fix(data status): handle missing DVC repo at Git HEAD (#10862)

Avoid raising errors in `data_status()` when Git HEAD does not contain a DVC
repository. Instead, fall back to an empty index.

Tests now cover both unborn and committed Git repos, with and without subdirs.

.github/workflows/benchmarks.yaml

@@ -16,14 +16,14 @@ jobs:
     steps:
       - uses: actions/setup-python@v5
         with:
-            python-version: "3.12"
+            python-version: "3.13"
 
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           ref: ${{ github.event.pull_request.base.sha }}
           fetch-depth: 0
 
-      - uses: astral-sh/setup-uv@v4
+      - uses: astral-sh/setup-uv@v6
         with:
           enable-cache: true
           cache-dependency-glob: pyproject.toml
@@ -32,7 +32,7 @@ jobs:
       - name: run benchmarks on base branch
         run: pytest --benchmark-autosave dvc/testing/benchmarks/ -k 'test_init or test_help'
 
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           fetch-depth: 0
           clean: false

.github/workflows/build.yaml

@@ -19,31 +19,33 @@ jobs:
     name: Build & Verify package
     runs-on: ubuntu-latest
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v5
       with:
         fetch-depth: 0
 
     - uses: actions/setup-python@v5
       with:
-        python-version: 3.12
+        python-version: 3.13
 
-    - uses: astral-sh/setup-uv@v4
-    - run: uv pip install --upgrade setuptools_scm twine --system
+    - uses: astral-sh/setup-uv@v6
 
-    - name: Force version for Test PyPI upload
-      if: ${{ !startsWith(github.ref, 'refs/tags') }}
-      run: echo version=$(python -m setuptools_scm | awk -F+ '{print $1}' | tail -1) >> $GITHUB_ENV
+    - name: Omit local version for Test PyPI upload
+      if: github.event_name == 'push' && github.ref == 'refs/heads/main'
+      run: echo SETUPTOOLS_SCM_OVERRIDES_FOR_DVC='{local_scheme="no-local-version"}' >> $GITHUB_ENV
 
-    - run: echo 'PKG = "pip"'>dvc/_build.py
-    - run: uv build
-      env:
-        SETUPTOOLS_SCM_PRETEND_VERSION_FOR_DVC: ${{ env.version }}
-    - run: twine check --strict dist/*
+    - name: Build Python Package
+      run: |
+        echo 'PKG = "pip"'>dvc/_build.py
+        uv build
+
+    - name: Check dist
+      run: uv tool run twine check --strict dist/*
 
     - uses: actions/upload-artifact@v4
       with:
         name: Packages
         path: dist/
+        if-no-files-found: error
 
   test-pypi-publish:
     name: Publish dev package to test.pypi.org
@@ -53,13 +55,13 @@ jobs:
 
     environment:
       name: test-pypi
-      url: https://test.pypi.org/p/dvc
+      url: https://test.pypi.org/p/dvc/${{ github.event.release.tag_name }}
 
     permissions:
       id-token: write
 
     steps:
-    - uses: actions/download-artifact@v4
+    - uses: actions/download-artifact@v5
       with:
         name: Packages
         path: dist
@@ -78,13 +80,13 @@ jobs:
 
     environment:
       name: pypi
-      url: https://pypi.org/p/dvc
+      url: https://pypi.org/p/dvc/${{ github.event.release.tag_name }}
 
     permissions:
       id-token: write
 
     steps:
-    - uses: actions/download-artifact@v4
+    - uses: actions/download-artifact@v5
       with:
         name: Packages
         path: dist

.github/workflows/codeql.yml

@@ -28,7 +28,7 @@ jobs:
 
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v4
+      uses: actions/checkout@v5
 
     - name: Initialize CodeQL
       uses: github/codeql-action/init@v3

.github/workflows/plugin_tests.yaml

@@ -19,15 +19,15 @@ jobs:
       fail-fast: false
       matrix:
         os: [ubuntu-latest]
-        pyv: ["3.12"]
+        pyv: ["3.13"]
         plugin: ["dvc-s3"]
 
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v5
       with:
         path: dvc
 
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v5
       with:
         repository: iterative/${{ matrix.plugin }}
         ref: main
@@ -38,7 +38,7 @@ jobs:
       with:
         python-version: ${{ matrix.pyv }}
 
-    - uses: astral-sh/setup-uv@v4
+    - uses: astral-sh/setup-uv@v6
       with:
         enable-cache: true
         cache-dependency-glob: "**/pyproject.toml"

.github/workflows/tests.yaml

@@ -26,14 +26,14 @@ jobs:
       matrix:
         os: [ubuntu-latest, macos-latest, windows-latest]
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v5
       with:
         fetch-depth: 0
     - name: Set up Python
       uses: actions/setup-python@v5
       with:
-        python-version: '3.12'
-    - uses: astral-sh/setup-uv@v4
+        python-version: '3.13'
+    - uses: astral-sh/setup-uv@v6
       with:
         enable-cache: true
         cache-dependency-glob: pyproject.toml
@@ -55,7 +55,7 @@ jobs:
       fail-fast: false
       matrix:
         os: [ubuntu-latest, macos-latest]
-        pyv: ["3.9", "3.10", "3.11", "3.12"]
+        pyv: ["3.9", "3.10", "3.11", "3.12", "3.13"]
         include:
         - os: windows-latest
           pyv: "3.9"
@@ -69,6 +69,30 @@ jobs:
         - os: windows-latest
           pyv: "3.9"
           pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 4"
+        - os: windows-latest
+          pyv: "3.10"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 1"
+        - os: windows-latest
+          pyv: "3.10"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 2"
+        - os: windows-latest
+          pyv: "3.10"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 3"
+        - os: windows-latest
+          pyv: "3.10"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 4"
+        - os: windows-latest
+          pyv: "3.11"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 1"
+        - os: windows-latest
+          pyv: "3.11"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 2"
+        - os: windows-latest
+          pyv: "3.11"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 3"
+        - os: windows-latest
+          pyv: "3.11"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 4"
         - os: windows-latest
           pyv: "3.12"
           pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 1"
@@ -81,15 +105,27 @@ jobs:
         - os: windows-latest
           pyv: "3.12"
           pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 4"
+        - os: windows-latest
+          pyv: "3.13"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 1"
+        - os: windows-latest
+          pyv: "3.13"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 2"
+        - os: windows-latest
+          pyv: "3.13"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 3"
+        - os: windows-latest
+          pyv: "3.13"
+          pytestargs: "--splitting-algorithm=least_duration --splits 4 --group 4"
     steps:
-    - uses: actions/checkout@v4
+    - uses: actions/checkout@v5
       with:
         fetch-depth: 0
     - name: Set up Python
       uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.pyv }}
-    - uses: astral-sh/setup-uv@v4
+    - uses: astral-sh/setup-uv@v6
       with:
         enable-cache: true
         cache-dependency-glob: pyproject.toml
@@ -101,15 +137,13 @@ jobs:
     - name: run tests
       timeout-minutes: 40
       env:
+        COVERAGE_CORE: sysmon
         PYTHONUTF8: 1
       run: >
         pytest ${{ matrix.pytestargs }} -n=logical --dist=worksteal --timeout=300 --durations=0
         --cov --cov-report=xml --cov-report=term  --durations-path=./.github/.test_durations
     - name: upload coverage report
       uses: codecov/codecov-action@v5
-      with:
-        file: ./coverage.xml
-        fail_ci_if_error: false
   check:
     if: always()
     needs: [tests]
@@ -124,7 +158,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
     - name: Slack Notification
-      uses: rtCamp/[email protected].2
+      uses: rtCamp/[email protected].3
       env:
         SLACK_WEBHOOK: ${{ secrets.SLACK_WEBHOOK }}
         SLACK_COLOR: ${{ job.status }}

.pre-commit-config.yaml

@@ -6,7 +6,7 @@ ci:
 
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v5.0.0
+    rev: v6.0.0
     hooks:
       - id: check-added-large-files
       - id: check-case-conflict
@@ -22,13 +22,13 @@ repos:
       - id: sort-simple-yaml
       - id: trailing-whitespace
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: 'v0.8.0'
+    rev: 'v0.12.11'
     hooks:
-      - id: ruff
+      - id: ruff-check
         args: [--fix, --exit-non-zero-on-fix]
       - id: ruff-format
   - repo: https://github.com/codespell-project/codespell
-    rev: v2.3.0
+    rev: v2.4.1
     hooks:
       - id: codespell
         additional_dependencies: ["tomli"]

README.rst

@@ -1,6 +1,4 @@
-🚀 Check out our new product `DataChain <https://github.com/iterative/datachain>`_ (and give it a ⭐!) if you need to version and process a large number of files. Contact us at [email protected] to discuss commercial solutions and support for AI reproducibility and data management scenarios.
-
---------------------------
+|Banner|
 
 `Website <https://dvc.org>`_
 • `Docs <https://dvc.org/doc>`_
@@ -215,7 +213,7 @@ Community and Support
 * `Forum <https://discuss.dvc.org/>`_
 * `Discord Chat <https://dvc.org/chat>`_
 * `Email <mailto:[email protected]>`_
-* `Mailing List <https://sweedom.us10.list-manage.com/subscribe/post?u=a08bf93caae4063c4e6a351f6&id=24c0ecc49a>`_
+* `Mailing List <https://dvc.org/community#subscribe>`_
 
 Copyright
 =========
@@ -242,11 +240,11 @@ Barrak, A., Eghan, E.E. and Adams, B. `On the Co-evolution of ML Pipelines and S
 .. |VS Code Extension Overview| image:: https://raw.githubusercontent.com/iterative/vscode-dvc/main/extension/docs/overview.gif
    :alt: DVC Extension for VS Code
 
-.. |CI| image:: https://github.com/iterative/dvc/workflows/Tests/badge.svg?branch=main
-   :target: https://github.com/iterative/dvc/actions
+.. |CI| image:: https://github.com/iterative/dvc/actions/workflows/tests.yaml/badge.svg
+   :target: https://github.com/iterative/dvc/actions/workflows/tests.yaml
    :alt: GHA Tests
 
-.. |Maintainability| image:: https://codeclimate.com/github/iterative/dvc/badges/gpa.svg
+.. |Maintainability| image:: https://img.shields.io/codeclimate/maintainability/iterative/dvc
    :target: https://codeclimate.com/github/iterative/dvc
    :alt: Code Climate
 
@@ -270,7 +268,7 @@ Barrak, A., Eghan, E.E. and Adams, B. `On the Co-evolution of ML Pipelines and S
    :target: https://formulae.brew.sh/formula/dvc
    :alt: Homebrew
 
-.. |Conda| image:: https://img.shields.io/conda/v/conda-forge/dvc.svg?label=conda&logo=conda-forge
+.. |Conda| image:: https://anaconda.org/conda-forge/dvc/badges/version.svg
    :target: https://anaconda.org/conda-forge/dvc
    :alt: Conda-forge
 

dvc/_debug.py

@@ -14,7 +14,7 @@ def viztracer_profile(
     log_async: bool = False,
 ):
     try:
-        import viztracer
+        import viztracer  # ty: ignore[unresolved-import]
     except ImportError:
         print("Failed to run profiler, viztracer is not installed")  # noqa: T201
         yield
@@ -37,7 +37,7 @@ def yappi_profile(
     separate_threads: Optional[bool] = False,
 ):
     try:
-        import yappi
+        import yappi  # ty: ignore[unresolved-import]
     except ImportError:
         print("Failed to run profiler, yappi is not installed")  # noqa: T201
         yield
@@ -79,7 +79,7 @@ def yappi_profile(
 def instrument(html_output=False):
     """Run a statistical profiler"""
     try:
-        from pyinstrument import Profiler
+        from pyinstrument import Profiler  # ty: ignore[unresolved-import]
     except ImportError:
         print("Failed to run profiler, pyinstrument is not installed")  # noqa: T201
         yield
@@ -123,7 +123,7 @@ def debug():
         yield
     except Exception:
         try:
-            import ipdb as pdb  # noqa: T100
+            import ipdb as pdb  # noqa: T100  # ty: ignore[unresolved-import]
         except ImportError:
             import pdb  # type: ignore[no-redef]  # noqa: T100
         pdb.post_mortem()

dvc/analytics.py

@@ -1,10 +1,14 @@
 import json
 import os
+from typing import TYPE_CHECKING, Optional
 
 from dvc.log import logger
 
 from .env import DVC_ANALYTICS_ENDPOINT, DVC_NO_ANALYTICS
 
+if TYPE_CHECKING:
+    from dvc.scm import Base
+
 logger = logger.getChild(__name__)
 
 
@@ -67,6 +71,7 @@ def send(path):
     thus, the need of removing such file afterwards.
     """
     import requests
+    from requests.exceptions import RequestException
 
     url = os.environ.get(DVC_ANALYTICS_ENDPOINT, "https://analytics.dvc.org")
     headers = {"content-type": "application/json"}
@@ -81,7 +86,7 @@ def send(path):
 
     try:
         requests.post(url, json=report, headers=headers, timeout=5)
-    except requests.exceptions.RequestException as e:
+    except RequestException as e:
         logger.trace("", exc_info=True)
         logger.debug("failed to send analytics report %s", str(e))
 
@@ -89,20 +94,62 @@ def send(path):
     os.remove(path)
 
 
-def _scm_in_use():
-    from dvc.exceptions import NotDvcRepoError
-    from dvc.repo import Repo
-    from dvc.scm import NoSCM
+def _git_remote_url(scm: Optional["Base"]) -> Optional[str]:
+    from dvc.scm import Git
 
-    from .scm import SCM, SCMError
+    if not isinstance(scm, Git):
+        return None
+
+    from dulwich.porcelain import get_remote_repo
+
+    dulwich_repo = scm.dulwich.repo
+    try:
+        _remote, url = get_remote_repo(dulwich_repo)
+    except IndexError:
+        # IndexError happens when the head is detached
+        _remote, url = get_remote_repo(dulwich_repo, b"origin")
+    # Dulwich returns (None, "origin") if no remote set
+    if (_remote, url) == (None, "origin"):
+        return None
+    return url
+
+
+def _scm_in_use(scm: Optional["Base"]) -> Optional[str]:
+    return type(scm).__name__ if scm else None
+
+
+def _parse_git_remote_path(remote_url: str) -> str:
+    from urllib.parse import urlparse
+
+    from scmrepo.urls import is_scp_style_url
+
+    parsed = urlparse(remote_url)
+    # Windows Path also gets parsed with a drive letter as scheme
+    # https://github.com/python/cpython/issues/86381
+    if parsed.scheme and parsed.scheme in ("http", "https", "git", "ssh"):
+        return parsed.path.strip("/")
+
+    if is_scp_style_url(remote_url):
+        # handle scp-style URL
+        parts = remote_url.split(":", 1)
+        if len(parts) == 2:
+            _, path = parts
+            return path.rstrip("/")
+    return remote_url
+
+
+def _git_remote_path_hash(scm: Optional["Base"]) -> Optional[str]:
+    """Return a hash of the git remote path."""
+    import hashlib
 
     try:
-        scm = SCM(root_dir=Repo.find_root())
-        return type(scm).__name__
-    except SCMError:
-        return NoSCM.__name__
-    except NotDvcRepoError:
-        pass
+        if remote_url := _git_remote_url(scm):
+            path = _parse_git_remote_path(remote_url)
+            h = hashlib.md5(path.encode("utf-8"), usedforsecurity=False)  # for FIPS
+            return h.hexdigest()
+    except Exception:  # noqa: BLE001
+        logger.debug("Failed to get git remote path", exc_info=True)
+    return None
 
 
 def _runtime_info():
@@ -112,6 +159,8 @@ def _runtime_info():
     from iterative_telemetry import _generate_ci_id, find_or_create_user_id
 
     from dvc import __version__
+    from dvc.info import _get_remotes
+    from dvc.repo import Repo
     from dvc.utils import is_binary
 
     ci_id = _generate_ci_id()
@@ -120,13 +169,24 @@ def _runtime_info():
     else:
         group_id, user_id = None, find_or_create_user_id()
 
+    scm = None
+    remotes = None
+    try:
+        repo = Repo()
+        scm = repo.scm
+        remotes = _get_remotes(repo.config)
+    except Exception as exc:  # noqa: BLE001
+        logger.debug("failed to open repo: %s", exc)
+
     return {
         "dvc_version": __version__,
         "is_binary": is_binary(),
-        "scm_class": _scm_in_use(),
+        "scm_class": _scm_in_use(scm),
         "system_info": _system_info(),
         "user_id": user_id,
         "group_id": group_id,
+        "remotes": remotes,
+        "git_remote_hash": _git_remote_path_hash(scm),
     }
 
 

dvc/api/show.py

@@ -27,8 +27,7 @@ def _postprocess(results):
             }
             processed[rev] = processed[rev] | to_merge
 
