repo: move state/index/data_index to site_cache_dir

We use all of these for optimization and should not depend on a potentially slow or broken
filesystem on which repository resides (could be HDD or CIFS/NFS/overlayfs/etc).

This PR means a few things:
    * all odd filesystems that dvc repo might be on will work quickly out of the box
    * when we start caching cloned repositories for `dvc get/import`, we will be
      able to cache indexes for them as well (note that bare clones don't have a
      worktree so there is no place for .dvc/tmp with the old approach).
    * `index.dir` and `state.dir` are now obsolete.
    * `dvc init` will cleanup site_cache_dir.

Related #5562

dvc/config_schema.py

@@ -273,12 +273,12 @@ SCHEMA = {
         )
     },
     "state": {
-        "dir": str,
+        "dir": str,  # obsoleted
         "row_limit": All(Coerce(int), Range(1)),  # obsoleted
         "row_cleanup_quota": All(Coerce(int), Range(0, 100)),  # obsoleted
     },
     "index": {
-        "dir": str,
+        "dir": str,  # obsoleted
     },
     "machine": {
         str: {

dvc/data_cloud.py

@@ -75,7 +75,7 @@ class DataCloud:
                     config["version_aware"] = True
 
             fs = cls(**config)
-            config["tmp_dir"] = self.repo.index_db_dir
+            config["tmp_dir"] = self.repo.site_cache_dir
             if self.repo.data_index is not None:
                 index = self.repo.data_index.view(("remote", name))
             else:

dvc/info.py

@@ -54,6 +54,7 @@ def get_dvc_info():
             fs_root = get_fs_type(os.path.abspath(root_directory))
             info.append(f"Workspace directory: {fs_root}")
             info.append(f"Repo: {_get_dvc_repo_info(repo)}")
+            info.append(f"Repo.site_cache_dir: {repo.site_cache_dir}")
     except NotDvcRepoError:
         pass
     except SCMError:

dvc/repo/__init__.py

@@ -134,38 +134,6 @@ class Repo:
         assert root_dir
         return root_dir, dvc_dir
 
-    def _get_database_dir(self, db_name: str) -> Optional[str]:
-        from dvc.fs import localfs
-
-        # NOTE: by default, store SQLite-based remote indexes and state's
-        # `links` and `md5s` caches in the repository itself to avoid any
-        # possible state corruption in 'shared cache dir' scenario, but allow
-        # user to override this through config when, say, the repository is
-        # located on a mounted volume — see
-        # https://github.com/iterative/dvc/issues/4420
-        base_db_dir = self.config.get(db_name, {}).get("dir", None)
-        if not base_db_dir:
-            return self.tmp_dir
-
-        import hashlib
-
-        if self.local_dvc_dir:
-            fs: "FileSystem" = localfs
-            local_root = fs.path.parent(self.local_dvc_dir)
-        else:
-            fs = self.fs
-            local_root = self.root_dir
-        root_dir_hash = hashlib.sha224(local_root.encode("utf-8")).hexdigest()
-
-        db_dir = fs.path.join(
-            base_db_dir,
-            self.DVC_DIR,
-            f"{fs.path.name(local_root)}-{root_dir_hash[0:7]}",
-        )
-
-        fs.makedirs(db_dir, exist_ok=True)
-        return db_dir
-
     def __init__(  # noqa: PLR0915
         self,
         root_dir: Optional[str] = None,
@@ -241,8 +209,8 @@ class Repo:
                     hardlink_lock=self.config["core"].get("hardlink_lock", False),
                     friendly=True,
                 )
-                state_db_dir = self._get_database_dir("state")
+                os.makedirs(self.site_cache_dir, exist_ok=True)
-                self.state = State(self.root_dir, state_db_dir, self.dvcignore)
+                self.state = State(self.root_dir, self.site_cache_dir, self.dvcignore)
             else:
                 self.lock = LockNoop()
                 self.state = StateNoop()
@@ -401,13 +369,9 @@ class Repo:
         from dvc_data.index import DataIndex
 
         if self._data_index is None:
-            if self.index_db_dir:
+            index_dir = os.path.join(self.site_cache_dir, "index", "data")
-                index_dir = os.path.join(self.index_db_dir, "index", "data")
+            os.makedirs(index_dir, exist_ok=True)
-                os.makedirs(index_dir, exist_ok=True)
+            self._data_index = DataIndex.open(os.path.join(index_dir, "db.db"))
-
-                self._data_index = DataIndex.open(os.path.join(index_dir, "db.db"))
-            else:
-                self._data_index = DataIndex()
 
         return self._data_index
 
@@ -590,8 +554,27 @@ class Repo:
         return DVCFileSystem(repo=self, subrepos=self.subrepos, **self._fs_conf)
 
