Implement --exclude-if-present

Add a new --exclude-if-present command-line flag to ``borg create``.  If
specified, directories containing the specified tag file will be excluded from
the backup. The flag can be repeated to ignore more than a single tag file,
irregardless of the contents.

This is taken from a attic PR (and adapted for borg):
    commit 3462a9ca90388dc5d8b4fa4218a32769676b3623
    Author: Yuri D'Elia <yuri.delia@eurac.edu>
    Date:   Sun Dec 7 19:15:17 2014 +0100

borg/archiver.py

@@ -20,7 +20,7 @@ from .helpers import Error, location_validator, format_time, format_file_size, \
     format_file_mode, ExcludePattern, IncludePattern, exclude_path, adjust_patterns, to_localtime, timestamp, \
     get_cache_dir, get_keys_dir, prune_within, prune_split, unhexlify, \
     Manifest, remove_surrogates, update_excludes, format_archive, check_extension_modules, Statistics, \
-    is_cachedir, bigint_to_int, ChunkerParams, CompressionSpec, have_cython, is_slow_msgpack, yes, \
+    dir_is_tagged, bigint_to_int, ChunkerParams, CompressionSpec, have_cython, is_slow_msgpack, yes, \
     EXIT_SUCCESS, EXIT_WARNING, EXIT_ERROR
 from .logger import create_logger, setup_logging
 logger = create_logger()
@@ -166,8 +166,8 @@ class Archiver:
                     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,
-                          read_special=args.read_special, dry_run=dry_run)
+                          skip_inodes, path, restrict_dev, read_special=args.read_special, dry_run=dry_run)
         if not dry_run:
             archive.save(timestamp=args.timestamp)
             if args.progress:
@@ -182,8 +182,8 @@ class Archiver:
                 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,
-                 read_special=False, dry_run=False):
+                 skip_inodes, path, restrict_dev, read_special=False, dry_run=False):
         if exclude_path(path, excludes):
             return
         try:
@@ -209,7 +209,7 @@ class Archiver:
                     status = 'E'
                     self.print_warning('%s: %s', path, e)
         elif stat.S_ISDIR(st.st_mode):
-            if exclude_caches and is_cachedir(path):
+            if dir_is_tagged(path, exclude_caches, exclude_if_present):
                 return
             if not dry_run:
                 status = archive.process_dir(path, st)
@@ -221,8 +221,8 @@ class Archiver:
             else:
                 for filename in sorted(entries):
                     entry_path = os.path.normpath(os.path.join(path, filename))
-                    self._process(archive, cache, excludes, exclude_caches, skip_inodes,
+                    self._process(archive, cache, excludes, exclude_caches, exclude_if_present,
-                                  entry_path, restrict_dev, read_special=read_special,
+                                  skip_inodes, entry_path, restrict_dev, read_special=read_special,
                                   dry_run=dry_run)
         elif stat.S_ISLNK(st.st_mode):
             if not dry_run:
@@ -785,6 +785,9 @@ class Archiver:
         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('-c', '--checkpoint-interval', dest='checkpoint_interval',
                                type=int, default=300, metavar='SECONDS',
                                help='write checkpoint every SECONDS seconds (Default: 300)')

borg/helpers.py

@@ -428,7 +428,7 @@ def CompressionSpec(s):
         raise ValueError
 
 
-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
@@ -448,6 +448,20 @@ 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.
+    """
+    if exclude_caches and dir_is_cachedir(path):
+        return True
+    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 True
+    return False
+
+
 def format_time(t):
     """Format datetime suitable for fixed length list output
     """

borg/testsuite/archiver.py

@@ -499,6 +499,17 @@ class ArchiverTestCase(ArchiverTestCaseBase):
         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.cmd('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.cmd('create', '--exclude-if-present', '.NOBACKUP', '--exclude-if-present', '00-NOBACKUP', self.repository_location + '::test', 'input')
+        with changedir('output'):
+            self.cmd('extract', self.repository_location + '::test')
+        self.assert_equal(sorted(os.listdir('output/input')), ['file1', 'tagged3'])
+
     def test_path_normalization(self):
         self.cmd('init', self.repository_location)
         self.create_regular_file('dir1/dir2/file', size=1024 * 80)