-    if "workspace" in processed:
-        del processed["workspace"]
+    processed.pop("workspace", None)
 
     return processed
 

dvc/cachemgr.py

@@ -86,7 +86,7 @@ class CacheManager:
         return FileStorage(
             key=(),
             fs=self.local.fs,
-            path=self.local.fs.join(self.default_local_cache_dir, self.FS_DIR),
+            path=self.local.fs.join(self.local_cache_dir, self.FS_DIR),
         )
 
     def _init_odb(self, schemes):

dvc/cli/__init__.py

@@ -157,6 +157,8 @@ def main(argv=None):  # noqa: C901, PLR0912, PLR0915
     Returns:
         int: command's return code.
     """
+    from contextlib import ExitStack
+
     from dvc._debug import debugtools
     from dvc.config import ConfigError
     from dvc.exceptions import DvcException, NotDvcRepoError
@@ -173,12 +175,11 @@ def main(argv=None):  # noqa: C901, PLR0912, PLR0915
         logging.disable(logging.INFO)
 
     args = None
-
-    outer_log_level = logger.level
-    level = None
+    stack = ExitStack()
     try:
         args = parse_args(argv)
 
+        level = None
         if args.quiet:
             level = logging.CRITICAL
         elif args.verbose == 1:
@@ -187,7 +188,7 @@ def main(argv=None):  # noqa: C901, PLR0912, PLR0915
             level = logging.TRACE  # type: ignore[attr-defined]
 
         if level is not None:
-            set_loggers_level(level)
+            stack.enter_context(set_loggers_level(level))
 
         if level and level <= logging.DEBUG:
             from platform import platform, python_implementation, python_version
@@ -247,7 +248,7 @@ def main(argv=None):  # noqa: C901, PLR0912, PLR0915
 
         return ret
     finally:
-        logger.setLevel(outer_log_level)
+        stack.close()
 
         from dvc.repo.open_repo import clean_repos
 

dvc/cli/command.py

@@ -18,7 +18,10 @@ class CmdBase(ABC):
 
         os.chdir(args.cd)
 
-        self.repo: Repo = Repo(uninitialized=self.UNINITIALIZED)
+        self.repo: Repo = Repo(
+            uninitialized=self.UNINITIALIZED,
+            _wait_for_lock=args.wait_for_lock,
+        )
         self.config: Config = self.repo.config
         self.args = args
 

dvc/cli/completion.py

@@ -22,6 +22,11 @@ _dvc_compgen_exps() {
     local _dvc_exps=($(dvc exp list -q --all-commits --names-only))
     compgen -W "${_dvc_exps[*]}" -- $1
 }
+
+_dvc_compgen_remotes() {
+    local _dvc_remotes=($(dvc remote list | cut -d' ' -f1))
+    compgen -W "${_dvc_remotes[*]}" -- $1
+}
 """
 
 ZSH_PREAMBLE = """
@@ -55,6 +60,10 @@ _dvc_compadd_stages_and_files() {
 _dvc_compadd_exps() {
     _describe 'experiments' "($(dvc exp list -q --all-commits --names-only))"
 }
+
+_dvc_compadd_remotes() {
+    _describe 'remotes' "($(dvc remote list | cut -d' ' -f1))"
+}
 """
 
 PREAMBLE = {
@@ -71,3 +80,4 @@ DVCFILES_AND_STAGE = {
     "zsh": "_dvc_compadd_stages_and_files",
 }
 EXPERIMENT = {"bash": "_dvc_compgen_exps", "zsh": "_dvc_compadd_exps"}
+REMOTE = {"bash": "_dvc_compgen_remotes", "zsh": "_dvc_compadd_remotes"}

dvc/cli/parser.py

@@ -200,6 +200,14 @@ def get_main_parser():
         type=str,
     )
 
+    parser.add_argument(
+        "--wait-for-lock",
+        action="store_true",
+        default=False,
+        help="Wait for the lock if it is already held by another process, instead of"
+        " failing immediately.",
+    )
+
     # Sub commands
     subparsers = parser.add_subparsers(
         title="Available Commands",

dvc/cli/utils.py

@@ -40,7 +40,7 @@ def hide_subparsers_from_help(subparsers):
     # from the 'positional arguments' choices list
     # see: https://bugs.python.org/issue22848
     # Need to set `add_help=False`, but avoid setting `help`
-    # (not even to `argparse.SUPPPRESS`).
+    # (not even to `argparse.SUPPRESS`).
     # NOTE: The argument is the parent subparser, not the subcommand parser.
     cmds = [cmd for cmd, parser in subparsers.choices.items() if parser.add_help]
     subparsers.metavar = "{{{}}}".format(",".join(cmds))

dvc/commands/add.py

@@ -101,7 +101,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         help="Remote storage to download to",
         metavar="<name>",
-    )
+    ).complete = completion.REMOTE
     parser.add_argument(
         "--remote-jobs",
         type=int,

dvc/commands/artifacts.py

@@ -149,7 +149,7 @@ def add_parser(subparsers, parent_parser):
             "Remote name to set as a default in the target repository "
             "(only applicable when downloading from DVC remote)."
         ),
-    )
+    ).complete = completion.REMOTE
     get_parser.add_argument(
         "--remote-config",
         type=str,

dvc/commands/check_ignore.py

@@ -1,20 +1,28 @@
+from typing import TYPE_CHECKING
+
 from dvc.cli import completion, formatter
 from dvc.cli.command import CmdBase
 from dvc.cli.utils import append_doc_link
 from dvc.ui import ui
 
+if TYPE_CHECKING:
+    from dvc.ignore import CheckIgnoreResult
+
 
 class CmdCheckIgnore(CmdBase):
     def __init__(self, args):
         super().__init__(args)
         self.ignore_filter = self.repo.dvcignore
 
-    def _show_results(self, result):
-        if not result.match and not self.args.non_matching:
+    def _show_results(self, result: "CheckIgnoreResult"):
+        if not result.match and not self.args.details:
             return
 
         if self.args.details:
-            patterns = result.patterns
+            pattern_infos = result.pattern_infos
+            patterns = [str(pi) for pi in pattern_infos]
+            if not patterns and self.args.non_matching:
+                patterns = ["::"]
             if not self.args.all:
                 patterns = patterns[-1:]
 
@@ -22,13 +30,11 @@ class CmdCheckIgnore(CmdBase):
                 ui.write(pattern, result.file, sep="\t")
         else:
             ui.write(result.file)
+        return bool(result.pattern_infos)
 
     def _check_one_file(self, target):
         result = self.ignore_filter.check_ignore(target)
-        self._show_results(result)
-        if result.match:
-            return 0
-        return 1
+        return bool(self._show_results(result))
 
     def _interactive_mode(self):
         ret = 1
@@ -39,14 +45,14 @@ class CmdCheckIgnore(CmdBase):
                 break
             if not target:
                 break
-            if not self._check_one_file(target):
+            if self._check_one_file(target):
                 ret = 0
         return ret
 
     def _normal_mode(self):
         ret = 1
         for target in self.args.targets:
-            if not self._check_one_file(target):
+            if self._check_one_file(target):
                 ret = 0
         return ret
 
@@ -76,6 +82,8 @@ class CmdCheckIgnore(CmdBase):
 
 
 def add_parser(subparsers, parent_parser):
+    import argparse
+
     ADD_HELP = "Check whether files or directories are excluded due to `.dvcignore`."
 
     parser = subparsers.add_parser(
@@ -93,14 +101,7 @@ def add_parser(subparsers, parent_parser):
         help="Show the exclude patterns along with each target path.",
     )
     parser.add_argument(
-        "-a",
-        "--all",
-        action="store_true",
-        default=False,
-        help=(
-            "Include the target paths which don't match any pattern "
-            "in the `--details` list."
-        ),
+        "-a", "--all", action="store_true", default=False, help=argparse.SUPPRESS
     )
     parser.add_argument(
         "-n",

dvc/commands/checkout.py

@@ -1,5 +1,3 @@
-import operator
-
 from dvc.cli import completion, formatter
 from dvc.cli.command import CmdBase
 from dvc.cli.utils import append_doc_link
@@ -30,7 +28,7 @@ class CmdCheckout(CmdBase):
 
         stats, exc = None, None
         try:
-            stats = self.repo.checkout(
+            result = self.repo.checkout(
                 targets=self.args.targets,
                 with_deps=self.args.with_deps,
                 force=self.args.force,
@@ -40,14 +38,17 @@ class CmdCheckout(CmdBase):
             )
         except CheckoutError as _exc:
             exc = _exc
-            stats = exc.stats
+            result = exc.result
 
         if self.args.summary:
             default_message = "No changes."
-            msg = get_summary(sorted(stats.items(), key=operator.itemgetter(0)))
+            stats = result["stats"]
+            assert isinstance(stats, dict)
+            msg = get_summary(stats.items())
             ui.write(msg or default_message)
         else:
-            log_changes(stats)
+            result.pop("stats", {})
+            log_changes(result)
 
         if exc:
             raise exc

dvc/commands/data.py

@@ -2,7 +2,7 @@ from typing import TYPE_CHECKING, ClassVar
 
 from funcy import chunks, compact, log_durations
 
-from dvc.cli import formatter
+from dvc.cli import completion, formatter
 from dvc.cli.command import CmdBase
 from dvc.cli.utils import append_doc_link
 from dvc.log import logger
@@ -10,6 +10,7 @@ from dvc.ui import ui
 from dvc.utils import colorize
 
 if TYPE_CHECKING:
+    from dvc.repo.data import GitInfo
     from dvc.repo.data import Status as DataStatus
 
 
@@ -55,20 +56,15 @@ class CmdDataStatus(CmdBase):
     def _process_status(status: "DataStatus"):
         """Flatten stage status, and filter empty stage status contents."""
         for stage, stage_status in status.items():
-            items = stage_status
-            if isinstance(stage_status, dict):
-                items = {
-                    file: state
-                    for state, files in stage_status.items()
-                    for file in files
-                }
-            if not items:
+            if not stage_status or (
+                isinstance(stage_status, dict) and not any(stage_status.values())
+            ):
                 continue
-            yield stage, items
+            yield stage, stage_status
 
     @classmethod
     def _show_status(cls, status: "DataStatus") -> int:  # noqa: C901
-        git_info = status.pop("git")  # type: ignore[misc]
+        git_info: GitInfo = status.pop("git")  # type: ignore[misc]
         result = dict(cls._process_status(status))
         if not result:
             no_changes = "No changes"
@@ -90,7 +86,16 @@ class CmdDataStatus(CmdBase):
                     ui.write(f"  ({hint})")
 
             if isinstance(stage_status, dict):
-                items = [f"{state}: {file}" for file, state in stage_status.items()]
+                items = [
+                    f"{state}: "
+                    + (
+                        " -> ".join(change.values())
+                        if isinstance(change, dict)
+                        else change
+                    )
+                    for state, changes in stage_status.items()
+                    for change in changes
+                ]
             else:
                 items = stage_status
 
@@ -108,8 +113,10 @@ class CmdDataStatus(CmdBase):
     def run(self) -> int:
         with log_durations(logger.trace, "in data_status"):
             status = self.repo.data_status(
+                targets=self.args.targets,
                 granular=self.args.granular,
                 untracked_files=self.args.untracked_files,
+                remote=self.args.remote,
                 not_in_remote=self.args.not_in_remote,
                 remote_refresh=self.args.remote_refresh,
             )
@@ -126,9 +133,12 @@ class CmdDataStatus(CmdBase):
 
 
 def add_parser(subparsers, parent_parser):
+    data_help = "Commands related to data management."
     data_parser = subparsers.add_parser(
         "data",
         parents=[parent_parser],
+        description=append_doc_link(data_help, "data/status"),
+        help=data_help,
         formatter_class=formatter.RawDescriptionHelpFormatter,
     )
     data_subparsers = data_parser.add_subparsers(
@@ -147,6 +157,14 @@ def add_parser(subparsers, parent_parser):
         formatter_class=formatter.RawDescriptionHelpFormatter,
         help=DATA_STATUS_HELP,
     )
+    data_status_parser.add_argument(
+        "targets",
+        nargs="*",
+        help=(
+            "Limit command scope to these tracked files/directories, "
+            ".dvc files and stage names."
+        ),
+    ).complete = completion.FILE  # type: ignore[attr-defined]
     data_status_parser.add_argument(
         "--json",
         action="store_true",
@@ -173,6 +191,12 @@ def add_parser(subparsers, parent_parser):
         nargs="?",
         help="Show untracked files.",
     )
+    data_status_parser.add_argument(
+        "-r",
+        "--remote",
+        help="Remote storage to check (only applicable with --not-in-remote).",
+        metavar="<name>",
+    ).complete = completion.REMOTE
     data_status_parser.add_argument(
         "--not-in-remote",
         action="store_true",

dvc/commands/data_sync.py

@@ -22,17 +22,18 @@ class CmdDataBase(CmdBase):
 
 
 class CmdDataPull(CmdDataBase):
-    def log_summary(self, stats):
+    def log_summary(self, result):
         from dvc.commands.checkout import log_changes
 
-        log_changes(stats)
+        stats = result.pop("stats", {})
+        log_changes(result)
         super().log_summary(stats)
 
     def run(self):
         from dvc.exceptions import CheckoutError, DvcException
 
         try:
-            stats = self.repo.pull(
+            result = self.repo.pull(
                 targets=self.args.targets,
                 jobs=self.args.jobs,
                 remote=self.args.remote,
@@ -46,10 +47,10 @@ class CmdDataPull(CmdDataBase):
                 glob=self.args.glob,
                 allow_missing=self.args.allow_missing,
             )
-            self.log_summary(stats)
+            self.log_summary(result)
         except (CheckoutError, DvcException) as exc:
-            if stats := getattr(exc, "stats", {}):
-                self.log_summary(stats)
+            if result := getattr(exc, "result", {}):
+                self.log_summary(result)
             logger.exception("failed to pull data from the cloud")
             return 1
 
@@ -149,7 +150,7 @@ def add_parser(subparsers, _parent_parser):
     )
     pull_parser.add_argument(
         "-r", "--remote", help="Remote storage to pull from", metavar="<name>"
-    )
+    ).complete = completion.REMOTE
     pull_parser.add_argument(
         "-a",
         "--all-branches",
@@ -224,7 +225,7 @@ def add_parser(subparsers, _parent_parser):
     )
     push_parser.add_argument(
         "-r", "--remote", help="Remote storage to push to", metavar="<name>"
-    )
+    ).complete = completion.REMOTE
     push_parser.add_argument(
         "-a",
         "--all-branches",
@@ -286,7 +287,7 @@ def add_parser(subparsers, _parent_parser):
     )
     fetch_parser.add_argument(
         "-r", "--remote", help="Remote storage to fetch from", metavar="<name>"
-    )
+    ).complete = completion.REMOTE
     fetch_parser.add_argument(
         "-a",
         "--all-branches",
@@ -379,7 +380,7 @@ def add_parser(subparsers, _parent_parser):
         "--remote",
         help="Remote storage to compare local cache to",
         metavar="<name>",
-    )
+    ).complete = completion.REMOTE
     status_parser.add_argument(
         "-a",
         "--all-branches",

dvc/commands/du.py

@@ -60,7 +60,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         type=str,
         help="Remote name to set as a default in the target repository.",
-    )
+    ).complete = completion.REMOTE
     du_parser.add_argument(
         "--remote-config",
         type=str,

dvc/commands/experiments/pull.py

@@ -1,6 +1,6 @@
 import argparse
 
-from dvc.cli import formatter
+from dvc.cli import completion, formatter
 from dvc.cli.command import CmdBase
 from dvc.cli.utils import append_doc_link
 from dvc.log import logger
@@ -71,7 +71,7 @@ def add_parser(experiments_subparsers, parent_parser):
         dest="dvc_remote",
         metavar="<name>",
         help="Name of the DVC remote to use when pulling cached outputs.",
-    )
+    ).complete = completion.REMOTE
     experiments_pull_parser.add_argument(
         "-j",
         "--jobs",

dvc/commands/experiments/push.py

@@ -111,7 +111,7 @@ def add_parser(experiments_subparsers, parent_parser):
         dest="dvc_remote",
         metavar="<name>",
         help="Name of the DVC remote to use when pushing cached outputs.",
-    )
+    ).complete = completion.REMOTE
     experiments_push_parser.add_argument(
         "-j",
         "--jobs",

dvc/commands/experiments/show.py

@@ -1,6 +1,6 @@
 import argparse
 import re
-from collections.abc import Iterable
+from collections.abc import Collection, Iterable
 from datetime import date, datetime
 from typing import TYPE_CHECKING
 
