deps: bump dvc-data to 0.31.3

dvc/dependency/param.py

@@ -2,7 +2,7 @@
 import os
 import typing
 from collections import defaultdict
-from typing import Dict
+from typing import Dict, cast
 
 import dpath.util
 from voluptuous import Any
@@ -41,7 +41,10 @@ def __init__(self, stage, path, params=None, repo=None):
         self.params = list(params) if params else []
         hash_info = HashInfo()
         if isinstance(params, dict):
-            hash_info = HashInfo(self.PARAM_PARAMS, params)
+            hash_info = HashInfo(
+                self.PARAM_PARAMS,
+                params,  # type: ignore[arg-type]
+            )
         repo = repo or stage.repo
         path = path or os.path.join(repo.root_dir, self.DEFAULT_PARAMS_FILE)
         super().__init__(stage, path, repo=repo)
@@ -64,7 +67,10 @@ def fill_values(self, values=None):
         for param in self.params:
             if param in values:
                 info[param] = values[param]
-        self.hash_info = HashInfo(self.PARAM_PARAMS, info)
+        self.hash_info = HashInfo(
+            self.PARAM_PARAMS,
+            info,  # type: ignore[arg-type]
+        )
 
     def read_params(
         self, flatten: bool = True, **kwargs: typing.Any
@@ -105,7 +111,7 @@ def workspace_status(self):
         from funcy import ldistinct
 
         status: Dict[str, Any] = defaultdict(dict)
-        info = self.hash_info.value if self.hash_info else {}
+        info = cast(dict, self.hash_info.value) if self.hash_info else {}
         actual = self.read_params()
 
         # NOTE: we want to preserve the order of params as specified in the
@@ -159,7 +165,7 @@ def get_hash(self):
                 )
             )
 
-        return HashInfo(self.PARAM_PARAMS, info)
+        return HashInfo(self.PARAM_PARAMS, info)  # type: ignore[arg-type]
 
     def save(self):
         if not self.exists:

dvc/dependency/repo.py

@@ -184,6 +184,7 @@ def iter_objs():
                     tree = Tree.load(odb, hash_info)
                     yield from (odb.get(hi.value) for _, _, hi in tree)
                 else:
+                    assert hash_info.value
                     yield odb.get(hash_info.value)
 
         checked_urls = set()

dvc/output.py

@@ -1,7 +1,7 @@
 import logging
 import os
 from collections import defaultdict
-from typing import TYPE_CHECKING, Dict, List, Optional, Set, Tuple, Type
+from typing import TYPE_CHECKING, Dict, List, Optional, Set, Tuple, Type, cast
 from urllib.parse import urlparse
 
 from funcy import collecting, project
@@ -732,7 +732,8 @@ def _commit_granular_dir(self, filter_info) -> Optional["HashFile"]:
             self.hash_name,
             ignore=self.dvcignore,
         )
-        save_obj = save_obj.filter(prefix)
+        save_obj = cast(Tree, save_obj)
+        save_obj = cast(Tree, save_obj.filter(prefix))
         checkout_obj = save_obj.get_obj(self.odb, prefix)
         otransfer(
             staging,
@@ -746,7 +747,7 @@ def _commit_granular_dir(self, filter_info) -> Optional["HashFile"]:
     def dumpd(self, **kwargs):  # noqa: C901
         meta = self.meta.to_dict()
         meta.pop("isdir", None)
-        ret = {**self.hash_info.to_dict(), **meta}
+        ret: Dict[str, Any] = {**self.hash_info.to_dict(), **meta}
 
         if self.is_in_repo:
             path = self.fs.path.as_posix(
@@ -792,11 +793,14 @@ def dumpd(self, **kwargs):  # noqa: C901
             and self.hash_info.isdir
             and (kwargs.get("with_files") or self.files is not None)
         ):
+            obj: Optional["HashFile"]
             if self.obj:
                 obj = self.obj
             else:
                 obj = self.get_obj()
-            ret[self.PARAM_FILES] = obj.as_list(with_meta=True)
+            if obj:
+                obj = cast(Tree, obj)
+                ret[self.PARAM_FILES] = obj.as_list(with_meta=True)
 
