Enforce correct filename normalization (fixes #96)

5 files changed, 34 insertions, 5 deletions

diff --git a/cmd/syncthing/model.go b/cmd/syncthing/model.go
index acab979cc3..db6dd29d8b 100644
--- a/cmd/syncthing/model.go
+++ b/cmd/syncthing/model.go
@@ -839,8 +839,8 @@ func (m *Model) recomputeNeedForFile(gf scanner.File, toAdd []addOrder, toDelete
 		} else {
 			local, remote := scanner.BlockDiff(lf.Blocks, gf.Blocks)
 			fm := fileMonitor{
-				name:        gf.Name,
-				path:        path.Clean(path.Join(m.dir, gf.Name)),
+				name:        FSNormalize(gf.Name),
+				path:        FSNormalize(path.Clean(path.Join(m.dir, gf.Name))),
 				global:      gf,
 				model:       m,
 				localBlocks: local,
@@ -875,7 +875,7 @@ func (m *Model) deleteLoop() {
 		if debugPull {
 			dlog.Println("delete", file.Name)
 		}
-		path := path.Clean(path.Join(m.dir, file.Name))
+		path := FSNormalize(path.Clean(path.Join(m.dir, file.Name)))
 		err := os.Remove(path)
 		if err != nil {
 			warnf("%s: %v", file.Name, err)
diff --git a/cmd/syncthing/normalize.go b/cmd/syncthing/normalize.go
new file mode 100644
index 0000000000..ccedaccd42
--- /dev/null
+++ b/cmd/syncthing/normalize.go
@@ -0,0 +1,11 @@
+//+build !darwin
+
+package main
+
+import "code.google.com/p/go.text/unicode/norm"
+
+// FSNormalize returns the string with the required unicode normalization for
+// the host operating system.
+func FSNormalize(s string) string {
+	return norm.NFC.String(s)
+}
diff --git a/cmd/syncthing/normalize_darwin.go b/cmd/syncthing/normalize_darwin.go
new file mode 100644
index 0000000000..77ba8e5cb9
--- /dev/null
+++ b/cmd/syncthing/normalize_darwin.go
@@ -0,0 +1,11 @@
+//+build darwin
+
+package main
+
+import "code.google.com/p/go.text/unicode/norm"
+
+// FSNormalize returns the string with the required unicode normalization for
+// the host operating system.
+func FSNormalize(s string) string {
+	return norm.NFD.String(s)
+}
diff --git a/protocol/PROTOCOL.md b/protocol/PROTOCOL.md
index eb1383c2e2..1a5cb300e7 100644
--- a/protocol/PROTOCOL.md
+++ b/protocol/PROTOCOL.md
@@ -163,8 +163,9 @@ response to the Index message.
 The Repository field identifies the repository that the index message
 pertains to. For single repository implementations an empty repository
 ID is acceptable, or the word "default". The Name is the file name path
-relative to the repository root. The combination of Repository and Name
-uniquely identifies each file in a cluster.
+relative to the repository root. The Name is always in UTF-8 NFC regardless
+of operating system or file system specific conventions. The combination of
+Repository and Name uniquely identifies each file in a cluster.
 
 The Version field is a counter that is initially zero for each file. It
 is incremented each time a change is detected. The combination of
diff --git a/scanner/walk.go b/scanner/walk.go
index e32cd0830b..1bd4124aa8 100644
--- a/scanner/walk.go
+++ b/scanner/walk.go
@@ -9,6 +9,8 @@ import (
 	"path/filepath"
 	"strings"
 	"time"
+
+	"code.google.com/p/go.text/unicode/norm"
 )
 
 type Walker struct {
@@ -136,6 +138,7 @@ func (w *Walker) loadIgnoreFiles(dir string, ign map[string][]string) filepath.W
 
 func (w *Walker) walkAndHashFiles(res *[]File, ign map[string][]string) filepath.WalkFunc {
 	return func(p string, info os.FileInfo, err error) error {
+
 		if err != nil {
 			if debug {
 				dlog.Println("error:", p, info, err)
@@ -151,6 +154,9 @@ func (w *Walker) walkAndHashFiles(res *[]File, ign map[string][]string) filepath
 			return nil
 		}
 
+		// Internally, we always use unicode normalization form C
+		rn = norm.NFC.String(rn)
+
 		if w.TempNamer != nil && w.TempNamer.IsTemporary(rn) {
 			if debug {
 				dlog.Println("temporary:", rn)