@@ -59,17 +59,16 @@ def baseline_styler(typ):
 def show_experiments(
     td: "TabularData",
     headers: dict[str, Iterable[str]],
-    keep=None,
-    drop=None,
+    keep: Collection[str] = (),
+    drop: Collection[str] = (),
     pager=True,
     csv=False,
     markdown=False,
     **kwargs,
 ):
     if keep:
-        for col in td.keys():  # noqa: SIM118
-            if re.match(keep, col):
-                td.protect(col)
+        keep_re = re.compile("|".join(keep))
+        td.protect(*(col for col in td.keys() if keep_re.match(col)))  # noqa: SIM118
 
     for col in ("State", "Executor"):
         if td.is_empty(col):
@@ -110,8 +109,9 @@ def show_experiments(
         td.drop_duplicates("cols", ignore_empty=False)
 
     cols_to_drop = set()
-    if drop is not None:
-        cols_to_drop = {col for col in td.keys() if re.match(drop, col)}  # noqa: SIM118
+    if drop:
+        drop_re = re.compile("|".join(drop))
+        cols_to_drop = {col for col in td.keys() if drop_re.match(col)}  # noqa: SIM118
     td.drop(*cols_to_drop)
 
     td.render(
@@ -150,6 +150,7 @@ class CmdExperimentsShow(CmdBase):
                 all_commits=self.args.all_commits,
                 hide_queued=self.args.hide_queued,
                 hide_failed=self.args.hide_failed,
+                hide_workspace=self.args.hide_workspace,
                 revs=self.args.rev,
                 num=self.args.num,
                 sha_only=self.args.sha,
@@ -236,11 +237,15 @@ def add_parser(experiments_subparsers, parent_parser):
     )
     experiments_show_parser.add_argument(
         "--drop",
+        action="append",
+        default=[],
         help="Remove the columns matching the specified regex pattern.",
         metavar="<regex_pattern>",
     )
     experiments_show_parser.add_argument(
         "--keep",
+        action="append",
+        default=[],
         help="Preserve the columns matching the specified regex pattern.",
         metavar="<regex_pattern>",
     )
@@ -279,6 +284,12 @@ def add_parser(experiments_subparsers, parent_parser):
         default=False,
         help="Hide queued experiments in the table.",
     )
+    experiments_show_parser.add_argument(
+        "--hide-workspace",
+        action="store_true",
+        default=False,
+        help="Hide workspace row in the table.",
+    )
     experiments_show_parser.add_argument(
         "--json",
         action="store_true",

dvc/commands/gc.py

@@ -1,6 +1,6 @@
 import os
 
-from dvc.cli import formatter
+from dvc.cli import completion, formatter
 from dvc.cli.command import CmdBase
 from dvc.cli.utils import append_doc_link
 from dvc.log import logger
@@ -188,7 +188,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         help="Remote storage to collect garbage in",
         metavar="<name>",
-    )
+    ).complete = completion.REMOTE
     gc_parser.add_argument(
         "--skip-failed",
         action="store_true",

dvc/commands/get.py

@@ -121,7 +121,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         type=str,
         help="Remote name to set as a default in the target repository.",
-    )
+    ).complete = completion.REMOTE
     get_parser.add_argument(
         "--remote-config",
         type=str,

dvc/commands/imp.py

@@ -115,7 +115,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         type=str,
         help="Remote name to set as a default in the target repository.",
-    )
+    ).complete = completion.REMOTE
     import_parser.add_argument(
         "--remote-config",
         type=str,

dvc/commands/imp_url.py

@@ -73,7 +73,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         help="Remote storage to download to",
         metavar="<name>",
-    )
+    ).complete = completion.REMOTE
     no_download_exec_group = import_parser.add_mutually_exclusive_group()
     no_download_exec_group.add_argument(
         "--no-exec",

dvc/commands/init.py

@@ -42,7 +42,7 @@ class CmdInit(CmdBaseNoRepo):
 
         try:
             with Repo.init(
-                ".",
+                self.args.directory,
                 no_scm=self.args.no_scm,
                 force=self.args.force,
                 subdir=self.args.subdir,
@@ -57,10 +57,11 @@ class CmdInit(CmdBaseNoRepo):
 
 def add_parser(subparsers, parent_parser):
     """Setup parser for `dvc init`."""
-    INIT_HELP = "Initialize DVC in the current directory."
+    INIT_HELP = "Initialize DVC repository."
     INIT_DESCRIPTION = (
-        "Initialize DVC in the current directory. Expects directory\n"
-        "to be a Git repository unless --no-scm option is specified."
+        "Initialize DVC repository in the given directory (defaults to the current "
+        "working directory).\n"
+        "Expects directory to be a Git repository unless --no-scm option is specified."
     )
 
     init_parser = subparsers.add_parser(
@@ -70,6 +71,14 @@ def add_parser(subparsers, parent_parser):
         help=INIT_HELP,
         formatter_class=formatter.RawDescriptionHelpFormatter,
     )
+    init_parser.add_argument(
+        "directory",
+        nargs="?",
+        default=".",
+        help=(
+            "Directory to initialize DVC in. Defaults to the current working directory."
+        ),
+    )
     init_parser.add_argument(
         "--no-scm",
         action="store_true",

dvc/commands/ls/__init__.py

@@ -28,7 +28,7 @@ def _format_entry(entry, name, with_size=True, with_hash=False):
     ret = []
     if with_size:
         size = entry.get("size")
-        if size is None:
+        if size is None or (size <= 0 and entry.get("isdir")):
             size = ""
         else:
             size = naturalsize(size)
@@ -77,10 +77,7 @@ def _build_tree_structure(
 
     num_entries = len(entries)
     for i, (name, entry) in enumerate(entries.items()):
-        # show full path for root, otherwise only show the name
-        if _depth > 0:
-            entry["path"] = name
-
+        entry["path"] = name
         is_last = i >= num_entries - 1
         tree_part = ""
         if _depth > 0:
@@ -253,7 +250,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         type=str,
         help="Remote name to set as a default in the target repository.",
-    )
+    ).complete = completion.REMOTE
     list_parser.add_argument(
         "--remote-config",
         type=str,

dvc/commands/ls_url.py

@@ -3,26 +3,45 @@ from dvc.cli.command import CmdBaseNoRepo
 from dvc.cli.utils import DictAction, append_doc_link
 from dvc.log import logger
 
-from .ls import show_entries
+from .ls import show_entries, show_tree
 
 logger = logger.getChild(__name__)
 
 
 class CmdListUrl(CmdBaseNoRepo):
-    def run(self):
-        from dvc.config import Config
+    def _show_tree(self, config):
+        from dvc.fs import parse_external_url
+        from dvc.repo.ls import _ls_tree
+
+        fs, fs_path = parse_external_url(
+            self.args.url, fs_config=self.args.fs_config, config=config
+        )
+        entries = _ls_tree(fs, fs_path, maxdepth=self.args.level)
+        show_tree(entries, with_color=True, with_size=self.args.size)
+        return 0
+
+    def _show_list(self, config):
         from dvc.repo import Repo
 
         entries = Repo.ls_url(
             self.args.url,
             recursive=self.args.recursive,
+            maxdepth=self.args.level,
             fs_config=self.args.fs_config,
-            config=Config.from_cwd(),
+            config=config,
         )
         if entries:
             show_entries(entries, with_color=True, with_size=self.args.size)
         return 0
 
+    def run(self):
+        from dvc.config import Config
+
+        config = Config.from_cwd()
+        if self.args.tree:
+            return self._show_tree(config=config)
+        return self._show_list(config=config)
+
 
 def add_parser(subparsers, parent_parser):
     LS_HELP = "List directory contents from URL."
@@ -40,6 +59,19 @@ def add_parser(subparsers, parent_parser):
     lsurl_parser.add_argument(
         "-R", "--recursive", action="store_true", help="Recursively list files."
     )
+    lsurl_parser.add_argument(
+        "-T",
+        "--tree",
+        action="store_true",
+        help="Recurse into directories as a tree.",
+    )
+    lsurl_parser.add_argument(
+        "-L",
+        "--level",
+        metavar="depth",
+        type=int,
+        help="Limit the depth of recursion.",
+    )
     lsurl_parser.add_argument("--size", action="store_true", help="Show sizes.")
     lsurl_parser.add_argument(
         "--fs-config",

dvc/commands/remote.py

@@ -1,4 +1,4 @@
-from dvc.cli import formatter
+from dvc.cli import completion, formatter
 from dvc.cli.utils import append_doc_link
 from dvc.commands.config import CmdConfig
 from dvc.ui import ui
@@ -110,8 +110,16 @@ class CmdRemoteDefault(CmdRemote):
 class CmdRemoteList(CmdRemote):
     def run(self):
         conf = self.config.read(self.args.level)
+        default_remote = conf["core"].get("remote")
+
         for name, remote_conf in conf["remote"].items():
-            ui.write(name, remote_conf["url"], sep="\t")
+            if name == default_remote:
+                text = f"{name}\t{remote_conf['url']}\t(default)"
+                color = "green"
+            else:
+                text = f"{name}\t{remote_conf['url']}"
+                color = ""
+            ui.write(ui.rich_text(text, style=color), styled=True)
         return 0
 
 
@@ -202,7 +210,9 @@ def add_parser(subparsers, parent_parser):
         help=REMOTE_DEFAULT_HELP,
         formatter_class=formatter.RawDescriptionHelpFormatter,
     )
-    remote_default_parser.add_argument("name", nargs="?", help="Name of the remote")
+    remote_default_parser.add_argument(
+        "name", nargs="?", help="Name of the remote"
+    ).complete = completion.REMOTE
     remote_default_parser.add_argument(
         "-u",
         "--unset",
@@ -220,7 +230,9 @@ def add_parser(subparsers, parent_parser):
         help=REMOTE_MODIFY_HELP,
         formatter_class=formatter.RawDescriptionHelpFormatter,
     )
-    remote_modify_parser.add_argument("name", help="Name of the remote")
+    remote_modify_parser.add_argument(
+        "name", help="Name of the remote"
+    ).complete = completion.REMOTE
     remote_modify_parser.add_argument("option", help="Name of the option to modify.")
     remote_modify_parser.add_argument(
         "value", nargs="?", help="(optional) Value of the option."
@@ -252,7 +264,9 @@ def add_parser(subparsers, parent_parser):
         help=REMOTE_REMOVE_HELP,
         formatter_class=formatter.RawDescriptionHelpFormatter,
     )
-    remote_remove_parser.add_argument("name", help="Name of the remote to remove.")
+    remote_remove_parser.add_argument(
+        "name", help="Name of the remote to remove."
+    ).complete = completion.REMOTE
     remote_remove_parser.set_defaults(func=CmdRemoteRemove)
     REMOTE_RENAME_HELP = "Rename a DVC remote"
     remote_rename_parser = remote_subparsers.add_parser(

dvc/commands/status.py

@@ -71,7 +71,7 @@ class CmdDataStatus(CmdDataBase):
                 return 0
 
             if self.args.quiet:
-                return bool(st)
+                return int(bool(st))
 
             if st:
                 self._show(st, indent)

dvc/commands/update.py

@@ -74,7 +74,7 @@ def add_parser(subparsers, parent_parser):
         "--remote",
         help="Remote storage to perform updates to",
         metavar="<name>",
-    )
+    ).complete = completion.REMOTE
     update_parser.add_argument(
         "-j",
         "--jobs",

dvc/config.py

@@ -1,6 +1,8 @@
 """DVC config objects."""
 
+import ntpath
 import os
+import posixpath
 import re
 from contextlib import contextmanager
 from functools import partial
@@ -268,12 +270,17 @@ class Config(dict):
         if re.match(r"\w+://", path):
             return path
 
+        if os.name == "nt" and posixpath.isabs(path) and ntpath.sep not in path:
+            return path
+
         if os.path.isabs(path):
             return path
 
         # on windows convert slashes to backslashes
         # to have path compatible with abs_conf_dir
         if os.path.sep == "\\" and "/" in path:
+            if path.startswith("/"):
+                path = path.replace("/", "\\\\", 1)
             path = path.replace("/", "\\")
 
         expanded = os.path.expanduser(path)
@@ -305,6 +312,9 @@ class Config(dict):
         if os.path.expanduser(path) != path:
             return localfs.as_posix(path)
 
+        if os.name == "nt" and posixpath.isabs(path) and ntpath.sep not in path:
+            return path
+
         if isinstance(path, RelPath) or not os.path.isabs(path):
             path = relpath(path, conf_dir)
             return localfs.as_posix(path)
@@ -330,6 +340,7 @@ class Config(dict):
                     "gdrive_user_credentials_file": func,
                     "gdrive_service_account_json_file_path": func,
                     "credentialpath": func,
+                    "configpath": func,
                     "keyfile": func,
                     "cert_path": func,
                     "key_path": func,

dvc/config_schema.py

@@ -215,6 +215,7 @@ SCHEMA = {
                     "credentialpath": str,
                     "endpointurl": str,
                     Optional("verify", default=False): Bool,
+                    Optional("allow_anonymous_login", default=False): Bool,
                     **REMOTE_COMMON,
                 },
                 "ssh": {

dvc/daemon.py

@@ -32,10 +32,10 @@ def _win_detached_subprocess(args: Sequence[str], **kwargs) -> int:
     assert os.name == "nt"
 
     from subprocess import (  # type: ignore[attr-defined]
-        CREATE_NEW_PROCESS_GROUP,
-        CREATE_NO_WINDOW,
-        STARTF_USESHOWWINDOW,
-        STARTUPINFO,
+        CREATE_NEW_PROCESS_GROUP,  # ty: ignore[unresolved-import]
+        CREATE_NO_WINDOW,  # ty: ignore[unresolved-import]
+        STARTF_USESHOWWINDOW,  # ty: ignore[unresolved-import]
+        STARTUPINFO,  # ty: ignore[unresolved-import]
     )
 
     # https://stackoverflow.com/a/7006424

dvc/dagascii.py

@@ -104,21 +104,21 @@ class AsciiCanvas:
                 if dx == 0:
                     y = y0
                 else:
-                    y = y0 + int(round((x - x0) * dy / float(dx)))
+                    y = y0 + round((x - x0) * dy / float(dx))
                 self.point(x, y, char)
         elif y0 < y1:
             for y in range(y0, y1 + 1):
                 if dy == 0:
                     x = x0
                 else:
-                    x = x0 + int(round((y - y0) * dx / float(dy)))
+                    x = x0 + round((y - y0) * dx / float(dy))
                 self.point(x, y, char)
         else:
             for y in range(y1, y0 + 1):
                 if dy == 0:
                     x = x0
                 else:
-                    x = x1 + int(round((y - y1) * dx / float(dy)))
+                    x = x1 + round((y - y1) * dx / float(dy))
                 self.point(x, y, char)
 
     def text(self, x, y, text):
@@ -195,7 +195,7 @@ def _build_sugiyama_layout(vertices, edges):
 
     sug.yspace = VertexViewer.HEIGHT
     sug.xspace = minw
-    sug.route_edge = route_with_lines
+    sug.route_edge = route_with_lines  # ty: ignore[unresolved-attribute]
 
     sug.draw()
 
@@ -258,8 +258,8 @@ def draw(vertices, edges):
     maxx = max(Xs)
     maxy = max(Ys)
 
-    canvas_cols = int(math.ceil(math.ceil(maxx) - math.floor(minx))) + 1
-    canvas_lines = int(round(maxy - miny))
+    canvas_cols = math.ceil(math.ceil(maxx) - math.floor(minx)) + 1
+    canvas_lines = round(maxy - miny)
 
     canvas = AsciiCanvas(canvas_cols, canvas_lines)
 
@@ -270,10 +270,10 @@ def draw(vertices, edges):
             start = edge.view._pts[index - 1]
             end = edge.view._pts[index]
 
-            start_x = int(round(start[0] - minx))
-            start_y = int(round(start[1] - miny))
-            end_x = int(round(end[0] - minx))
-            end_y = int(round(end[1] - miny))
+            start_x = round(start[0] - minx)
+            start_y = round(start[1] - miny)
+            end_x = round(end[0] - minx)
+            end_y = round(end[1] - miny)
 
             assert start_x >= 0
             assert start_y >= 0
@@ -288,12 +288,12 @@ def draw(vertices, edges):
         y = vertex.view.xy[1]
 
         canvas.box(
-            int(round(x - minx)),
-            int(round(y - miny)),
+            round(x - minx),
+            round(y - miny),
             vertex.view.w,
             vertex.view.h,
         )
 
-        canvas.text(int(round(x - minx)) + 1, int(round(y - miny)) + 1, vertex.data)
+        canvas.text(round(x - minx) + 1, round(y - miny) + 1, vertex.data)
 
     return canvas.draw()

dvc/dependency/repo.py

@@ -146,11 +146,15 @@ class RepoDependency(Dependency):
         else:
             config = Config.load_file(conf)
 
+        # Setup config to the new DVCFileSystem to use the remote repo, but rely on the
+        # local cache instead of the remote's cache. This avoids re-streaming of data,
+        # but messes up the call to `_get_remote_config()` downstream, which will need
+        # to ignore cache parameters.
         config["cache"] = self.repo.config["cache"]
         config["cache"]["dir"] = self.repo.cache.local_cache_dir
 
         return DVCFileSystem(
-            url=self.def_repo[self.PARAM_URL],
+            repo=self.def_repo[self.PARAM_URL],
             rev=rev or self._get_rev(locked=locked),
             subrepos=True,
             config=config,

dvc/dirs.py

@@ -1,4 +1,5 @@
 import os
+from typing import Optional
 
 import platformdirs
 
@@ -20,10 +21,16 @@ def global_config_dir():
     )
 