         return ret
 
@@ -844,6 +848,7 @@ def get_obj(
         fs_path = self.fs.path
         if filter_info and filter_info != self.fs_path:
             prefix = fs_path.relparts(filter_info, self.fs_path)
+            obj = cast(Tree, obj)
             obj = obj.get_obj(self.odb, prefix)
 
         return obj
@@ -1032,8 +1037,9 @@ def _collect_used_dir_cache(
             prefix = self.fs.path.parts(
                 self.fs.path.relpath(filter_info, self.fs_path)
             )
+            obj = cast(Tree, obj)
             return obj.filter(prefix)
-        return obj
+        return cast(Tree, obj)
 
     def get_used_objs(  # noqa: C901
         self, **kwargs
@@ -1071,6 +1077,7 @@ def get_used_objs(  # noqa: C901
             logger.warning(msg)
             return {}
 
+        obj: Optional["HashFile"]
         if self.is_dir_checksum:
             obj = self._collect_used_dir_cache(**kwargs)
         else:

dvc/repo/__init__.py

@@ -11,6 +11,7 @@
     Optional,
     Tuple,
     Union,
+    cast,
 )
 
 from dvc.exceptions import FileMissingError
@@ -22,14 +23,15 @@
 from dvc.utils.objects import cached_property
 
 if TYPE_CHECKING:
-    from dvc.fs import FileSystem
+    from dvc.fs import FileSystem, GitFileSystem
     from dvc.fs.data import DataFileSystem
     from dvc.fs.dvc import DVCFileSystem
     from dvc.lock import LockBase
     from dvc.machine import MachineManager
     from dvc.scm import Base, Git, NoSCM
     from dvc.stage import Stage
     from dvc.types import DictStrAny
+    from dvc_data.hashfile.state import StateBase
 
     from .experiments import Experiments
     from .index import Index
@@ -215,13 +217,14 @@ def __init__(
 
         self.lock: "LockBase"
         self.odb: ODBManager
+        self.state: "StateBase"
         if isinstance(self.fs, GitFileSystem) or not self.dvc_dir:
             self.lock = LockNoop()
             self.state = StateNoop()
             self.odb = ODBManager(self)
             self.tmp_dir = None
         else:
-            self.fs.makedirs(self.tmp_dir, exist_ok=True)
+            self.fs.makedirs(cast(str, self.tmp_dir), exist_ok=True)
 
             if isinstance(self.fs, LocalFileSystem):
                 self.lock = make_lock(
@@ -322,7 +325,7 @@ def get_rev(self):
 
             with map_scm_exception():
                 return self.scm.get_rev()
-        return self.fs.rev
+        return cast("GitFileSystem", self.fs).rev
 
     @cached_property
     def experiments(self) -> "Experiments":
@@ -541,6 +544,7 @@ def open_by_relpath(self, path, remote=None, mode="r", encoding=None):
         from dvc.fs.data import DataFileSystem
         from dvc.fs.dvc import DVCFileSystem
 
+        fs: Union["FileSystem", DataFileSystem, DVCFileSystem]
         if os.path.isabs(path):
             fs = DataFileSystem(index=self.index.data["local"])
             fs_path = path

dvc/repo/data.py

@@ -214,7 +214,7 @@ def _diff_head_to_index(
     repo: "Repo", head: str = "HEAD", **kwargs: Any
 ) -> Dict[str, List[str]]:
     # we need to store objects from index and the HEAD to diff later
-    objs: Dict[str, Dict[str, Tuple["HashFile", "HashInfo"]]]
+    objs: Dict[str, Dict[str, Tuple[Optional["HashFile"], "HashInfo"]]]
     objs = defaultdict(dict)
 