     @cached_property
-    def index_db_dir(self):
+    def site_cache_dir(self) -> str:
-        return self._get_database_dir("index")
+        import hashlib
+
+        import platformdirs
+
+        from dvc.fs import GitFileSystem
+
+        cache_dir = platformdirs.site_cache_dir("dvc", "iterative", opinion=True)
+
+        if isinstance(self.fs, GitFileSystem):
+            relparts = ()
+            if self.root_dir != "/":
+                # subrepo
+                relparts = self.fs.path.relparts(self.root_dir, "/")
+            root_dir = os.path.join(self.scm.root_dir, *relparts)
+        else:
+            root_dir = self.root_dir
+
+        repo_token = hashlib.md5(os.fsencode(root_dir)).hexdigest()  # noqa: S324
+
+        return os.path.join(cache_dir, "repo", repo_token)
 
     @contextmanager
     def open_by_relpath(self, path, remote=None, mode="r", encoding=None):

dvc/repo/init.py

@@ -12,7 +12,7 @@ from dvc.utils.fs import remove
 logger = logging.getLogger(__name__)
 
 
-def init(root_dir=os.curdir, no_scm=False, force=False, subdir=False):
+def init(root_dir=os.curdir, no_scm=False, force=False, subdir=False):  # noqa: C901
     """
     Creates an empty repo on the given directory -- basically a
     `.dvc` directory with subdirectories for configuration and cache.
@@ -75,6 +75,11 @@ def init(root_dir=os.curdir, no_scm=False, force=False, subdir=False):
 
     proj = Repo(root_dir)
 
+    if os.path.isdir(proj.site_cache_dir):
+        proj.close()
+        remove(proj.site_cache_dir)
+        proj = Repo(root_dir)
+
     with proj.scm_context(autostage=True) as context:
         files = [
             config.files["repo"],

pyproject.toml

@@ -48,7 +48,7 @@ dependencies = [
     "networkx>=2.5",
     "packaging>=19",
     "pathspec>=0.10.3",
-    "platformdirs<4,>=3",
+    "platformdirs<4,>=3.1.1",
     "psutil>=5.8",
     "pydot>=1.2.4",
     "pygtrie>=2.3.2",

tests/func/test_state.py

@@ -1,7 +1,5 @@
 import os
-import re
 
-from dvc.repo import Repo
 from dvc_data.hashfile.hash import file_md5
 from dvc_data.hashfile.hash_info import HashInfo
 from dvc_data.hashfile.state import State
@@ -71,15 +69,3 @@ def test_get_unused_links(tmp_dir, dvc):
             dvc.fs,
         )
     ) == {"bar"}
-
-
-def test_state_dir_config(make_tmp_dir, dvc):
-    assert dvc.state.tmp_dir == dvc.tmp_dir
-
-    index_dir = str(make_tmp_dir("tmp_index"))
-    repo = Repo(config={"state": {"dir": index_dir}})
-    assert os.path.dirname(repo.state.tmp_dir) == os.path.join(index_dir, ".dvc")
-    assert re.match(
-        r"^test_state_dir_config0-([0-9a-f]+)$",
-        os.path.basename(repo.state.tmp_dir),
-    )

tests/unit/remote/test_remote.py

@@ -1,11 +1,8 @@
-import os
-
 import pytest
 from dvc_gs import GSFileSystem
 from dvc_s3 import S3FileSystem
 
 from dvc.fs import get_cloud_fs
-from dvc_data.hashfile.db import get_index
 
 
 def test_remote_with_hash_jobs(dvc):
@@ -57,18 +54,3 @@ def test_makedirs_not_create_for_top_level_path(fs_cls, dvc, mocker):
 
     fs.makedirs(url)
     assert not mocked_client.called
-
-
-def test_remote_index_dir_config(make_tmp_dir, dvc):
-    index_dir = str(make_tmp_dir("tmp_index"))
-    with dvc.config.edit() as conf:
-        conf["index"]["dir"] = index_dir
-        conf["remote"]["s3"] = {"url": "s3://bucket/name"}
-
-    dvc.root_dir = "/usr/local/test_repo"
-    dvc.dvc_dir = "/usr/local/test_repo/.dvc"
-    dvc.__dict__.pop("local_dvc_dir")
-
-    assert os.path.dirname(
-        get_index(dvc.cloud.get_remote_odb(name="s3")).index_dir
-    ) == os.path.join(index_dir, ".dvc", "test_repo-a473718", "index")