 
-def site_cache_dir():
+def site_cache_dir(config_site_cache_dir: Optional[str] = None):
     from platformdirs import PlatformDirs
     from platformdirs.unix import Unix
 
+    if dvc_site_cache_dir := os.getenv(env.DVC_SITE_CACHE_DIR):
+        return dvc_site_cache_dir
+
+    if config_site_cache_dir is not None:
+        return config_site_cache_dir
+
     if issubclass(Unix, PlatformDirs):
         # Return the cache directory shared by users, e.g. `/var/tmp/$appname`
         # NOTE: platformdirs>=5 changed `site_cache_dir` to return /var/cache/$appname.
@@ -34,6 +41,4 @@ def site_cache_dir():
         #        In the future, consider migrating to a more secure directory.
         return f"/var/tmp/{APPNAME}"  # noqa: S108
 
-    return os.getenv(env.DVC_SITE_CACHE_DIR) or platformdirs.site_cache_dir(
-        APPNAME, APPAUTHOR, opinion=True
-    )
+    return platformdirs.site_cache_dir(APPNAME, APPAUTHOR, opinion=True)

dvc/dvcfile.py

@@ -162,6 +162,9 @@ class FileMixin:
     def dump(self, stage, **kwargs):
         raise NotImplementedError
 
+    def dump_stages(self, stages, **kwargs):
+        raise NotImplementedError
+
     def merge(self, ancestor, other, allowed=None):
         raise NotImplementedError
 
@@ -198,6 +201,13 @@ class SingleStageFile(FileMixin):
         dump_yaml(self.path, serialize.to_single_stage_file(stage, **kwargs))
         self.repo.scm_context.track_file(self.relpath)
 
+    def dump_stages(self, stages, **kwargs) -> None:
+        if not stages:
+            return
+
+        assert len(stages) == 1, "SingleStageFile can only dump one stage."
+        return self.dump(stages[0], **kwargs)
+
     def remove_stage(self, stage):  # noqa: ARG002
         self.remove()
 
@@ -228,17 +238,27 @@ class ProjectFile(FileMixin):
 
     def dump(self, stage, update_pipeline=True, update_lock=True, **kwargs):
         """Dumps given stage appropriately in the dvcfile."""
+        return self.dump_stages(
+            [stage], update_pipeline=update_pipeline, update_lock=update_lock, **kwargs
+        )
+
+    def dump_stages(self, stages, update_pipeline=True, update_lock=True, **kwargs):
         from dvc.stage import PipelineStage
 
-        assert isinstance(stage, PipelineStage)
+        if not stages:
+            return
+
+        for stage in stages:
+            assert isinstance(stage, PipelineStage)
+
         if self.verify:
             check_dvcfile_path(self.repo, self.path)
 
-        if update_pipeline and not stage.is_data_source:
-            self._dump_pipeline_file(stage)
+        if update_pipeline:
+            self._dump_pipeline_file(stages)
 
         if update_lock:
-            self._dump_lockfile(stage, **kwargs)
+            self._dump_lockfile(stages, **kwargs)
 
     def dump_dataset(self, dataset):
         with modify_yaml(self.path, fs=self.repo.fs) as data:
@@ -260,32 +280,37 @@ class ProjectFile(FileMixin):
                 raw.append(dataset)
         self.repo.scm_context.track_file(self.relpath)
 
-    def _dump_lockfile(self, stage, **kwargs):
-        self._lockfile.dump(stage, **kwargs)
+    def _dump_lockfile(self, stages, **kwargs):
+        self._lockfile.dump_stages(stages, **kwargs)
 
     @staticmethod
     def _check_if_parametrized(stage, action: str = "dump") -> None:
         if stage.raw_data.parametrized:
             raise ParametrizedDumpError(f"cannot {action} a parametrized {stage}")
 
-    def _dump_pipeline_file(self, stage):
-        self._check_if_parametrized(stage)
-        stage_data = serialize.to_pipeline_file(stage)
+    def _dump_pipeline_file(self, stages):
+        stages = stages if isinstance(stages, list) else [stages]
+        if not stages:
+            return
+
+        for stage in stages:
+            self._check_if_parametrized(stage)
 
         with modify_yaml(self.path, fs=self.repo.fs) as data:
             if not data:
                 logger.info("Creating '%s'", self.relpath)
 
             data["stages"] = data.get("stages", {})
-            existing_entry = stage.name in data["stages"]
-            action = "Modifying" if existing_entry else "Adding"
-            logger.info("%s stage '%s' in '%s'", action, stage.name, self.relpath)
-
-            if existing_entry:
-                orig_stage_data = data["stages"][stage.name]
-                apply_diff(stage_data[stage.name], orig_stage_data)
-            else:
-                data["stages"].update(stage_data)
+            for stage in stages:
+                stage_data = serialize.to_pipeline_file(stage)
+                existing_entry = stage.name in data["stages"]
+                action = "Modifying" if existing_entry else "Adding"
+                logger.info("%s stage '%s' in '%s'", action, stage.name, self.relpath)
+                if existing_entry:
+                    orig_stage_data = data["stages"][stage.name]
+                    apply_diff(stage_data[stage.name], orig_stage_data)
+                else:
+                    data["stages"].update(stage_data)
 
         self.repo.scm_context.track_file(self.relpath)
 
@@ -399,9 +424,12 @@ class Lockfile(FileMixin):
             data.setdefault("stages", {})
         self.repo.scm_context.track_file(self.relpath)
 
-    def dump(self, stage, **kwargs):
-        stage_data = serialize.to_lockfile(stage, **kwargs)
+    def dump_stages(self, stages, **kwargs):
+        if not stages:
+            return
 
+        is_modified = False
+        log_updated = False
         with modify_yaml(self.path, fs=self.repo.fs) as data:
             if not data:
                 data.update({"schema": "2.0"})
@@ -409,17 +437,24 @@ class Lockfile(FileMixin):
                 logger.info("Generating lock file '%s'", self.relpath)
 
             data["stages"] = data.get("stages", {})
-            modified = data["stages"].get(stage.name, {}) != stage_data.get(
-                stage.name, {}
-            )
-            if modified:
-                logger.info("Updating lock file '%s'", self.relpath)
+            for stage in stages:
+                stage_data = serialize.to_lockfile(stage, **kwargs)
+                modified = data["stages"].get(stage.name, {}) != stage_data.get(
+                    stage.name, {}
+                )
+                if modified:
+                    is_modified = True
+                    if not log_updated:
+                        logger.info("Updating lock file '%s'", self.relpath)
+                        log_updated = True
+                data["stages"].update(stage_data)
 
-            data["stages"].update(stage_data)
-
-        if modified:
+        if is_modified:
             self.repo.scm_context.track_file(self.relpath)
 
+    def dump(self, stage, **kwargs):
+        self.dump_stages([stage], **kwargs)
+
     def remove_stage(self, stage):
         if not self.exists():
             return

dvc/exceptions.py

@@ -242,11 +242,11 @@ class UploadError(FileTransferError):
 
 
 class CheckoutError(DvcException):
-    def __init__(self, target_infos: list[str], stats: dict[str, list[str]]):
+    def __init__(self, target_infos: list[str], result: dict):
         from dvc.utils import error_link
 
         self.target_infos = target_infos
-        self.stats = stats
+        self.result = result
         targets = [str(t) for t in target_infos]
         m = (
             "Checkout failed for following targets:\n{}\nIs your "

dvc/fs/callbacks.py

@@ -1,8 +1,8 @@
 from contextlib import ExitStack
 from typing import TYPE_CHECKING, BinaryIO, Optional, Union
 
-import fsspec
 from fsspec.callbacks import DEFAULT_CALLBACK, Callback  # noqa: F401
+from fsspec.callbacks import TqdmCallback as _TqdmCallback
 
 from dvc.progress import Tqdm
 from dvc.utils.objects import cached_property
@@ -14,7 +14,7 @@ if TYPE_CHECKING:
     from dvc.ui._rich_progress import RichTransferProgress
 
 
-class TqdmCallback(fsspec.callbacks.TqdmCallback):
+class TqdmCallback(_TqdmCallback):
     def __init__(
         self,
         size: Optional[int] = None,

dvc/fs/dvc.py

@@ -83,11 +83,10 @@ class _DVCFileSystem(AbstractFileSystem):
     cachable = False
     root_marker = "/"
 
-    def __init__(  # noqa: PLR0913
+    def __init__(
         self,
-        url: Optional[str] = None,
+        repo: Union["Repo", os.PathLike[str], str, None] = None,
         rev: Optional[str] = None,
-        repo: Optional["Repo"] = None,
         subrepos: bool = False,
         repo_factory: Optional[RepoFactory] = None,
         fo: Optional[str] = None,
@@ -101,7 +100,8 @@ class _DVCFileSystem(AbstractFileSystem):
         """DVC + git-tracked files fs.
 
         Args:
-            path (str, optional): URL or path to a DVC/Git repository.
+            repo (str | os.PathLike[str] | Repo, optional): A url or a path to a DVC/Git
+                repository, or a `Repo` instance.
                 Defaults to a DVC repository in the current working directory.
                 Both HTTP and SSH protocols are supported for remote Git repos
                 (e.g. [user@]server:project.git).
@@ -111,7 +111,6 @@ class _DVCFileSystem(AbstractFileSystem):
                 In case of a local repository, if rev is unspecified, it will
                 default to the working directory.
                 If the repo is not a Git repo, this option is ignored.
-            repo (:obj:`Repo`, optional): `Repo` instance.
             subrepos (bool): traverse to subrepos.
                 By default, it ignores subrepos.
             repo_factory (callable): A function to initialize subrepo with.
@@ -136,13 +135,23 @@ class _DVCFileSystem(AbstractFileSystem):
             ...    rev="main",
             ... )
         """
+        from dvc.repo import Repo
+
+        # kwargs.get("url") is for maintaining backward compatibility
+        repo = repo or fo or kwargs.get("url")
+        if isinstance(repo, Repo):
+            self._repo: Optional[Repo] = repo
+            url = None
+        else:
+            self._repo = None
+            url = os.fspath(repo) if repo else None
+
         super().__init__()
-        self._repo = repo
         self._repo_factory = repo_factory
         self._traverse_subrepos = subrepos
         self._repo_stack = ExitStack()
         self._repo_kwargs = {
-            "url": url if url is not None else fo,
+            "url": url,
             "rev": rev,
             "subrepos": subrepos,
             "config": config,
@@ -500,7 +509,7 @@ class _DVCFileSystem(AbstractFileSystem):
             **kwargs,
         )
 
-    def _get(  # noqa: C901, PLR0912
+    def _get(  # noqa: C901, PLR0912, PLR0915
         self,
         rpath,
         lpath,
@@ -578,7 +587,8 @@ class _DVCFileSystem(AbstractFileSystem):
             kw = kwargs
             if isinstance(fs, DataFileSystem):
                 kw = kw | {"info": info}
-            return fs.get_file(src, dest, callback=callback, **kw)
+            with callback.branched(src, dest) as child:
+                fs.get_file(src, dest, callback=child, **kw)
 
         if batch_size == 1:
             ctx: AbstractContextManager = nullcontext()
@@ -589,7 +599,7 @@ class _DVCFileSystem(AbstractFileSystem):
 
         with ctx:
             it = ((fs, f) for fs, files in _files.items() for f in files)
-            deque(map_fn(get_file, it), maxlen=0)
+            deque(callback.wrap(map_fn(get_file, it)), maxlen=0)
         return result
 
     def get_file(self, rpath, lpath, **kwargs):
@@ -732,9 +742,10 @@ class DVCFileSystem(FileSystem):
         return self.fs.repo_url
 
     def from_os_path(self, path: str) -> str:
-        if os.path.isabs(path):
+        if os.path.isabs(path) or (
+            os.name == "nt" and posixpath.isabs(path) and ntpath.sep not in path
+        ):
             path = os.path.relpath(path, self.repo.root_dir)
-
         return as_posix(path)
 
     def close(self):

dvc/fs/dvc_path.py

@@ -15,7 +15,7 @@ Examples
 
 from urllib.parse import urlsplit
 
-from upath import UPath
+from upath import UPath  # ty: ignore[unresolved-import]
 
 
 class DVCPath(UPath):

dvc/ignore.py

@@ -1,8 +1,9 @@
+import functools
 import os
 import re
-from collections import namedtuple
+from collections.abc import Iterable, Iterator
 from itertools import chain, groupby, takewhile
-from typing import TYPE_CHECKING, Optional
+from typing import TYPE_CHECKING, Any, Literal, NamedTuple, Optional, Union, overload
 
 from pathspec.patterns import GitWildMatchPattern
 from pathspec.util import normalize_file
@@ -13,7 +14,9 @@ from dvc.log import logger
 from dvc.pathspec_math import PatternInfo, merge_patterns
 
 if TYPE_CHECKING:
-    from dvc.fs import AnyFSPath, FileSystem
+    from typing_extensions import Self
+
+    from dvc.fs import FileSystem
 
 logger = logger.getChild(__name__)
 
@@ -21,36 +24,79 @@ logger = logger.getChild(__name__)
 class DvcIgnore:
     DVCIGNORE_FILE = ".dvcignore"
 
-    def __call__(self, root, dirs, files):
+    def __call__(
+        self, root: str, dirs: list[str], files: list[str]
+    ) -> tuple[list[str], list[str]]:
         raise NotImplementedError
 
 
 class DvcIgnorePatterns(DvcIgnore):
-    def __init__(self, pattern_list, dirname, sep):
+    def __init__(
+        self, pattern_list: Iterable[Union[PatternInfo, str]], dirname: str, sep: str
+    ) -> None:
         from pathspec.patterns.gitwildmatch import _DIR_MARK
 
-        if pattern_list and isinstance(pattern_list[0], str):
-            pattern_list = [PatternInfo(pattern, "") for pattern in pattern_list]
-
-        self.sep = sep
-        self.pattern_list = pattern_list
-        self.dirname = dirname
-
-        self.regex_pattern_list = []
-        for count, pattern in enumerate(pattern_list):
-            pattern, group = GitWildMatchPattern.pattern_to_regex(pattern.patterns)
-            if pattern:
-                pattern = pattern.replace(f"<{_DIR_MARK}>", f"<{_DIR_MARK}{count}>")
-                self.regex_pattern_list.append((pattern, group))
-
-        self.ignore_spec = [
-            (ignore, re.compile("|".join(item[0] for item in group)))
-            for ignore, group in groupby(self.regex_pattern_list, lambda x: x[1])
-            if ignore is not None
+        pattern_infos = [
+            pattern if isinstance(pattern, PatternInfo) else PatternInfo(pattern, "")
+            for pattern in pattern_list
         ]
 
+        self.sep = sep
+        self.pattern_list: list[PatternInfo] = []
+        self.dirname = dirname
+        self.find_matching_pattern = functools.cache(self._find_matching_pattern)
+
+        regex_pattern_list: list[tuple[str, bool, bool, PatternInfo]] = []
+        for count, pattern_info in enumerate(pattern_infos):
+            regex, ignore = GitWildMatchPattern.pattern_to_regex(pattern_info.patterns)
+            if regex is not None and ignore is not None:
+                self.pattern_list.append(pattern_info)
+                regex = regex.replace(f"<{_DIR_MARK}>", f"<{_DIR_MARK}{count}>")
+                regex_pattern_list.append(
+                    (regex, ignore, pattern_info.patterns.endswith("/"), pattern_info)
+                )
+
+        def keyfunc(item: tuple[str, bool, bool, PatternInfo]) -> tuple[bool, bool]:
+            _, ignore, dir_only_pattern, _ = item
+            return ignore, dir_only_pattern
+
+        self.ignore_spec: list[
+            tuple[
+                re.Pattern[str],
+                bool,
+                bool,
+                dict[Optional[str], tuple[str, PatternInfo]],
+            ]
+        ]
+        self.ignore_spec = []
+        for (ignore, dir_only_pattern), group in groupby(
+            regex_pattern_list, key=keyfunc
+        ):
+            if ignore:
+                # For performance, we combine all exclude patterns.
+                # But we still need to figure out which pattern matched which rule,
+                # (eg: to show in `dvc check-ignore`).
+                # So, we use named groups and keep a map of group name to pattern.
+                pattern_map: dict[Optional[str], tuple[str, PatternInfo]] = {
+                    f"rule_{i}": (regex, pi)
+                    for i, (regex, _, _, pi) in enumerate(group)
+                }
+                combined_regex = "|".join(
+                    f"(?P<{name}>{regex})" for name, (regex, _) in pattern_map.items()
+                )
+                self.ignore_spec.append(
+                    (re.compile(combined_regex), ignore, dir_only_pattern, pattern_map)
+                )
+            else:
+                # unignored patterns are not combined with `|`.
+                for regex, _, _, pi in group:
+                    pattern_map = {None: (regex, pi)}
+                    self.ignore_spec.append(
+                        (re.compile(regex), ignore, dir_only_pattern, pattern_map)
+                    )
+
     @classmethod
-    def from_file(cls, path, fs, name):
+    def from_file(cls, path: str, fs: "FileSystem", name: str) -> "Self":
         assert fs.isabs(path)
         dirname = fs.normpath(fs.dirname(path))
         with fs.open(path, encoding="utf-8") as fobj:
@@ -62,13 +108,15 @@ class DvcIgnorePatterns(DvcIgnore):
 
         return cls(path_spec_lines, dirname, fs.sep)
 
-    def __call__(self, root: list[str], dirs: list[str], files: list[str]):
+    def __call__(
+        self, root: str, dirs: list[str], files: list[str]
+    ) -> tuple[list[str], list[str]]:
         files = [f for f in files if not self.matches(root, f)]
         dirs = [d for d in dirs if not self.matches(root, d, True)]
 
         return dirs, files
 
-    def _get_normalize_path(self, dirname, basename):
+    def _get_normalize_path(self, dirname: str, basename: str) -> Optional[str]:
         # NOTE: `relpath` is too slow, so we have to assume that both
         # `dirname` and `self.dirname` are relative or absolute together.
 