     staged_diff = defaultdict(list)

dvc/repo/diff.py

@@ -11,6 +11,7 @@
     Optional,
     Set,
     Tuple,
+    cast,
 )
 
 from dvc.exceptions import PathMissingError
@@ -21,7 +22,8 @@
     from dvc.fs import DVCFileSystem, FileSystem
     from dvc.output import Output
     from dvc.repo import Repo
-    from dvc_objects.obj import Object
+    from dvc_data.hashfile.obj import HashFile
+    from dvc_data.hashfile.tree import Tree
 
 logger = logging.getLogger(__name__)
 
@@ -164,6 +166,7 @@ def _to_path(output: "Output") -> str:
             for target in targets
         )
 
+        obj: Optional["HashFile"]
         if on_working_fs:
             _, _, obj = build(
                 repo.odb.repo,
@@ -194,6 +197,7 @@ def _to_path(output: "Output") -> str:
                 )
             )
         ):
+            obj = cast("Tree", obj)
             yield from _dir_output_paths(
                 output.fs, output.fs_path, obj, targets
             )
@@ -202,7 +206,7 @@ def _to_path(output: "Output") -> str:
 def _dir_output_paths(
     fs: "FileSystem",
     fs_path: str,
-    obj: "Object",
+    obj: "Tree",
     targets: Optional[Iterable[str]] = None,
 ) -> Iterator[Tuple[str, str]]:
     base = os.path.join(*fs.path.relparts(fs_path))

dvc/repo/fetch.py

@@ -12,8 +12,8 @@
     from dvc.data_cloud import Remote
     from dvc.repo import Repo
     from dvc.types import TargetType
+    from dvc_data.hashfile.db import HashFileDB
     from dvc_data.hashfile.transfer import TransferResult
-    from dvc_objects.db import ObjectDB
 
 logger = logging.getLogger(__name__)
 
@@ -31,7 +31,7 @@ def fetch(  # noqa: C901
     all_commits=False,
     run_cache=False,
     revs=None,
-    odb: Optional["ObjectDB"] = None,
+    odb: Optional["HashFileDB"] = None,
 ) -> int:
     """Download data items from a cloud and imported repositories
 
@@ -133,7 +133,7 @@ def _fetch(
     targets: "TargetType",
     remote: Optional[str] = None,
     jobs: Optional[int] = None,
-    odb: Optional["ObjectDB"] = None,
+    odb: Optional["HashFileDB"] = None,
     **kwargs,
 ) -> "TransferResult":
     from dvc_data.hashfile.transfer import TransferResult

dvc/repo/get.py

@@ -1,10 +1,15 @@
 import logging
 import os
+from typing import TYPE_CHECKING, Union
 
 from dvc.exceptions import DvcException
 from dvc.utils import resolve_output
 from dvc.utils.fs import remove
 
+if TYPE_CHECKING:
+    from dvc.fs.dvc import DVCFileSystem
+
+
 logger = logging.getLogger(__name__)
 
 
@@ -50,9 +55,10 @@ def get(url, path, out=None, rev=None, jobs=None):
         with external_repo(
             url=url, rev=rev, cache_dir=tmp_dir, cache_types=cache_types
         ) as repo:
+            from dvc.fs.data import DataFileSystem
 
