEntries = append(fileEntries, remoteEntry)
 			totalFileSize += remoteEntry.Size
@@ -948,11 +975,32 @@ func (manager *BackupManager) Restore(top string, revision int, inPlace bool, qu
 	}
 
 	for _, linkEntry := range hardLinks {
-		i, _ := strconv.ParseUint(linkEntry.Link, 16, 64)
+
+		i, _ := linkEntry.GetHardlinkId()
 		sourcePath := joinPath(top, hardLinkTable[i].entry.Path)
 		fullPath := joinPath(top, linkEntry.Path)
-		LOG_INFO("RESTORE_HARDLINK", "Hard linking %s to %s", fullPath, sourcePath)
-		if err := os.Link(sourcePath, fullPath); err != nil {
+
+		if stat, _ := os.Lstat(fullPath); stat != nil {
+			sourceStat, _ := os.Lstat(sourcePath)
+			if os.SameFile(stat, sourceStat) {
+				continue
+			}
+
+			if sourceStat == nil {
+				LOG_WERROR(allowFailures, "RESTORE_HARDLINK",
+					"Target %s for hardlink %s is missing", sourcePath, linkEntry.Path)
+				continue
+			}
+			if !overwrite {
+				LOG_WERROR(allowFailures, "DOWNLOAD_OVERWRITE",
+					"File %s already exists.  Please specify the -overwrite option to overwrite", linkEntry.Path)
+				continue
+			}
+			os.Remove(fullPath)
+		}
+
+		LOG_DEBUG("RESTORE_HARDLINK", "Hard linking %s to %s", fullPath, sourcePath)
+		if err := MakeHardlink(sourcePath, fullPath); err != nil {
 			LOG_ERROR("RESTORE_HARDLINK", "Failed to create hard link %s to %s", fullPath, sourcePath)
 			return 0
 		}
@@ -1142,10 +1190,10 @@ func (manager *BackupManager) UploadSnapshot(chunkOperator *ChunkOperator, top s
 			entry.StartChunk -= lastEndChunk
 			lastEndChunk = entry.EndChunk
 			entry.EndChunk = delta
-		} else if entry.IsHardlinkedFrom() {
-			i, err := strconv.ParseUint(entry.Link, 16, 64)
+		} else if entry.IsHardlinkedFrom() && !entry.IsLink() {
+			i, err := entry.GetHardlinkId()
 			if err !=  nil {
-				LOG_ERROR("SNAPSHOT_UPLOAD", "Decode error in hardlink entry, expected hex int, got %s", entry.Link)
+				LOG_ERROR("SNAPSHOT_UPLOAD", "Decode error for hardlinked entry %s, %v", entry.Link, err)
 				return err
 			}
 

src/duplicacy_entry.go

@@ -8,6 +8,7 @@ import (
 	"crypto/sha256"
 	"encoding/base64"
 	"encoding/json"
+	"errors"
 	"fmt"
 	"io/ioutil"
 	"os"
@@ -23,6 +24,11 @@ import (
 	"github.com/vmihailenco/msgpack"
 )
 
+const (
+	entrySymHardLinkRootChunkMarker   = -72
+	entrySymHardLinkTargetChunkMarker = -73
+)
+
 // This is the hidden directory in the repository for storing various files.
 var DUPLICACY_DIRECTORY = ".duplicacy"
 var DUPLICACY_FILE = ".duplicacy"
@@ -518,11 +524,23 @@ func (entry *Entry) IsComplete() bool {
 }
 
 func (entry *Entry) IsHardlinkedFrom() bool {
-	return entry.IsFile() && len(entry.Link) > 0 && entry.Link != "/"
+	return (entry.IsFile() && len(entry.Link) > 0 && entry.Link != "/") || (entry.IsLink() && entry.StartChunk == entrySymHardLinkTargetChunkMarker)
 }
 
 func (entry *Entry) IsHardlinkRoot() bool {
-	return entry.IsFile() && entry.Link == "/"
+	return (entry.IsFile() && entry.Link == "/") || (entry.IsLink() && entry.StartChunk == entrySymHardLinkRootChunkMarker)
+}
+
+func (entry *Entry) GetHardlinkId() (int, error) {
+	if entry.IsLink() {
+		if entry.StartChunk != entrySymHardLinkTargetChunkMarker {
+			return 0, errors.New("Symlink entry not marked as hardlinked")
+		}
+		return entry.StartOffset, nil
+	} else {
+		i, err := strconv.ParseUint(entry.Link, 16, 64)
+		return int(i), err
+	}
 }
 