@@ -81,62 +129,105 @@ class DvcIgnorePatterns(DvcIgnore):
             # NOTE: `os.path.join` is ~x5.5 slower
             path = f"{rel}{self.sep}{basename}"
         else:
-            return False
+            return None
 
         if os.name == "nt":
             return normalize_file(path)
         return path
 
-    def matches(self, dirname, basename, is_dir=False, details: bool = False):
+    @overload
+    def matches(
+        self,
+        dirname: str,
+        basename: str,
+        is_dir: bool = False,
+        details: Literal[False] = ...,
+    ) -> bool: ...
+
+    @overload
+    def matches(
+        self,
+        dirname: str,
+        basename: str,
+        is_dir: bool = False,
+        details: Literal[True] = ...,
+    ) -> tuple[bool, list[PatternInfo]]: ...
+
+    @overload
+    def matches(
+        self,
+        dirname: str,
+        basename: str,
+        is_dir: bool = False,
+        details: bool = False,
+    ) -> Union[bool, tuple[bool, list[PatternInfo]]]: ...
+
+    def matches(
+        self,
+        dirname: str,
+        basename: str,
+        is_dir: bool = False,
+        details: bool = False,
+    ) -> Union[bool, tuple[bool, list[PatternInfo]]]:
         path = self._get_normalize_path(dirname, basename)
-        if not path:
-            return False
-
-        if details:
-            return self._ignore_details(path, is_dir)
-        return self.ignore(path, is_dir)
-
-    def ignore(self, path, is_dir):
-        def matches(pattern, path, is_dir) -> bool:
-            matches_ = bool(pattern.match(path))
-
-            if is_dir:
-                matches_ |= bool(pattern.match(f"{path}/"))
-
-            return matches_
-
         result = False
+        _match: list[PatternInfo] = []
+        if path:
+            result, _match = self._ignore(path, is_dir)
+        return (result, _match) if details else result
 
-        for ignore, pattern in self.ignore_spec[::-1]:
-            if matches(pattern, path, is_dir):
-                result = ignore
-                break
-        return result
+    def _find_matching_pattern(
+        self, path: str, is_dir: bool
+    ) -> tuple[bool, list[PatternInfo]]:
+        paths = [path]
+        if is_dir and not path.endswith("/"):
+            paths.append(f"{path}/")
 
-    def _ignore_details(self, path, is_dir: bool):
-        result = []
-        for (regex, _), pattern_info in list(
-            zip(self.regex_pattern_list, self.pattern_list)
+        for pattern, ignore, dir_only_pattern, pattern_map in reversed(
+            self.ignore_spec
         ):
-            # skip system pattern
-            if not pattern_info.file_info:
+            if dir_only_pattern and not is_dir:
                 continue
+            for p in paths:
+                match = pattern.match(p)
+                if not match:
+                    continue
+                if ignore:
+                    group_name, _match = next(
+                        (
+                            (name, _match)
+                            for name, _match in match.groupdict().items()
+                            if name.startswith("rule_") and _match is not None
+                        )
+                    )
+                else:
+                    # unignored patterns are not combined with `|`,
+                    # so there are no groups.
+                    group_name = None
+                _regex, pattern_info = pattern_map[group_name]
+                return ignore, [pattern_info]
+        return False, []
 
-            regex = re.compile(regex)
+    def _ignore(self, path: str, is_dir: bool) -> tuple[bool, list[PatternInfo]]:
+        parts = path.split("/")
+        result = False
+        matches: list[PatternInfo] = []
+        for i in range(1, len(parts) + 1):
+            rel_path = "/".join(parts[:i])
+            result, _matches = self.find_matching_pattern(
+                rel_path, is_dir or i < len(parts)
+            )
+            if i < len(parts) and not result:
+                continue
+            matches.extend(_matches)
+            if result:
+                break
+        return result, matches
 
-            matches = bool(regex.match(path))
-            if is_dir:
-                matches |= bool(regex.match(f"{path}/"))
-
-            if matches:
-                result.append(pattern_info.file_info)
-
-        return result
-
-    def __hash__(self):
+    def __hash__(self) -> int:
         return hash(self.dirname + ":" + str(self.pattern_list))
 
-    def __eq__(self, other):
+    def __eq__(self, other: object) -> bool:
         if not isinstance(other, DvcIgnorePatterns):
             return NotImplemented
         return (self.dirname == other.dirname) & (
@@ -144,21 +235,18 @@ class DvcIgnorePatterns(DvcIgnore):
             == [pattern.patterns for pattern in other.pattern_list]
         )
 
-    def __bool__(self):
+    def __bool__(self) -> bool:
         return bool(self.pattern_list)
 
 
-CheckIgnoreResult = namedtuple(  # noqa: PYI024
-    "CheckIgnoreResult", ["file", "match", "patterns"]
-)
-
-
-def _no_match(path):
-    return CheckIgnoreResult(path, False, ["::"])
+class CheckIgnoreResult(NamedTuple):
+    file: str
+    match: bool
+    pattern_infos: list[PatternInfo]
 
 
 class DvcIgnoreFilter:
-    def __init__(self, fs, root_dir):
+    def __init__(self, fs: "FileSystem", root_dir: str) -> None:
         from dvc.repo import Repo
 
         default_ignore_patterns = [
@@ -193,7 +281,7 @@ class DvcIgnoreFilter:
             ignore_subrepos=True,
         )
 
-    def _get_key(self, path):
+    def _get_key(self, path: str) -> tuple[str, ...]:
         parts = self.fs.relparts(path, self.root_dir)
         if parts == (os.curdir,):
             return ()
@@ -241,7 +329,7 @@ class DvcIgnoreFilter:
             for dname in dnames:
                 self._update_sub_repo(self.fs.join(dirname, dname), ignore_trie)
 
-    def _update_sub_repo(self, path, ignore_trie: Trie):
+    def _update_sub_repo(self, path: str, ignore_trie: Trie) -> None:
         from dvc.repo import Repo
 
         if path == self.root_dir:
@@ -268,7 +356,9 @@ class DvcIgnoreFilter:
         else:
             ignore_trie[key] = new_pattern
 
-    def __call__(self, root, dirs, files, ignore_subrepos=True):
+    def __call__(
+        self, root: str, dirs: list[str], files: list[str], ignore_subrepos: bool = True
+    ) -> tuple[list[str], list[str]]:
         abs_root = self.fs.abspath(root)
         ignore_pattern = self._get_trie_pattern(
             abs_root, dnames=dirs, ignore_subrepos=ignore_subrepos
@@ -277,7 +367,24 @@ class DvcIgnoreFilter:
             dirs, files = ignore_pattern(abs_root, dirs, files)
         return dirs, files
 
-    def ls(self, fs, path, detail=True, **kwargs):
+    @overload
+    def ls(
+        self, fs: "FileSystem", path: str, detail: Literal[True], **kwargs: Any
+    ) -> list[dict[str, Any]]: ...
+
+    @overload
+    def ls(
+        self, fs: "FileSystem", path: str, detail: Literal[False], **kwargs
+    ) -> list[str]: ...
+
+    @overload
+    def ls(
+        self, fs: "FileSystem", path: str, detail: bool = True, **kwargs
+    ) -> Union[list[str], list[dict[str, Any]]]: ...
+
+    def ls(
+        self, fs: "FileSystem", path: str, detail: bool = True, **kwargs: Any
+    ) -> Union[list[str], list[dict[str, Any]]]:
         fs_dict = {}
         dirs = []
         nondirs = []
@@ -297,22 +404,29 @@ class DvcIgnoreFilter:
 
         return [fs_dict[name] for name in chain(dirs, nondirs)]
 
-    def walk(self, fs: "FileSystem", path: "AnyFSPath", **kwargs):
+    def walk(
+        self, fs: "FileSystem", path: str, **kwargs: Any
+    ) -> Iterator[
+        Union[
+            tuple[str, list[str], list[str]],
+            tuple[str, dict[str, dict], dict[str, dict]],
+        ]
+    ]:
         detail = kwargs.get("detail", False)
         ignore_subrepos = kwargs.pop("ignore_subrepos", True)
         if fs.protocol == Schemes.LOCAL:
             for root, dirs, files in fs.walk(path, **kwargs):
                 if detail:
-                    all_dnames = set(dirs.keys())
-                    all_fnames = set(files.keys())
+                    assert isinstance(dirs, dict)
+                    assert isinstance(files, dict)
                     dnames, fnames = self(
                         root,
-                        all_dnames,
-                        all_fnames,
+                        list(dirs),
+                        list(files),
                         ignore_subrepos=ignore_subrepos,
                     )
-                    list(map(dirs.pop, all_dnames - set(dnames)))
-                    list(map(files.pop, all_fnames - set(fnames)))
+                    list(map(dirs.pop, dirs.keys() - set(dnames)))
+                    list(map(files.pop, files.keys() - set(fnames)))
                 else:
                     dirs[:], files[:] = self(
                         root, dirs, files, ignore_subrepos=ignore_subrepos
@@ -321,7 +435,7 @@ class DvcIgnoreFilter:
         else:
             yield from fs.walk(path, **kwargs)
 
-    def find(self, fs: "FileSystem", path: "AnyFSPath", **kwargs):
+    def find(self, fs: "FileSystem", path: str, **kwargs: Any) -> Iterator[str]:
         if fs.protocol == Schemes.LOCAL:
             for root, _, files in self.walk(fs, path, **kwargs):
                 for file in files:
@@ -331,7 +445,7 @@ class DvcIgnoreFilter:
             yield from fs.find(path)
 
     def _get_trie_pattern(
-        self, dirname, dnames: Optional["list"] = None, ignore_subrepos=True
+        self, dirname: str, dnames: Optional[list[str]] = None, ignore_subrepos=True
     ) -> Optional["DvcIgnorePatterns"]:
         if ignore_subrepos:
             ignores_trie = self.ignores_trie_fs
@@ -367,7 +481,7 @@ class DvcIgnoreFilter:
 
     def _is_ignored(
         self, path: str, is_dir: bool = False, ignore_subrepos: bool = True
-    ):
+    ) -> bool:
         if self._outside_repo(path):
             return False
         dirname, basename = self.fs.split(self.fs.normpath(path))
@@ -389,24 +503,23 @@ class DvcIgnoreFilter:
         path = self.fs.abspath(path)
         return self._is_ignored(path, False, ignore_subrepos=ignore_subrepos)
 
-    def _outside_repo(self, path):
+    def _outside_repo(self, path: str) -> bool:
         return not self.fs.isin_or_eq(path, self.root_dir)
 
-    def check_ignore(self, target):
+    def check_ignore(self, target: str) -> CheckIgnoreResult:
         # NOTE: can only be used in `dvc check-ignore`, see
         # https://github.com/iterative/dvc/issues/5046
         full_target = self.fs.abspath(target)
+        matched_patterns: list[PatternInfo] = []
+        ignore = False
         if not self._outside_repo(full_target):
             dirname, basename = self.fs.split(self.fs.normpath(full_target))
             pattern = self._get_trie_pattern(dirname)
             if pattern:
-                matches = pattern.matches(
-                    dirname, basename, self.fs.isdir(full_target), True
+                ignore, matched_patterns = pattern.matches(
+                    dirname, basename, self.fs.isdir(full_target), details=True
                 )
-
-                if matches:
-                    return CheckIgnoreResult(target, True, matches)
-        return _no_match(target)
+        return CheckIgnoreResult(target, ignore, matched_patterns)
 
     def is_ignored(
         self, fs: "FileSystem", path: str, ignore_subrepos: bool = True
@@ -424,7 +537,7 @@ class DvcIgnoreFilter:
         )
 
 
-def init(path):
+def init(path: Union[str, os.PathLike[str]]) -> str:
     dvcignore = os.path.join(path, DvcIgnore.DVCIGNORE_FILE)
     if os.path.exists(dvcignore):
         return dvcignore
@@ -439,7 +552,7 @@ def init(path):
     return dvcignore
 
 
-def destroy(path):
+def destroy(path: Union[str, os.PathLike[str]]) -> None:
     from dvc.utils.fs import remove
 
     dvcignore = os.path.join(path, DvcIgnore.DVCIGNORE_FILE)

dvc/info.py

@@ -48,7 +48,10 @@ def get_dvc_info():
                 info.append("Cache types: " + error_link("no-dvc-cache"))
 
             info.append(f"Caches: {_get_caches(repo.cache)}")
-            info.append(f"Remotes: {_get_remotes(repo.config)}")
+
+            configured_remotes = _get_remotes(repo.config)
+            remotes = ", ".join(configured_remotes) if configured_remotes else None
+            info.append(f"Remotes: {remotes}")
 
             root_directory = repo.root_dir
             fs_root = _get_fs_type(os.path.abspath(root_directory))
@@ -75,12 +78,10 @@ def _get_caches(cache):
 
 
 def _get_remotes(config):
-    schemes = (
+    return [
         get_fs_cls(get_fs_config(config, name=remote)).protocol
         for remote in config["remote"]
-    )
-
-    return ", ".join(schemes) or "None"
+    ]
 
 
 def _get_linktype_support_info(repo):

dvc/lock.py

@@ -2,13 +2,13 @@
 
 import hashlib
 import os
+import time
 from abc import ABC, abstractmethod
 from datetime import timedelta
 from typing import Optional, Union
 
 import flufl.lock
 import zc.lockfile
-from funcy import retry
 
 from dvc.exceptions import DvcException
 from dvc.progress import Tqdm
@@ -88,9 +88,10 @@ class Lock(LockBase):
     Uses zc.lockfile as backend.
     """
 
-    def __init__(self, lockfile, friendly=False, **kwargs):
+    def __init__(self, lockfile, friendly=False, wait=False, **kwargs):
         super().__init__(lockfile)
         self._friendly = friendly
+        self._wait = wait
         self._lock = None
         self._lock_failed = False
 
@@ -101,23 +102,38 @@ class Lock(LockBase):
     def _do_lock(self):
         try:
             self._lock_failed = False
-            with Tqdm(
-                bar_format="{desc}",
-                disable=not self._friendly,
-                desc="If DVC froze, see `hardlink_lock` in {}".format(
-                    format_link("https://man.dvc.org/config#core")
-                ),
-            ):
-                self._lock = zc.lockfile.LockFile(self._lockfile)
+            self._lock = zc.lockfile.LockFile(self._lockfile)
         except zc.lockfile.LockError:
             self._lock_failed = True
             raise LockError(FAILED_TO_LOCK_MESSAGE)  # noqa: B904
 
     def lock(self):
-        retries = 6
-        delay = DEFAULT_TIMEOUT / retries
-        lock_retry = retry(retries, LockError, timeout=delay)(self._do_lock)
-        lock_retry()
+        """Acquire the lock, either waiting forever, or after default_retries."""
+        default_retries = 6
+        delay = DEFAULT_TIMEOUT / default_retries
+        attempts = 0
+
+        max_retries = float("inf") if self._wait else default_retries
+
+        with Tqdm(
+            bar_format="{desc}",
+            disable=not self._friendly,
+            desc="Waiting to acquire lock. "
+            "If DVC froze, see `hardlink_lock` in {}.".format(
+                format_link("https://man.dvc.org/config#core")
+            ),
+        ) as pbar:
+            while True:
+                try:
+                    self._do_lock()
+                    return
+                except LockError:
+                    attempts += 1
+                    if attempts > max_retries:
+                        raise
+                    time.sleep(delay)
+                finally:
+                    pbar.update()
 
     def unlock(self):
         if self._lock_failed:
@@ -150,7 +166,7 @@ class HardlinkLock(flufl.lock.Lock, LockBase):
         tmp_dir (str): a directory to store claim files.
     """
 
