dvc: update gc to remove unpacked dir

In local remote its found that gc does not remove the unpacked dir. This change set helps to remove it. Signed-off-by: Sujith H <[email protected]>
iterative · Jan 6, 2020 · 1e3abe2 · 1e3abe2
1 parent 8041d70
commit 1e3abe2
Show file tree

Hide file tree

Showing 3 changed files with 25 additions and 0 deletions.
diff --git a/dvc/remote/base.py b/dvc/remote/base.py
@@ -697,6 +697,8 @@ def gc(self, named_cache):
             if checksum in used:
                 continue
             path_info = self.checksum_to_path_info(checksum)
+            if self.is_dir_checksum(checksum):
+                self._remove_unpacked_dir(checksum)
             self.remove(path_info)
             removed = True
         return removed
@@ -1009,3 +1011,6 @@ def _changed_unpacked_dir(self, checksum):
 
     def _update_unpacked_dir(self, checksum):
         pass
+
+    def _remove_unpacked_dir(self, checksum):
+        pass
diff --git a/dvc/remote/local.py b/dvc/remote/local.py
@@ -464,6 +464,10 @@ def _get_unpacked_dir_path_info(self, checksum):
         info = self.checksum_to_path_info(checksum)
         return info.with_name(info.name + self.UNPACKED_DIR_SUFFIX)
 
+    def _remove_unpacked_dir(self, checksum):
+        path_info = self._get_unpacked_dir_path_info(checksum)
+        self.remove(path_info)
+
     def _path_info_changed(self, path_info):
         if self.exists(path_info) and self.state.get(path_info):
             return False

diff --git a/tests/func/test_gc.py b/tests/func/test_gc.py
@@ -7,6 +7,7 @@
 from dvc.exceptions import CollectCacheError
 from dvc.main import main
 from dvc.repo import Repo as DvcRepo
+from dvc.remote.local import RemoteLOCAL
 from tests.basic_env import TestDir, TestDvcGit
 
 
@@ -204,3 +205,18 @@ def test_gc_no_dir_cache(tmp_dir, dvc, repo_template):
 
 def _count_files(path):
     return sum(len(files) for _, _, files in os.walk(path))
+
+
+def test_gc_no_unpacked_dir(tmp_dir, dvc, repo_template):
+    dir_stages = dvc.add("dir")
+    dvc.status()
+
+    os.remove("dir.dvc")
+    unpackeddir = (
+        dir_stages[0].outs[0].cache_path + RemoteLOCAL.UNPACKED_DIR_SUFFIX
+    )
+
+    assert os.path.exists(unpackeddir)
+
+    dvc.gc(force=True)
+    assert not os.path.exists(unpackeddir)