+            fs: Union[DataFileSystem, "DVCFileSystem"]
             if os.path.isabs(path):
-                from dvc.fs.data import DataFileSystem
 
                 fs = DataFileSystem(index=repo.index.data["local"])
                 fs_path = fs.from_os_path(path)

dvc/repo/imports.py

@@ -1,7 +1,7 @@
 import logging
 import os
 from tempfile import TemporaryDirectory
-from typing import TYPE_CHECKING, List, Set, Tuple, Union
+from typing import TYPE_CHECKING, List, Set, Tuple, Union, cast
 
 if TYPE_CHECKING:
     from dvc.dependency.base import Dependency
@@ -77,7 +77,7 @@ def unpartial_imports(index: Union["Index", "IndexView"]) -> int:
                 out.hash_info = dep.get_obj().hash_info
                 out.meta = dep.get_meta()
             else:
-                out.hash_info = entry.hash_info
+                out.hash_info = cast("HashInfo", entry.hash_info)
                 out.meta = entry.meta
             out.stage.dump()
             updated += out.meta.nfiles if out.meta.nfiles is not None else 1
@@ -123,7 +123,9 @@ def save_imports(
         downloaded.update(
             entry.hash_info
             for _, entry in data_view.iteritems()
-            if not entry.meta.isdir and entry.hash_info is not None
+            if entry.meta is not None
+            and not entry.meta.isdir
+            and entry.hash_info is not None
         )
 
     if unpartial:

dvc/repo/worktree.py

@@ -25,7 +25,7 @@
 def _meta_checksum(fs: "FileSystem", meta: "Meta") -> Any:
     if not meta or meta.isdir:
         return meta
-    return getattr(meta, fs.PARAM_CHECKSUM)
+    return getattr(meta, cast(str, fs.PARAM_CHECKSUM))
 
 
 def worktree_view(
@@ -161,14 +161,14 @@ def _update_out_meta(
         entry.hash_info = old_tree.hash_info
         entry.meta = out.meta
         for subkey, entry in index.iteritems(key):
-            if entry.meta.isdir:
+            if entry.meta is not None and entry.meta.isdir:
                 continue
             fs_path = repo.fs.path.join(repo.root_dir, *subkey)
             meta, hash_info = old_tree.get(
                 repo.fs.path.relparts(fs_path, out.fs_path)
-            )
+            ) or (None, None)
             entry.hash_info = hash_info
-            if meta.version_id is not None:
+            if meta is not None and meta.version_id is not None:
                 # preserve existing version IDs for unchanged files in
                 # this dir (entry will have the latest remote version
                 # ID after checkout)
@@ -236,8 +236,8 @@ def outs_filter(out: "Output") -> bool:
 
 def _fetch_out_changes(
     out: "Output",
-    local_index: "DataIndex",
-    remote_index: "DataIndex",
+    local_index: Union["DataIndex", "DataIndexView"],
+    remote_index: Union["DataIndex", "DataIndexView"],
     remote: "Remote",
 ):
     from dvc_data.index import DataIndex, checkout

dvc/utils/serialize/_common.py

@@ -11,6 +11,7 @@
     Protocol,
     TextIO,
     Union,
+    cast,
 )
 
 from funcy import reraise
@@ -76,7 +77,7 @@ def _load_data(
 ):
     open_fn = fs.open if fs else open
     encoding = "utf-8"
-    with open_fn(path, encoding=encoding) as fd:
+    with open_fn(path, encoding=encoding) as fd:  # type: ignore[operator]
         with reraise(UnicodeDecodeError, EncodingError(path, encoding)):
             return parser(fd.read(), path)
 
@@ -89,7 +90,7 @@ def _dump_data(
     **dumper_args,
 ):
     open_fn = fs.open if fs else open
-    with open_fn(path, "w+", encoding="utf-8") as fd:
+    with open_fn(path, "w+", encoding="utf-8") as fd:  # type: ignore[operator]
         dumper(data, fd, **dumper_args)
 
 
@@ -101,7 +102,7 @@ def _modify_data(
     fs: Optional["FileSystem"] = None,
 ):
     exists_fn = fs.exists if fs else os.path.exists
-    file_exists = exists_fn(path)
+    file_exists = exists_fn(cast(str, path))
     data = _load_data(path, parser=parser, fs=fs) if file_exists else {}
     yield data
     _dump_data(path, data, dumper=dumper, fs=fs)