-    def __init__(self, lockfile, tmp_dir=None, **kwargs):
+    def __init__(self, lockfile, tmp_dir=None, wait=False, **kwargs):
         import socket
 
         self._tmp_dir = tmp_dir
@@ -173,11 +189,21 @@ class HardlinkLock(flufl.lock.Lock, LockBase):
         self._separator = flufl.lock.SEP
         self._set_claimfile()
         self._owned = True
+        self._wait = wait
         self._retry_errnos = []
+        self._friendly = kwargs.get("friendly", False)
 
     def lock(self, timeout: Optional[Union[timedelta, int]] = None):
         try:
-            super().lock(timeout or timedelta(seconds=DEFAULT_TIMEOUT))
+            if not self._wait:
+                timeout = timeout or timedelta(seconds=DEFAULT_TIMEOUT)
+
+            with Tqdm(
+                bar_format="{desc}",
+                disable=not (self._wait and self._friendly),
+                desc="Waiting to acquire lock",
+            ):
+                super().lock(timeout)
         except flufl.lock.TimeOutError:
             raise LockError(FAILED_TO_LOCK_MESSAGE)  # noqa: B904
 
@@ -186,11 +212,11 @@ class HardlinkLock(flufl.lock.Lock, LockBase):
 
         if self._tmp_dir is not None:
             # Under Windows file path length is limited so we hash it
-            hasher = hashlib.md5(self._claimfile.encode())  # noqa: S324
+            hasher = hashlib.md5(self._claimfile.encode(), usedforsecurity=False)
             filename = hasher.hexdigest()
             self._claimfile = os.path.join(self._tmp_dir, filename + ".lock")
 
 
-def make_lock(lockfile, tmp_dir=None, friendly=False, hardlink_lock=False):
+def make_lock(lockfile, tmp_dir=None, friendly=False, hardlink_lock=False, wait=False):
     cls = HardlinkLock if hardlink_lock else Lock
-    return cls(lockfile, tmp_dir=tmp_dir, friendly=friendly)
+    return cls(lockfile, tmp_dir=tmp_dir, friendly=friendly, wait=wait)

dvc/logger.py

@@ -5,6 +5,8 @@ import logging.config
 import logging.handlers
 import os
 import sys
+from collections.abc import Iterator
+from contextlib import contextmanager
 from typing import ClassVar
 
 import colorama
@@ -36,7 +38,7 @@ def add_logging_level(level_name, level_num, method_name=None):
             self._log(level_num, message, args, **kwargs)
 
     def log_to_root(message, *args, **kwargs):
-        logging.log(level_num, message, *args, **kwargs)
+        logging.log(level_num, message, *args, **kwargs)  # noqa: LOG015
 
     # getLevelName resolves the numeric log level if already defined,
     # otherwise returns a string
@@ -174,9 +176,19 @@ def _iter_causes(exc):
         exc = exc.__cause__
 
 
-def set_loggers_level(level: int = logging.INFO) -> None:
+@contextmanager
+def set_loggers_level(level: int = logging.INFO) -> Iterator[None]:
+    ret: dict[logging.Logger, int] = {}
     for name in ["dvc", "dvc_objects", "dvc_data"]:
-        logging.getLogger(name).setLevel(level)
+        _logger = logging.getLogger(name)
+        ret[_logger] = _logger.getEffectiveLevel()
+        _logger.setLevel(level)
+
+    try:
+        yield
+    finally:
+        for _logger, old_lvl in ret.items():
+            _logger.setLevel(old_lvl)
 
 
 def setup(level: int = logging.INFO, log_colors: bool = True) -> None:

dvc/output.py

@@ -45,7 +45,7 @@ if TYPE_CHECKING:
     from dvc_data.hashfile.obj import HashFile
     from dvc_data.index import DataIndexKey
 
-    from .ignore import DvcIgnoreFilter
+    from .ignore import CheckIgnoreResult, DvcIgnoreFilter
 
 logger = logger.getChild(__name__)
 
@@ -267,9 +267,9 @@ class OutputIsStageFileError(DvcException):
 
 
 class OutputIsIgnoredError(DvcException):
-    def __init__(self, match):
-        lines = "\n".join(match.patterns)
-        super().__init__(f"Path '{match.file}' is ignored by\n{lines}")
+    def __init__(self, result: "CheckIgnoreResult"):
+        pi = result.pattern_infos[-1]
+        super().__init__(f"Path {result.file!r} is ignored by {pi!s}")
 
 
 class CheckoutCallback(TqdmCallback):
@@ -1203,8 +1203,9 @@ class Output:
         if stage:
             abs_path = os.path.join(stage.wdir, path)
             if self._is_path_dvcignore(abs_path):
-                check = stage.repo.dvcignore.check_ignore(abs_path)
-                raise self.IsIgnoredError(check)
+                check: CheckIgnoreResult = stage.repo.dvcignore.check_ignore(abs_path)
+                if check.match:
+                    raise self.IsIgnoredError(check)
 
     def _check_can_merge(self, out):
         if self.protocol != out.protocol:
@@ -1452,25 +1453,43 @@ class Output:
         self.remote = other.remote
         self.can_push = other.can_push
 
-    def merge_version_meta(self, other: "Output"):
+    def _get_versioned_meta(
+        self,
+    ) -> Optional[
+        tuple["HashInfo", Optional["Meta"], Optional[Union["HashFile", "Tree"]]]
+    ]:
+        if self.files is not None or (
+            self.meta is not None and self.meta.version_id is not None
+        ):
+            old_obj = self.obj if self.obj is not None else self.get_obj()
+            return self.hash_info, self.meta, old_obj
+        return None
+
+    def merge_version_meta(
+        self,
+        old_hi: "HashInfo",
+        old_meta: Optional["Meta"],
+        old_obj: Optional[Union["HashFile", "Tree"]],
+    ):
         """Merge version meta for files which are unchanged from other."""
         if not self.hash_info:
             return
         if self.hash_info.isdir:
-            return self._merge_dir_version_meta(other)
-        if self.hash_info != other.hash_info:
+            return self._merge_dir_version_meta(old_hi, old_obj)
+        if self.hash_info != old_hi:
             return
-        self.meta = other.meta
+        self.meta = old_meta
 
-    def _merge_dir_version_meta(self, other: "Output"):
+    def _merge_dir_version_meta(
+        self, old_hi: "HashInfo", old_obj: Optional[Union["HashFile", "Tree"]]
+    ):
         from dvc_data.hashfile.tree import update_meta
 
-        if not self.obj or not other.hash_info.isdir:
+        if not self.obj or not old_hi.isdir:
             return
-        other_obj = other.obj if other.obj is not None else other.get_obj()
         assert isinstance(self.obj, Tree)
-        assert isinstance(other_obj, Tree)
-        updated = update_meta(self.obj, other_obj)
+        assert isinstance(old_obj, Tree)
+        updated = update_meta(self.obj, old_obj)
         assert updated.hash_info == self.obj.hash_info
         self.obj = updated
         self.files = updated.as_list(with_meta=True)

dvc/parsing/context.py

@@ -105,7 +105,7 @@ class Meta:
 
     @staticmethod
     def update_path(meta: "Meta", path: Union[str, int]):
-        dpaths = meta.dpaths[:] + [str(path)]
+        dpaths = [*meta.dpaths, str(path)]
         return replace(meta, dpaths=dpaths)
 
     def __str__(self):
@@ -155,7 +155,7 @@ class Value(Node):
 PRIMITIVES = (int, float, str, bytes, bool)
 
 
-class Container(Node, ABC):
+class Container(Node, ABC):  # noqa: PLW1641
     meta: Meta
     data: Union[list, dict]
     _key_transform = staticmethod(identity)

dvc/pathspec_math.py

@@ -2,13 +2,19 @@
 # Including changing base dir of path specification patterns and merging
 # of two path specification patterns with different base
 # All the operations follow the documents of `gitignore`
-from collections import namedtuple
+from typing import NamedTuple
 
 from pathspec.util import normalize_file
 
 from dvc.utils import relpath
 
-PatternInfo = namedtuple("PatternInfo", ["patterns", "file_info"])  # noqa: PYI024
+
+class PatternInfo(NamedTuple):
+    patterns: str
+    file_info: str
+
+    def __str__(self) -> str:
+        return self.file_info or f":{self.patterns}"
 
 
 def _not_ignore(rule):
@@ -27,11 +33,9 @@ def _remove_slash(rule):
 
 def _match_all_level(rule):
     if rule[:-1].find("/") >= 0 and not rule.startswith("**/"):
-        if rule.startswith("/"):
-            rule = rule[1:]
+        rule = rule.removeprefix("/")
         return False, rule
-    if rule.startswith("**/"):
-        rule = rule[3:]
+    rule = rule.removeprefix("**/")
     return True, rule
 
 

dvc/repo/__init__.py

@@ -144,6 +144,7 @@ class Repo:
         scm: Optional[Union["Git", "NoSCM"]] = None,
         remote: Optional[str] = None,
         remote_config: Optional["DictStrAny"] = None,
+        _wait_for_lock: bool = False,
     ):
         from dvc.cachemgr import CacheManager
         from dvc.data_cloud import DataCloud
@@ -166,7 +167,8 @@ class Repo:
         self._config = config
         self._remote = remote
         self._remote_config = remote_config
-        self._data_index = None
+        self._data_index: Optional[DataIndex] = None
+        self._wait_for_lock = _wait_for_lock
 
         if rev and not fs:
             self._scm = scm = SCM(root_dir or os.curdir)
@@ -204,6 +206,7 @@ class Repo:
                     tmp_dir=self.tmp_dir,
                     hardlink_lock=self.config["core"].get("hardlink_lock", False),
                     friendly=True,
+                    wait=self._wait_for_lock,
                 )
                 os.makedirs(self.site_cache_dir, exist_ok=True)
                 if not fs and (
@@ -609,7 +612,7 @@ class Repo:
         from dvc.fs import GitFileSystem
         from dvc.version import version_tuple
 
-        cache_dir = self.config["core"].get("site_cache_dir") or site_cache_dir()
+        cache_dir = site_cache_dir(self.config["core"].get("site_cache_dir"))
 
         subdir = None
         if isinstance(self.fs, GitFileSystem):
@@ -638,10 +641,11 @@ class Repo:
         # that just happened to be at the same path as old deleted ones.
         btime = self._btime or getattr(os.stat(root_dir), "st_birthtime", None)
 
-        md5 = hashlib.md5(  # noqa: S324
+        md5 = hashlib.md5(
             str(
                 (root_dir, subdir, btime, getpass.getuser(), version_tuple[0], salt)
-            ).encode()
+            ).encode(),
+            usedforsecurity=False,
         )
         repo_token = md5.hexdigest()
         return os.path.join(repos_dir, repo_token)

dvc/repo/checkout.py

@@ -1,4 +1,5 @@
 import os
+from collections import defaultdict
 from typing import TYPE_CHECKING
 
 from dvc.exceptions import (
@@ -14,7 +15,9 @@ from dvc.utils import relpath
 from . import locked
 
 if TYPE_CHECKING:
-    from dvc_data.index import BaseDataIndex, DataIndexEntry
+    from dvc.repo.index import IndexView
+    from dvc_data.index import BaseDataIndex, DataIndexEntry, DataIndexKey
+    from dvc_data.index.diff import Change
     from dvc_objects.fs.base import FileSystem
 
 logger = logger.getChild(__name__)
@@ -36,10 +39,12 @@ def _remove_unused_links(repo):
     return ret
 
 
-def _build_out_changes(index, changes):
+def _build_out_changes(
+    index: "IndexView", changes: dict["DataIndexKey", "Change"]
+) -> dict["DataIndexKey", tuple[str, dict[str, int]]]:
     from dvc_data.index.checkout import MODIFY
 
-    out_keys = []
+    out_keys: list[DataIndexKey] = []
     for out in index.outs:
         if not out.use_cache:
             continue
@@ -47,19 +52,28 @@ def _build_out_changes(index, changes):
         ws, key = out.index_key
         if ws != "repo":
             continue
-
         out_keys.append(key)
 
-    out_changes = {}
+    out_stats: dict[DataIndexKey, dict[str, int]]
+    out_stats = defaultdict(lambda: defaultdict(int))
+
+    out_changes: dict[DataIndexKey, tuple[str, dict[str, int]]] = {}
     for key, change in changes.items():
+        typ = change.typ
+        isdir = change.new and change.new.isdir
         for out_key in out_keys:
             if len(out_key) > len(key) or key[: len(out_key)] != out_key:
                 continue
 
+            stats = out_stats[out_key]
+            if not isdir:
+                stats[typ] += 1
+
             if key == out_key:
-                out_changes[out_key] = change.typ
-            elif not out_changes.get(out_key):
-                out_changes[out_key] = MODIFY
+                out_changes[out_key] = typ, stats
+            elif out_key not in out_changes:
+                typ = MODIFY
+                out_changes[out_key] = typ, stats
             break
 
     return out_changes
@@ -107,14 +121,13 @@ def checkout(  # noqa: C901
     from dvc.stage.exceptions import StageFileBadNameError, StageFileDoesNotExistError
     from dvc_data.index.checkout import ADD, DELETE, MODIFY, apply, compare
 
-    stats: dict[str, list[str]] = {
-        "added": [],
-        "deleted": [],
-        "modified": [],
-    }
+    stats = {"modified": 0, "added": 0, "deleted": 0}
+    changes: dict[str, list[str]] = {"modified": [], "added": [], "deleted": []}
+
     if not targets:
         targets = [None]
-        stats["deleted"] = _remove_unused_links(self)
+        changes["deleted"] = _remove_unused_links(self)
+        stats["deleted"] = len(changes["deleted"])
 
     if isinstance(targets, str):
         targets = [targets]
@@ -127,8 +140,10 @@ def checkout(  # noqa: C901
             raise CheckoutErrorSuggestGit(target) from exc
         raise  # noqa: PLE0704
 
-    view = self.index.targets_view(
-        targets, recursive=recursive, with_deps=with_deps, onerror=onerror
+    from .index import index_from_targets
+
+    view = index_from_targets(
+        self, targets=targets, recursive=recursive, with_deps=with_deps, onerror=onerror
     )
 
     with ui.progress(unit="entry", desc="Building workspace index", leave=True) as pb:
@@ -149,7 +164,10 @@ def checkout(  # noqa: C901
 
     def checkout_onerror(src_path, dest_path, _exc):
         logger.debug(
-            "failed to create '%s' from '%s'", dest_path, src_path, exc_info=True
+            "failed to create '%s' from '%s'",
+            dest_path,
+            src_path,
+            exc_info=True,  # noqa: LOG014
         )
 
         for out_path in out_paths:
@@ -171,16 +189,26 @@ def checkout(  # noqa: C901
     out_changes = _build_out_changes(view, diff.changes)
 
     typ_map = {ADD: "added", DELETE: "deleted", MODIFY: "modified"}
-    for key, typ in out_changes.items():
+    for key, (typ, _stats) in out_changes.items():
         out_path = self.fs.join(self.root_dir, *key)
 
         if out_path in failed:
             self.fs.remove(out_path, recursive=True)
-        else:
-            self.state.save_link(out_path, self.fs)
-            stats[typ_map[typ]].append(_fspath_dir(out_path))
+            continue
 
+        self.state.save_link(out_path, self.fs)
+        for t, count in _stats.items():
+            stats_typ = typ_map[t]
+            stats[stats_typ] += count
+
+        changes[typ_map[typ]].append(_fspath_dir(out_path))
+
+    for changelist in changes.values():
+        # group directories first, then files. But keep them alphabetically sorted
+        changelist.sort(key=lambda p: (not p.endswith(os.sep), p))
+
+    result = changes | {"stats": stats}
     if failed and not allow_missing:
-        raise CheckoutError([relpath(out_path) for out_path in failed], stats)
-
-    return stats
+        result["failed"] = [relpath(out_path) for out_path in failed]
+        raise CheckoutError([relpath(out_path) for out_path in failed], result)
+    return result

dvc/repo/collect.py

@@ -43,13 +43,13 @@ def _collect_paths(
 
 
 def _filter_outs(
-    outs: Outputs, fs_paths: StrPaths, duplicates=False
+    repo: "Repo", outs: Outputs, fs_paths: StrPaths, duplicates=False
 ) -> tuple[Outputs, StrPaths]:
     res_outs: Outputs = []
     fs_res_paths = fs_paths
 
     for out in outs:
-        fs_path = out.repo.dvcfs.from_os_path(out.fs_path)
+        fs_path = repo.dvcfs.from_os_path(out.fs_path)
         if fs_path in fs_paths:
             res_outs.append(out)
             if not duplicates:
@@ -77,4 +77,4 @@ def collect(
 
     target_paths = _collect_paths(repo, targets, recursive=recursive)
 
