Implement --exclude-if-present

attic/archiver.py

@@ -17,8 +17,8 @@
 from attic.helpers import Error, location_validator, format_time, \
     format_file_mode, ExcludePattern, exclude_path, adjust_patterns, to_localtime, \
     get_cache_dir, get_keys_dir, format_timedelta, prune_within, prune_split, \
-    Manifest, remove_surrogates, update_excludes, format_archive, check_extension_modules, Statistics, \
-    is_cachedir, bigint_to_int
+    Manifest, remove_surrogates, update_excludes, format_archive, check_extension_modules, \
+    Statistics, dir_is_tagged, bigint_to_int
 from attic.remote import RepositoryServer, RemoteRepository
 
 
@@ -125,7 +125,8 @@ def do_create(self, args):
                     continue
             else:
                 restrict_dev = None
-            self._process(archive, cache, args.excludes, args.exclude_caches, skip_inodes, path, restrict_dev)
+            self._process(archive, cache, args.excludes, args.exclude_caches, args.exclude_if_present,
+                          args.keep_tag_files, skip_inodes, path, restrict_dev)
         archive.save()
         if args.stats:
             t = datetime.now()
@@ -141,7 +142,8 @@ def do_create(self, args):
             print('-' * 78)
         return self.exit_code
 
-    def _process(self, archive, cache, excludes, exclude_caches, skip_inodes, path, restrict_dev):
+    def _process(self, archive, cache, excludes, exclude_caches, exclude_if_present,
+                 keep_tag_files, skip_inodes, path, restrict_dev):
         if exclude_path(path, excludes):
             return
         try:
@@ -164,7 +166,11 @@ def _process(self, archive, cache, excludes, exclude_caches, skip_inodes, path,
             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):
+            tag_path = dir_is_tagged(path, exclude_caches, exclude_if_present)
+            if tag_path:
+                if keep_tag_files:
+                    archive.process_item(path, st)
+                    archive.process_item(tag_path, st)
                 return
             archive.process_item(path, st)
             try:
@@ -173,8 +179,8 @@ def _process(self, archive, cache, excludes, exclude_caches, skip_inodes, path,
                 self.print_error('%s: %s', path, e)
             else:
                 for filename in sorted(entries):
-                    self._process(archive, cache, excludes, exclude_caches, skip_inodes,
-                                  os.path.join(path, filename), restrict_dev)
+                    self._process(archive, cache, excludes, exclude_caches, exclude_if_present,
+                                  keep_tag_files, skip_inodes, os.path.join(path, filename), restrict_dev)
         elif stat.S_ISLNK(st.st_mode):
             archive.process_symlink(path, st)
         elif stat.S_ISFIFO(st.st_mode):
@@ -541,6 +547,12 @@ def run(self, args=None):
         subparser.add_argument('--exclude-caches', dest='exclude_caches',
                                action='store_true', default=False,
                                help='exclude directories that contain a CACHEDIR.TAG file (http://www.brynosaurus.com/cachedir/spec.html)')
+        subparser.add_argument('--exclude-if-present', dest='exclude_if_present',
+                               metavar='FILENAME', action='append', type=str,
+                               help='exclude directories that contain the specified file')
+        subparser.add_argument('--keep-tag-files', dest='keep_tag_files',
+                               action='store_true', default=False,
+                               help='keep tag files of excluded caches/directories')
         subparser.add_argument('-c', '--checkpoint-interval', dest='checkpoint_interval',
                                type=int, default=300, metavar='SECONDS',
                                help='write checkpoint every SECONDS seconds (Default: 300)')

attic/helpers.py

@@ -257,7 +257,7 @@ def __repr__(self):
         return '%s(%s)' % (type(self), self.pattern)
 
 
-def is_cachedir(path):
+def dir_is_cachedir(path):
     """Determines whether the specified path is a cache directory (and
     therefore should potentially be excluded from the backup) according to
     the CACHEDIR.TAG protocol
@@ -277,6 +277,22 @@ def is_cachedir(path):
     return False
 
 
+def dir_is_tagged(path, exclude_caches, exclude_if_present):
+    """Determines whether the specified path is excluded by being a cache
+    directory or containing the user-specified tag file. Returns the
+    path of the tag file (either CACHEDIR.TAG or the matching
+    user-specified file)
+    """
+    if exclude_caches and dir_is_cachedir(path):
+        return os.path.join(path, 'CACHEDIR.TAG')
+    if exclude_if_present is not None:
+        for tag in exclude_if_present:
+            tag_path = os.path.join(path, tag)
+            if os.path.isfile(tag_path):
+                return tag_path
+    return None
+
+
 def format_time(t):
     """Format datetime suitable for fixed length list output
     """

attic/testsuite/archiver.py

@@ -208,6 +208,31 @@ def test_exclude_caches(self):
         self.assert_equal(sorted(os.listdir('output/input')), ['cache2', 'file1'])
         self.assert_equal(sorted(os.listdir('output/input/cache2')), ['CACHEDIR.TAG'])
 
+    def test_exclude_tagged(self):
+        self.attic('init', self.repository_location)
+        self.create_regular_file('file1', size=1024 * 80)
+        self.create_regular_file('tagged1/.NOBACKUP')
+        self.create_regular_file('tagged2/00-NOBACKUP')
+        self.create_regular_file('tagged3/.NOBACKUP/file2')
+        self.attic('create', '--exclude-if-present', '.NOBACKUP', '--exclude-if-present', '00-NOBACKUP', self.repository_location + '::test', 'input')
+        with changedir('output'):
+            self.attic('extract', self.repository_location + '::test')
+        self.assert_equal(sorted(os.listdir('output/input')), ['file1', 'tagged3'])
+
+    def test_exclude_keep_tagged(self):
+        self.attic('init', self.repository_location)
+        self.create_regular_file('file1', size=1024 * 80)
+        self.create_regular_file('tagged1/.NOBACKUP')
+        self.create_regular_file('tagged1/file2', size=1024 * 80)
+        self.create_regular_file('tagged2/CACHEDIR.TAG', contents = b'Signature: 8a477f597d28d172789f06886806bc55 extra stuff')
+        self.create_regular_file('tagged2/file3', size=1024 * 80)
+        self.attic('create', '--exclude-if-present', '.NOBACKUP', '--exclude-caches', '--keep-tag-files', self.repository_location + '::test', 'input')
+        with changedir('output'):
+            self.attic('extract', self.repository_location + '::test')
+        self.assert_equal(sorted(os.listdir('output/input')), ['file1', 'tagged1', 'tagged2'])
+        self.assert_equal(sorted(os.listdir('output/input/tagged1')), ['.NOBACKUP'])
+        self.assert_equal(sorted(os.listdir('output/input/tagged2')), ['CACHEDIR.TAG'])
+
     def test_path_normalization(self):
         self.attic('init', self.repository_location)
         self.create_regular_file('dir1/dir2/file', size=1024 * 80)