PR #224 - Merge branch 'issue_55' into merge

attic/archive.py

@@ -369,21 +369,33 @@ def stat_attrs(self, st, path):
         acl_get(path, item, st, self.numeric_owner)
         return item
 
-    def process_item(self, path, st):
+    def process_dir(self, path, st):
         item = {b'path': make_path_safe(path)}
         item.update(self.stat_attrs(st, path))
         self.add_item(item)
+        return 'd'  # directory
+
+    def process_fifo(self, path, st):
+        item = {b'path': make_path_safe(path)}
+        item.update(self.stat_attrs(st, path))
+        self.add_item(item)
+        return 'f'  # fifo
 
     def process_dev(self, path, st):
         item = {b'path': make_path_safe(path), b'rdev': st.st_rdev}
         item.update(self.stat_attrs(st, path))
         self.add_item(item)
+        if stat.S_ISCHR(st.st_mode):
+            return 'c'  # char device
+        elif stat.S_ISBLK(st.st_mode):
+            return 'b'  # block device
 
     def process_symlink(self, path, st):
         source = os.readlink(path)
         item = {b'path': make_path_safe(path), b'source': source}
         item.update(self.stat_attrs(st, path))
         self.add_item(item)
+        return 's'  # symlink
 
     def process_stdin(self, path, cache):
         uid, gid = 0, 0
@@ -403,6 +415,7 @@ def process_stdin(self, path, cache):
         self.add_item(item)
 
     def process_file(self, path, st, cache):
+        status = None
         safe_path = make_path_safe(path)
         # Is it a hard link?
         if st.st_nlink > 1:
@@ -411,7 +424,8 @@ def process_file(self, path, st, cache):
                 item = self.stat_attrs(st, path)
                 item.update({b'path': safe_path, b'source': source})
                 self.add_item(item)
-                return
+                status = 'h'  # regular file, hardlink (to already seen inodes)
+                return status
             else:
                 self.hard_links[st.st_ino, st.st_dev] = safe_path
         path_hash = self.key.id_hash(os.path.join(self.cwd, path).encode('utf-8', 'surrogateescape'))
@@ -424,6 +438,9 @@ def process_file(self, path, st, cache):
                     break
             else:
                 chunks = [cache.chunk_incref(id_, self.stats) for id_ in ids]
+                status = 'U'  # regular file, unchanged
+        else:
+            status = 'A'  # regular file, added
         # Only chunkify the file if needed
         if chunks is None:
             with Archive._open_rb(path, st) as fd:
@@ -431,10 +448,12 @@ def process_file(self, path, st, cache):
                 for chunk in self.chunker.chunkify(fd):
                     chunks.append(cache.add_chunk(self.key.id_hash(chunk), chunk, self.stats))
             cache.memorize_file(path_hash, st, [c[0] for c in chunks])
+            status = status or 'M'  # regular file, modified (if not 'A' already)
         item = {b'path': safe_path, b'chunks': chunks}
         item.update(self.stat_attrs(st, path))
         self.stats.nfiles += 1
         self.add_item(item)
+        return status
 
     @staticmethod
     def list_archives(repository, key, manifest, cache=None):

attic/archiver.py

@@ -165,16 +165,16 @@ def _process(self, archive, cache, excludes, exclude_caches, skip_inodes, path,
         # Ignore unix sockets
         if stat.S_ISSOCK(st.st_mode):
             return
-        self.print_verbose(remove_surrogates(path))
+        status = None
         if stat.S_ISREG(st.st_mode):
             try:
-                archive.process_file(path, st, cache)
+                status = archive.process_file(path, st, cache)
             except IOError as e:
                 self.print_error('%s: %s', path, e)
         elif stat.S_ISDIR(st.st_mode):
             if exclude_caches and is_cachedir(path):
                 return
-            archive.process_item(path, st)
+            status = archive.process_dir(path, st)
             try:
                 entries = os.listdir(path)
             except OSError as e:
@@ -185,13 +185,27 @@ def _process(self, archive, cache, excludes, exclude_caches, skip_inodes, path,
                     self._process(archive, cache, excludes, exclude_caches, skip_inodes,
                                   entry_path, restrict_dev)
         elif stat.S_ISLNK(st.st_mode):
-            archive.process_symlink(path, st)
+            status = archive.process_symlink(path, st)
         elif stat.S_ISFIFO(st.st_mode):
-            archive.process_item(path, st)
+            status = archive.process_fifo(path, st)
         elif stat.S_ISCHR(st.st_mode) or stat.S_ISBLK(st.st_mode):
-            archive.process_dev(path, st)
+            status = archive.process_dev(path, st)
         else:
             self.print_error('Unknown file type: %s', path)
+            return
+        # Status output
+        # A lowercase character means a file type other than a regular file,
+        # attic usually just stores them. E.g. (d)irectory.
+        # Hardlinks to already seen content are indicated by (h).
+        # A uppercase character means a regular file that was (A)dded,
+        # (M)odified or was (U)nchanged.
+        # Note: A/M/U is relative to the "files" cache, not to the repo.
+        # This would be an issue if the files cache is not used.
+        if status is None:
+            status = '?'  # need to add a status code somewhere
+        # output ALL the stuff - it can be easily filtered using grep.
+        # even stuff considered unchanged might be interesting.
+        self.print_verbose("%1s %s", status, remove_surrogates(path))
 
     def do_extract(self, args):
         """Extract archive contents"""