-    return _filter_outs(outs, target_paths, duplicates=duplicates)
+    return _filter_outs(repo, outs, target_paths, duplicates=duplicates)

dvc/repo/commit.py

@@ -1,3 +1,4 @@
+from itertools import groupby
 from typing import TYPE_CHECKING
 
 from dvc import prompt
@@ -52,29 +53,38 @@ def commit(
     data_only=False,
     relink=True,
 ):
-    stages_info = [
-        info
-        for info in self.stage.collect_granular(
-            target, with_deps=with_deps, recursive=recursive
-        )
-        if not data_only or info.stage.is_data_source
-    ]
-    for stage_info in stages_info:
-        stage = stage_info.stage
-        if force:
-            stage.save(allow_missing=allow_missing)
-        else:
-            changes = stage.changed_entries()
-            if any(changes):
-                prompt_to_commit(stage, changes, force=force)
+    committed_stages = []
+    groups = groupby(
+        [
+            info
+            for info in self.stage.collect_granular(
+                target, with_deps=with_deps, recursive=recursive
+            )
+            if not data_only or info.stage.is_data_source
+        ],
+        key=lambda info: info.stage.dvcfile,
+    )
+
+    for dvcfile, stages_info_group in groups:
+        to_dump = []
+        for stage_info in stages_info_group:
+            stage = stage_info.stage
+            if force:
                 stage.save(allow_missing=allow_missing)
-        stage.commit(
-            filter_info=stage_info.filter_info,
-            allow_missing=allow_missing,
-            relink=relink,
-        )
-        stage.dump(update_pipeline=False)
-    return [s.stage for s in stages_info]
+            else:
+                changes = stage.changed_entries()
+                if any(changes):
+                    prompt_to_commit(stage, changes, force=force)
+                    stage.save(allow_missing=allow_missing)
+            stage.commit(
+                filter_info=stage_info.filter_info,
+                allow_missing=allow_missing,
+                relink=relink,
+            )
+            to_dump.append(stage)
+        dvcfile.dump_stages(to_dump, update_pipeline=False)
+        committed_stages.extend(to_dump)
+    return committed_stages
 
 
 @locked

dvc/repo/data.py

@@ -1,38 +1,34 @@
 import os
 import posixpath
-from collections.abc import Iterable
-from typing import TYPE_CHECKING, Any, TypedDict, Union
+from collections import defaultdict
+from collections.abc import Iterable, Iterator, Mapping
+from typing import TYPE_CHECKING, Optional, TypedDict, Union
 
-from dvc.fs.callbacks import DEFAULT_CALLBACK
+from dvc.fs.callbacks import DEFAULT_CALLBACK, Callback, TqdmCallback
+from dvc.log import logger
 from dvc.ui import ui
+from dvc_data.index import DataIndexDirError
 
 if TYPE_CHECKING:
-    from dvc.fs.callbacks import Callback
     from dvc.repo import Repo
     from dvc.scm import Git, NoSCM
-    from dvc_data.index import DataIndex
+    from dvc_data.index import (
+        BaseDataIndex,
+        DataIndex,
+        DataIndexEntry,
+        DataIndexKey,
+        DataIndexView,
+    )
     from dvc_data.index.diff import Change
+    from dvc_objects.fs.base import FileSystem
+
+logger = logger.getChild(__name__)
 
 
 def posixpath_to_os_path(path: str) -> str:
     return path.replace(posixpath.sep, os.path.sep)
 
 
-def _adapt_typ(typ: str) -> str:
-    from dvc_data.index.diff import ADD, DELETE, MODIFY
-
-    if typ == MODIFY:
-        return "modified"
-
-    if typ == ADD:
-        return "added"
-
-    if typ == DELETE:
-        return "deleted"
-
-    return typ
-
-
 def _adapt_path(change: "Change") -> str:
     isdir = False
     if change.new and change.new.meta:
@@ -45,27 +41,98 @@ def _adapt_path(change: "Change") -> str:
     return os.path.sep.join(key)
 
 
+def _adapt_path_from_entry(entry: "DataIndexEntry") -> str:
+    key = entry.key
+    assert key
+    if entry.meta and entry.meta.isdir:
+        key = (*key, "")
+    return os.sep.join(key)
+
+
+def _get_missing_paths(
+    to_check: Mapping["FileSystem", Mapping[str, Iterable["DataIndexEntry"]]],
+    batch_size: Optional[int] = None,
+    callback: "Callback" = DEFAULT_CALLBACK,
+) -> Iterator[str]:
+    for fs, paths_map in to_check.items():
+        if batch_size == 1 or (batch_size is None and fs.protocol == "local"):
+            results = list(callback.wrap(map(fs.exists, paths_map)))
+        else:
+            results = fs.exists(
+                list(paths_map), batch_size=batch_size, callback=callback
+            )
+
+        for cache_path, exists in zip(paths_map, results):
+            if exists:
+                continue
+
+            for entry in paths_map[cache_path]:
+                yield _adapt_path_from_entry(entry)
+
+
+class StorageCallback(Callback):
+    def __init__(self, parent_cb: Callback) -> None:
+        super().__init__(size=0, value=0)
+        self.parent_cb = parent_cb
+
+    def set_size(self, size: int) -> None:
+        # This is a no-op to prevent `fs.exists` from trying to set the size
+        pass
+
+    def relative_update(self, value: int = 1) -> None:
+        self.parent_cb.relative_update(value)
+
+    def absolute_update(self, value: int) -> None:
+        self.parent_cb.relative_update(value - self.value)
+
+
+class Rename(TypedDict):
+    old: str
+    new: str
+
+
+class DiffResult(TypedDict, total=False):
+    modified: list[str]
+    added: list[str]
+    deleted: list[str]
+    renamed: list[Rename]
+    unchanged: list[str]
+    unknown: list[str]
+    not_in_cache: list[str]
+
+
 def _diff(
-    old: "DataIndex",
-    new: "DataIndex",
+    old: "BaseDataIndex",
+    new: "BaseDataIndex",
     *,
+    filter_keys: Optional[Iterable["DataIndexKey"]] = None,
     granular: bool = False,
     not_in_cache: bool = False,
-    not_in_remote: bool = False,
-    remote_refresh: bool = False,
+    batch_size: Optional[int] = None,
     callback: "Callback" = DEFAULT_CALLBACK,
-) -> dict[str, list[str]]:
-    from dvc_data.index import StorageError
-    from dvc_data.index.diff import UNCHANGED, UNKNOWN, diff
+    with_renames: bool = False,
+) -> DiffResult:
+    from dvc_data.index.diff import (
+        ADD,
+        DELETE,
+        MODIFY,
+        RENAME,
+        UNCHANGED,
+        UNKNOWN,
+        diff,
+    )
 
-    ret: dict[str, list[str]] = {}
+    ret: DiffResult = defaultdict(list)  # type: ignore[assignment]
+    change_types = {
+        MODIFY: "modified",
+        ADD: "added",
+        DELETE: "deleted",
+        RENAME: "renamed",
+    }
 
-    def _add_change(typ, change):
-        typ = _adapt_typ(typ)
-        if typ not in ret:
-            ret[typ] = []
-
-        ret[typ].append(_adapt_path(change))
+    to_check: dict[FileSystem, dict[str, list[DataIndexEntry]]] = defaultdict(
+        lambda: defaultdict(list)
+    )
 
     for change in diff(
         old,
@@ -74,45 +141,72 @@ def _diff(
         shallow=not granular,
         hash_only=True,
         with_unknown=True,
+        with_renames=with_renames,
         callback=callback,
     ):
+        typ = change.typ
+
+        # The index is a trie, so even when we filter by a specific path
+        # like `dir/file`, all parent nodes leading to that path (e.g., `dir/`)
+        # still appear in the view. As a result, keys like `dir/` will be present
+        # even if only `dir/file` matches the filter.
+        # We need to skip such entries to avoid showing root of tracked directories.
+        if filter_keys:
+            # RENAME does not have a `change.key`
+            if typ == RENAME:
+                assert change.new
+                key = change.new.key
+                # match with "new" key only
+                assert key
+            else:
+                key = change.key
+
+            if not any(key[: len(fk)] == fk for fk in filter_keys):
+                continue
+
         if (
-            change.typ == UNCHANGED
+            typ == UNCHANGED
             and (not change.old or not change.old.hash_info)
             and (not change.new or not change.new.hash_info)
         ):
             # NOTE: emulating previous behaviour
             continue
 
-        if change.typ == UNKNOWN and not change.new:
+        if typ == UNKNOWN and not change.new:
             # NOTE: emulating previous behaviour
             continue
 
-        if (
-            not_in_cache
-            and change.old
-            and change.old.hash_info
-            and not old.storage_map.cache_exists(change.old)
-        ):
-            # NOTE: emulating previous behaviour
-            _add_change("not_in_cache", change)
+        if not_in_cache and change.old and change.old.hash_info:
+            old_entry = change.old
+            cache_fs, cache_path = old.storage_map.get_cache(old_entry)
+            # check later in batches
+            to_check[cache_fs][cache_path].append(old_entry)
 
-        try:
-            if (
-                not_in_remote
-                and change.old
-                and change.old.hash_info
-                and not old.storage_map.remote_exists(
-                    change.old, refresh=remote_refresh
-                )
-            ):
-                _add_change("not_in_remote", change)
-        except StorageError:
-            pass
+        change_typ = change_types.get(typ, typ)
+        if typ == RENAME:
+            assert change.old is not None
+            assert change.new is not None
+            ret["renamed"].append(
+                {
+                    "old": _adapt_path_from_entry(change.old),
+                    "new": _adapt_path_from_entry(change.new),
+                }
+            )
+        else:
+            ret[change_typ].append(_adapt_path(change))  # type: ignore[literal-required]
 
-        _add_change(change.typ, change)
-
-    return ret
+    total_items = sum(
+        len(entries) for paths in to_check.values() for entries in paths.values()
+    )
+    with TqdmCallback(size=total_items, desc="Checking cache", unit="entry") as cb:
+        missing_items = list(
+            _get_missing_paths(
+                to_check, batch_size=batch_size, callback=StorageCallback(cb)
+            ),
+        )
+        if missing_items:
+            ret["not_in_cache"] = missing_items
+    return dict(ret)  # type: ignore[return-value]
 
 
 class GitInfo(TypedDict, total=False):
@@ -151,7 +245,49 @@ def _git_info(scm: Union["Git", "NoSCM"], untracked_files: str = "all") -> GitIn
     )
 
 
-def _diff_index_to_wtree(repo: "Repo", **kwargs: Any) -> dict[str, list[str]]:
+def filter_index(
+    index: Union["DataIndex", "DataIndexView"],
+    filter_keys: Optional[Iterable["DataIndexKey"]] = None,
+) -> "BaseDataIndex":
+    from dvc_data.index.view import DataIndexView
+
+    if not filter_keys:
+        return index
+
+    if isinstance(index, DataIndexView):
+        orig_index = index._index
+        parent_filter_fn = index.filter_fn
+    else:
+        orig_index = index
+        parent_filter_fn = None
+
+    def filter_fn(key: "DataIndexKey") -> bool:
+        if parent_filter_fn is not None and not parent_filter_fn(key):
+            return False
+
+        for filter_key in filter_keys:
+            # eg: if key is "dir/file" and filter_key is "dir/", return True
+            if key[: len(filter_key)] == filter_key:
+                return True
+            # eg: if key is `dir/` and filter_key is `dir/file`, also return True.
+            # This ensures we include parent prefixes needed to reach matching leaves.
+            # Intermediate prefixes must be retained to access nested keys.
+            if filter_key[: len(key)] == key:
+                return True
+        return False
+
+    from dvc_data.index import view
+
+    return view(orig_index, filter_fn=filter_fn)
+
+
+def _diff_index_to_wtree(
+    repo: "Repo",
+    filter_keys: Optional[Iterable["DataIndexKey"]] = None,
+    granular: bool = False,
+    batch_size: Optional[int] = None,
+    with_renames: bool = False,
+) -> DiffResult:
     from .index import build_data_index
 
     with ui.progress(desc="Building workspace index", unit="entry") as pb:
@@ -162,37 +298,69 @@ def _diff_index_to_wtree(repo: "Repo", **kwargs: Any) -> dict[str, list[str]]:
             compute_hash=True,
             callback=pb.as_callback(),
         )
+        workspace_view = filter_index(workspace, filter_keys=filter_keys)
 
     with ui.progress(
         desc="Calculating diff between index/workspace",
         unit="entry",
     ) as pb:
+        index = repo.index.data["repo"]
+        view = filter_index(index, filter_keys=filter_keys)
         return _diff(
-            repo.index.data["repo"],
-            workspace,
+            view,
+            workspace_view,
+            filter_keys=filter_keys,
+            granular=granular,
             not_in_cache=True,
+            with_renames=with_renames,
+            batch_size=batch_size,
             callback=pb.as_callback(),
-            **kwargs,
         )
 
 
 def _diff_head_to_index(
-    repo: "Repo", head: str = "HEAD", **kwargs: Any
-) -> dict[str, list[str]]:
-    index = repo.index.data["repo"]
+    repo: "Repo",
+    head: str = "HEAD",
+    filter_keys: Optional[Iterable["DataIndexKey"]] = None,
+    granular: bool = False,
+    with_renames: bool = False,
+) -> DiffResult:
+    from dvc.exceptions import NotDvcRepoError
+    from dvc.scm import RevError
+    from dvc_data.index import DataIndex
 
-    with repo.switch(head):
-        head_index = repo.index.data["repo"]
+    index = repo.index.data["repo"]
+    index_view = filter_index(index, filter_keys=filter_keys)
+
+    try:
+        with repo.switch(head):
+            head_index = repo.index.data["repo"]
+            head_view = filter_index(head_index, filter_keys=filter_keys)
+    except RevError:
+        logger.debug("failed to switch to '%s'", head)
+        head_view = DataIndex()
+    except NotDvcRepoError as exc:
+        # NOTE: this only gets raised on subdir repos at the moment,
+        # which looks like a bug in `repo.switch`.
+        logger.warning(exc)
+        head_view = DataIndex()
 
     with ui.progress(desc="Calculating diff between head/index", unit="entry") as pb:
-        return _diff(head_index, index, callback=pb.as_callback(), **kwargs)
+        return _diff(
+            head_view,
+            index_view,
+            filter_keys=filter_keys,
+            granular=granular,
+            with_renames=with_renames,
+            callback=pb.as_callback(),
+        )
 
 
 class Status(TypedDict):
     not_in_cache: list[str]
     not_in_remote: list[str]
-    committed: dict[str, list[str]]
-    uncommitted: dict[str, list[str]]
+    committed: DiffResult
+    uncommitted: DiffResult
     untracked: list[str]
     unchanged: list[str]
     git: GitInfo
@@ -217,27 +385,158 @@ def _transform_git_paths_to_dvc(repo: "Repo", files: Iterable[str]) -> list[str]
     return [repo.fs.relpath(file, start) for file in files]
 
 
-def status(repo: "Repo", untracked_files: str = "no", **kwargs: Any) -> Status:
+def _get_entries_not_in_remote(
+    repo: "Repo",
+    filter_keys: Optional[Iterable["DataIndexKey"]] = None,
+    granular: bool = False,
+    remote_refresh: bool = False,
+) -> list[str]:
+    """Get entries that are not in remote storage."""
+    from dvc.repo.worktree import worktree_view
+    from dvc_data.index import StorageKeyError
+
+    entries: dict[DataIndexKey, DataIndexEntry] = {}
+
+    def _onerror(entry, exc):
+        if not isinstance(exc, DataIndexDirError):
+            raise exc
+        # We don't have the contents of this dir file, so we will only check this key.
+        entries[entry.key] = entry
+
+    # View into the index, with only pushable entries
+    index = worktree_view(repo.index, push=True)
+    data_index = index.data["repo"]
+
+    orig_data_index_onerror = data_index.onerror
+    data_index.onerror = _onerror
+
+    view = filter_index(data_index, filter_keys=filter_keys)  # type: ignore[arg-type]
+
+    missing_entries = []
+
+    storage_map = view.storage_map
+
+    n = 0
+    with TqdmCallback(size=n, desc="Checking remote", unit="entry") as cb:
+        for key, entry in view.iteritems(shallow=not granular):
+            if not (entry and entry.hash_info):
+                continue
+
+            # The index is a trie, so even when we filter by a specific path
+            # like `dir/file`, all parent nodes leading to that path (e.g., `dir/`)
+            # still appear in the view. As a result, keys like `dir/` will be present
+            # even if only `dir/file` matches the filter.
+            # We need to skip such entries to avoid showing root of tracked directories.
+            if filter_keys and not any(
+                key[: len(filter_key)] == filter_key for filter_key in filter_keys
+            ):
+                continue
+
+            entries[key] = entry
+            n += 1
+            cb.set_size(n)
+
+        for key, entry in entries.items():
+            k = (*key, "") if entry.meta and entry.meta.isdir else key
+            try:
+                if not storage_map.remote_exists(entry, refresh=remote_refresh):
+                    missing_entries.append(os.path.sep.join(k))
+            except StorageKeyError:
+                pass
+            finally:
+                cb.relative_update()
+    data_index.onerror = orig_data_index_onerror
+    return missing_entries
+
+
+def _matches_target(p: str, targets: Iterable[str]) -> bool:
+    sep = os.sep
+    return any(p == t or p.startswith(t + sep) for t in targets)
+
+
+def _prune_keys(filter_keys: Iterable["DataIndexKey"]) -> list["DataIndexKey"]:
+    sorted_keys = sorted(set(filter_keys), key=len)
+    result: list[DataIndexKey] = []
+
+    for key in sorted_keys:
+        if not any(key[: len(prefix)] == prefix for prefix in result):
+            result.append(key)
+    return result
+
+
+def status(  # noqa: PLR0913
+    repo: "Repo",
+    targets: Optional[Iterable[Union[os.PathLike[str], str]]] = None,
+    *,
+    granular: bool = False,
+    untracked_files: str = "no",
+    remote: Optional[str] = None,
+    not_in_remote: bool = False,
+    remote_refresh: bool = False,
+    config: Optional[dict] = None,
+    batch_size: Optional[int] = None,
+    head: str = "HEAD",
+    with_renames: bool = True,
+) -> Status:
     from dvc.scm import NoSCMError, SCMError
 