 func (entry *Entry) GetPermissions() os.FileMode {
@@ -789,6 +807,42 @@ func ListEntries(top string, path string, patterns []string, nobackupFile string
 		if len(patterns) > 0 && !MatchPath(entry.Path, patterns) {
 			continue
 		}
+
+		if f.Mode()&(os.ModeNamedPipe|os.ModeSocket|os.ModeDevice) != 0 {
+			LOG_WARN("LIST_SKIP", "Skipped non-regular file %s", entry.Path)
+			skippedFiles = append(skippedFiles, entry.Path)
+			continue
+		}
+
+		var linkKey *listEntryLinkKey
+
+		if runtime.GOOS != "windows" && !entry.IsDir() {
+			if stat := f.Sys().(*syscall.Stat_t); stat != nil && stat.Nlink > 1 {
+				k := listEntryLinkKey{dev: uint64(stat.Dev), ino: uint64(stat.Ino)}
+				if linkIndex, seen := listingState.linkTable[k]; seen {
+					if linkIndex == -1 {
+						LOG_DEBUG("LIST_EXCLUDE", "%s is excluded by attribute (hardlink)", entry.Path)
+						continue
+					}
+					entry.Size = 0
+					if entry.IsLink() {
+						entry.StartChunk = entrySymHardLinkTargetChunkMarker
+						entry.StartOffset = linkIndex
+					} else {
+						entry.Link = strconv.FormatInt(int64(linkIndex), 16)
+					}
+				} else {
+					if entry.IsLink() {
+						entry.StartChunk = entrySymHardLinkRootChunkMarker
+					} else {
+						entry.Link = "/"
+					}
+					listingState.linkTable[k] = -1
+					linkKey = &k
+				}
+			}
+		}
+
 		if entry.IsLink() {
 			isRegular := false
 			isRegular, entry.Link, err = Readlink(joinPath(top, entry.Path))
@@ -830,24 +884,6 @@ func ListEntries(top string, path string, patterns []string, nobackupFile string
 			}
 		}
 
-		var linkKey *listEntryLinkKey
-
-		if stat, ok := f.Sys().(*syscall.Stat_t); entry.IsFile() && ok && stat != nil && stat.Nlink > 1 {
-			k := listEntryLinkKey{dev: uint64(stat.Dev), ino: uint64(stat.Ino)}
-			if linkIndex, seen := listingState.linkTable[k]; seen {
-				if linkIndex == -1 {
-					LOG_DEBUG("LIST_EXCLUDE", "%s is excluded by attribute (hardlink)", entry.Path)
-					continue
-				}
-				entry.Size = 0
-				entry.Link = strconv.FormatInt(int64(linkIndex), 16)
-			} else {
-				entry.Link = "/"
-				listingState.linkTable[k] = -1
-				linkKey = &k
-			}
-		}
-
 		entry.ReadAttributes(top)
 
 		if excludeByAttribute && entry.Attributes != nil && excludedByAttribute(*entry.Attributes) {

src/duplicacy_utils.go

@@ -7,14 +7,15 @@ package duplicacy
 import (
 	"bufio"
 	"crypto/sha256"
+	"encoding/json"
 	"fmt"
 	"io"
 	"os"
 	"regexp"
+	"runtime"
 	"strconv"
 	"strings"
 	"time"
-	"runtime"
 
 	"github.com/gilbertchen/gopass"
 	"golang.org/x/crypto/pbkdf2"
@@ -275,7 +276,6 @@ func SavePassword(preference Preference, passwordType string, password string) {
 // The following code was modified from the online article  'Matching Wildcards: An Algorithm', by Kirk J. Krauss,
 // Dr. Dobb's, August 26, 2008. However, the version in the article doesn't handle cases like matching 'abcccd'
 // against '*ccd', and the version here fixed that issue.
-//
 func matchPattern(text string, pattern string) bool {
 
 	textLength := len(text)
@@ -474,3 +474,34 @@ func PrintMemoryUsage() {
 		time.Sleep(time.Second)
 	}
 }
+
+func (entry *Entry) dump() map[string]interface{} {
+
+	object := make(map[string]interface{})
+
+	object["path"] = entry.Path
+	object["size"] = entry.Size
+	object["time"] = entry.Time
+	object["mode"] = entry.Mode
+	object["hash"] = entry.Hash
+	object["link"] = entry.Link
+
+	object["content"] = fmt.Sprintf("%d:%d:%d:%d",
+		entry.StartChunk, entry.StartOffset, entry.EndChunk, entry.EndOffset)
+
+	if entry.UID != -1 && entry.GID != -1 {
+		object["uid"] = entry.UID
+		object["gid"] = entry.GID
+	}
+
+	if entry.Attributes != nil && len(*entry.Attributes) > 0 {
+		object["attributes"] = entry.Attributes
+	}
+
+	return object
+}
+
+func (entry *Entry) dumpString() string {
+	data, _ := json.Marshal(entry.dump())
+	return string(data)
+}

src/duplicacy_utils_others.go

@@ -11,6 +11,8 @@ import (
 	"os"
 	"path"
 	"syscall"
+
+	"golang.org/x/sys/unix"
 )
 
 func Readlink(path string) (isRegular bool, s string, err error) {
@@ -74,6 +76,10 @@ func (entry *Entry) RestoreSpecial(fullPath string) error {
 	return syscall.Mknod(fullPath, mode, int(uint64(entry.StartChunk)|uint64(entry.StartOffset)<<32))
 }
 
+func MakeHardlink(source string, target string) error {
+	return unix.Linkat(unix.AT_FDCWD, source, unix.AT_FDCWD, target, 0)
+}
+
 func joinPath(components ...string) string {
 	return path.Join(components...)
 }

src/duplicacy_utils_windows.go

@@ -125,6 +125,10 @@ func (entry *Entry) RestoreSpecial(fullPath string) error {
 	return nil
 }
 
+func MakeHardlink(source string, target string) error {
+	return os.Link(source, target)
+}
+
 func joinPath(components ...string) string {
 
 	combinedPath := `\\?\` + filepath.Join(components...)