-    head = kwargs.pop("head", "HEAD")
-    uncommitted_diff = _diff_index_to_wtree(repo, **kwargs)
+    config = config or {}
+    if remote and not_in_remote:
+        logger.debug("Using remote %r", remote)
+        core = config.setdefault("core", {})
+        core["remote"] = remote
+    repo.config.merge(config)
+
+    targets = targets or []
+    filter_keys: list[DataIndexKey] = [repo.fs.relparts(os.fspath(t)) for t in targets]
+    # try to remove duplicate and overlapping keys
+    filter_keys = _prune_keys(filter_keys)
+
+    uncommitted_diff = _diff_index_to_wtree(
+        repo,
+        filter_keys=filter_keys,
+        granular=granular,
+        batch_size=batch_size,
+        with_renames=with_renames,
+    )
     unchanged = set(uncommitted_diff.pop("unchanged", []))
 
+    entries_not_in_remote: list[str] = []
+    if not_in_remote:
+        entries_not_in_remote = _get_entries_not_in_remote(
+            repo,
+            filter_keys=filter_keys,
+            granular=granular,
+            remote_refresh=remote_refresh,
+        )
+
     try:
-        committed_diff = _diff_head_to_index(repo, head=head, **kwargs)
+        committed_diff = _diff_head_to_index(
+            repo,
+            filter_keys=filter_keys,
+            head=head,
+            granular=granular,
+            with_renames=with_renames,
+        )
     except (SCMError, NoSCMError):
         committed_diff = {}
     else:
         unchanged &= set(committed_diff.pop("unchanged", []))
 
     git_info = _git_info(repo.scm, untracked_files=untracked_files)
-    untracked = git_info.get("untracked", [])
-    untracked = _transform_git_paths_to_dvc(repo, untracked)
+    scm_filter_targets = {
+        os.path.relpath(os.path.abspath(t), repo.scm.root_dir) for t in targets
+    }
+    untracked_it: Iterable[str] = git_info.get("untracked", [])
+    if scm_filter_targets:
+        untracked_it = (
+            f for f in untracked_it if _matches_target(f, scm_filter_targets)
+        )
+    untracked = _transform_git_paths_to_dvc(repo, untracked_it)
     # order matters here
     return Status(
         not_in_cache=uncommitted_diff.pop("not_in_cache", []),
-        not_in_remote=uncommitted_diff.pop("not_in_remote", []),
+        not_in_remote=entries_not_in_remote,
         committed=committed_diff,
         uncommitted=uncommitted_diff,
         untracked=untracked,

dvc/repo/diff.py

@@ -45,6 +45,12 @@ def _diff(old, new, data_keys, with_missing=False):
         with_renames=True,
         meta_cmp_key=meta_cmp_key,
         roots=data_keys,
+        # Include unknown entries from missing dir entry, so that they don't
+        # get reported as added/modified/deleted.
+        # Also return unchanged entries so that we can check if they are missing
+        # from cache.
+        with_unknown=True,
+        with_unchanged=with_missing,
     ):
         if (change.old and change.old.isdir and not change.old.hash_info) or (
             change.new and change.new.isdir and not change.new.hash_info

dvc/repo/experiments/collect.py

@@ -274,6 +274,7 @@ def collect(
     num: int = 1,
     hide_queued: bool = False,
     hide_failed: bool = False,
+    hide_workspace: bool = False,
     sha_only: bool = False,
     **kwargs,
 ) -> list["ExpState"]:
@@ -303,8 +304,11 @@ def collect(
             repo.scm, baseline_revs, refs=cached_refs, logger=logger
         )
 
-    workspace_data = collect_rev(repo, "workspace", **kwargs)
-    result: list[ExpState] = [workspace_data]
+    result: list[ExpState] = []
+    if not hide_workspace:
+        workspace_data = collect_rev(repo, "workspace", **kwargs)
+        result.append(workspace_data)
+
     queued = collect_queued(repo, baseline_revs, **kwargs) if not hide_queued else {}
     active = collect_active(repo, baseline_revs, **kwargs)
     failed = collect_failed(repo, baseline_revs, **kwargs) if not hide_failed else {}

dvc/repo/experiments/executor/base.py

@@ -1,20 +1,21 @@
-import logging
 import os
 import pickle
 import shutil
 from abc import ABC, abstractmethod
 from collections.abc import Iterable, Iterator
-from contextlib import contextmanager
+from contextlib import AbstractContextManager, contextmanager
 from dataclasses import asdict, dataclass
 from enum import IntEnum
 from itertools import chain
 from typing import TYPE_CHECKING, Any, Callable, NamedTuple, Optional, Union
 
+from funcy import nullcontext
 from scmrepo.exceptions import SCMError
 
 from dvc.env import DVC_EXP_AUTO_PUSH, DVC_EXP_GIT_REMOTE
 from dvc.exceptions import DvcException
 from dvc.log import logger
+from dvc.logger import set_loggers_level
 from dvc.repo.experiments.exceptions import ExperimentExistsError
 from dvc.repo.experiments.refs import EXEC_BASELINE, EXEC_BRANCH, ExpRefInfo
 from dvc.repo.experiments.utils import to_studio_params
@@ -297,7 +298,9 @@ class BaseExecutor(ABC):
                 stages = dvc.commit([], recursive=recursive, force=True, relink=False)
             exp_hash = cls.hash_exp(stages)
             if include_untracked:
-                dvc.scm.add(include_untracked, force=True)  # type: ignore[call-arg]
+                from dvc.scm import add_no_submodules
+
+                add_no_submodules(dvc.scm, include_untracked, force=True)  # type: ignore[call-arg]
 
             with cls.auto_push(dvc):
                 cls.commit(
@@ -466,8 +469,8 @@ class BaseExecutor(ABC):
 
         if queue is not None:
             queue.put((rev, os.getpid()))
-        if log_errors and log_level is not None:
-            cls._set_log_level(log_level)
+
+        log_ctx = cls._set_log_level(log_level) if log_errors else nullcontext()
 
         exp_hash: Optional[str] = None
         exp_ref: Optional[ExpRefInfo] = None
@@ -476,14 +479,17 @@ class BaseExecutor(ABC):
         if info.name:
             ui.write(f"Reproducing experiment '{info.name}'")
 
-        with cls._repro_dvc(
-            info,
-            infofile,
-            log_errors=log_errors,
-            copy_paths=copy_paths,
-            message=message,
-            **kwargs,
-        ) as dvc:
+        with (
+            log_ctx,
+            cls._repro_dvc(
+                info,
+                infofile,
+                log_errors=log_errors,
+                copy_paths=copy_paths,
+                message=message,
+                **kwargs,
+            ) as dvc,
+        ):
             args, kwargs = cls._repro_args(dvc)
             if args:
                 targets: Optional[Union[list, str]] = args[0]
@@ -509,7 +515,9 @@ class BaseExecutor(ABC):
             stages = dvc.reproduce(*args, **kwargs)
             if paths := cls._get_top_level_paths(dvc):
                 logger.debug("Staging top-level files: %s", paths)
-                dvc.scm_context.add(paths)
+                from dvc.scm import add_no_submodules
+
+                add_no_submodules(dvc.scm, paths)
 
             exp_hash = cls.hash_exp(stages)
             if not repro_dry:
@@ -786,13 +794,10 @@ class BaseExecutor(ABC):
         return orig_rev
 
     @staticmethod
-    def _set_log_level(level):
-        # When executor.reproduce is run in a multiprocessing child process,
-        # dvc.cli.main will not be called for that child process so we need to
-        # setup logging ourselves
-        dvc_logger = logging.getLogger("dvc")
+    def _set_log_level(level: Optional[int]) -> AbstractContextManager[None]:
         if level is not None:
-            dvc_logger.setLevel(level)
+            return set_loggers_level(level)
+        return nullcontext()
 
     @staticmethod
     def _copy_path(src, dst):

dvc/repo/experiments/queue/base.py

@@ -38,7 +38,7 @@ logger = logger.getChild(__name__)
 
 
 @dataclass(frozen=True)
-class QueueEntry:
+class QueueEntry:  # noqa: PLW1641
     dvc_root: str
     scm_root: str
     stash_ref: str
@@ -372,7 +372,7 @@ class BaseStashQueue(ABC):
         )
 
     def _stash_commit_deps(self, *args, **kwargs):
-        if len(args):
+        if args:
             targets = args[0]
         else:
             targets = kwargs.get("targets")

dvc/repo/experiments/queue/celery.py

@@ -1,3 +1,4 @@
+import glob
 import hashlib
 import locale
 import logging
@@ -434,9 +435,13 @@ class LocalCeleryQueue(BaseStashQueue):
         try:
             proc_info = self.proc[queue_entry.stash_rev]
         except KeyError:
-            raise DvcException(  # noqa: B904
-                f"No output logs found for experiment '{rev}'"
-            )
+            message = f"No output logs found for experiment '{rev}'"
+            if self.match_queue_entry_by_name({rev}, self.iter_failed()).get(rev):
+                message += "\nExperiment likely failed during setup."
+                if celery_logs := glob.glob(os.path.join(self.wdir, "*.out")):
+                    message += " Check the celery logs for more details:"
+                    message += f"\n\tcat {' '.join(celery_logs)}".expandtabs(4)
+            raise DvcException(message)  # noqa: B904
         with open(
             proc_info.stdout, encoding=encoding or locale.getpreferredencoding()
         ) as fobj:

dvc/repo/experiments/run.py

@@ -93,7 +93,7 @@ def run(  # noqa: C901, PLR0912
 
     for idx, sweep_overrides in enumerate(sweeps):
         if hydra_sweep and name_prefix is not None:
-            kwargs["name"] = f"{name_prefix}-{idx+1}"
+            kwargs["name"] = f"{name_prefix}-{idx + 1}"
         queue_entry = repo.experiments.queue_one(
             repo.experiments.celery_queue,
             targets=targets,

dvc/repo/experiments/show.py

@@ -30,6 +30,7 @@ def show(
     num: int = 1,
     hide_queued: bool = False,
     hide_failed: bool = False,
+    hide_workspace: bool = False,
     sha_only: bool = False,
     **kwargs,
 ) -> list["ExpState"]:
@@ -42,6 +43,7 @@ def show(
         num=num,
         hide_queued=hide_queued,
         hide_failed=hide_failed,
+        hide_workspace=hide_workspace,
         sha_only=sha_only,
         **kwargs,
     )

dvc/repo/fetch.py

@@ -10,6 +10,7 @@ from . import locked
 
 if TYPE_CHECKING:
     from dvc.output import Output
+    from dvc.repo import Repo
     from dvc.stage import Stage
 
 logger = logger.getChild(__name__)
@@ -24,7 +25,7 @@ def _make_index_onerror(onerror, rev):
 
 
 def _collect_indexes(  # noqa: PLR0913
-    repo,
+    repo: "Repo",
     targets=None,
     remote=None,
     all_branches=False,
@@ -40,6 +41,8 @@ def _collect_indexes(  # noqa: PLR0913
     onerror=None,
     push=False,
 ):
+    from .index import index_from_targets
+
     indexes = {}
     collection_exc = None
 
@@ -67,7 +70,8 @@ def _collect_indexes(  # noqa: PLR0913
         try:
             repo.config.merge(config)
 
-            idx = repo.index.targets_view(
+            idx = index_from_targets(
+                repo,
                 targets,
                 with_deps=with_deps,
                 recursive=recursive,
@@ -94,7 +98,7 @@ def _collect_indexes(  # noqa: PLR0913
 
 @locked
 def fetch(  # noqa: PLR0913
-    self,
+    self: "Repo",
     targets=None,
     jobs=None,
     remote=None,

dvc/repo/index.py

@@ -17,6 +17,7 @@ from dvc.utils.objects import cached_property
 if TYPE_CHECKING:
     from networkx import DiGraph
     from pygtrie import Trie
+    from typing_extensions import Self
 
     from dvc.dependency import Dependency
     from dvc.fs.callbacks import Callback
@@ -66,14 +67,14 @@ def collect_files(
 
     is_local_fs = isinstance(fs, LocalFileSystem)
 
-    def is_ignored(path):
+    def is_ignored(path: str) -> bool:
         # apply only for the local fs
         return is_local_fs and scm.is_ignored(path)
 
-    def is_dvcfile_and_not_ignored(root, file):
+    def is_dvcfile_and_not_ignored(root: str, file: str) -> bool:
         return is_valid_filename(file) and not is_ignored(f"{root}{sep}{file}")
 
-    def is_out_or_ignored(root, directory):
+    def is_out_or_ignored(root: str, directory: str) -> bool:
         dir_path = f"{root}{sep}{directory}"
         # trailing slash needed to check if a directory is gitignored
         return dir_path in outs or is_ignored(f"{dir_path}{sep}")
@@ -83,6 +84,8 @@ def collect_files(
         walk_iter = log_walk(walk_iter)
 
     for root, dirs, files in walk_iter:
+        assert isinstance(dirs, list)
+        assert isinstance(files, list)
         dvcfile_filter = partial(is_dvcfile_and_not_ignored, root)
         for file in filter(dvcfile_filter, files):
             file_path = fs.join(root, file)
@@ -318,32 +321,9 @@ class Index:
         repo: "Repo",
         onerror: Optional[Callable[[str, Exception], None]] = None,
     ) -> "Index":
-        stages = []
-        metrics = {}
-        plots = {}
-        params = {}
-        artifacts = {}
-        datasets = {}
-        datasets_lock = {}
-
         onerror = onerror or repo.stage_collection_error_handler
-        for _, idx in collect_files(repo, onerror=onerror):
-            stages.extend(idx.stages)
-            metrics.update(idx._metrics)
-            plots.update(idx._plots)
-            params.update(idx._params)
-            artifacts.update(idx._artifacts)
-            datasets.update(idx._datasets)
-            datasets_lock.update(idx._datasets_lock)
-        return cls(
-            repo,
-            stages=stages,
-            metrics=metrics,
-            plots=plots,
-            params=params,
-            artifacts=artifacts,
-            datasets=datasets,
-            datasets_lock=datasets_lock,
+        return cls.from_indexes(
+            repo, (idx for _, idx in collect_files(repo, onerror=onerror))
         )
 
     @classmethod
@@ -364,7 +344,7 @@ class Index:
             else {},
         )
 
-    def update(self, stages: Iterable["Stage"]) -> "Index":
+    def update(self, stages: Iterable["Stage"]) -> "Self":
         stages = set(stages)
         # we remove existing stages with same hashes at first
         # and then re-add the new ones later.
@@ -379,6 +359,36 @@ class Index:
             datasets=self._datasets,
         )
 
+    @classmethod
+    def from_indexes(cls, repo, idxs: Iterable["Self"]) -> "Self":
+        stages = []
+        metrics = {}
+        plots = {}
+        params = {}
+        artifacts = {}
+        datasets = {}
+        datasets_lock = {}
+
+        for idx in idxs:
+            stages.extend(idx.stages)
+            metrics.update(idx._metrics)
+            plots.update(idx._plots)
+            params.update(idx._params)
+            artifacts.update(idx._artifacts)
+            datasets.update(idx._datasets)
+            datasets_lock.update(idx._datasets_lock)
+
+        return cls(
+            repo,
+            stages=stages,
+            metrics=metrics,
+            plots=plots,
+            params=params,
+            artifacts=artifacts,
+            datasets=datasets,
+            datasets_lock=datasets_lock,
+        )
+
     @cached_property
     def outs_trie(self) -> "Trie":
         from dvc.repo.trie import build_outs_trie
@@ -735,6 +745,10 @@ class IndexView:
         for stage in self.stages:
             yield from stage.deps
 
+    @property
+    def index(self) -> "Index":
+        return self._index
+
     @property
     def _filtered_outs(self) -> Iterator[tuple["Output", Optional[str]]]:
         for stage, filter_info in self._stage_infos:
@@ -927,3 +941,51 @@ def _get_entry_hash_name(
             return src_entry.hash_info.name
 
     return DEFAULT_ALGORITHM
+
+
+def index_from_targets(
+    repo: "Repo",
+    targets: Optional["TargetType"] = None,
+    stage_filter: Optional[Callable[["Stage"], bool]] = None,
+    outs_filter: Optional[Callable[["Output"], bool]] = None,
+    max_size: Optional[int] = None,
+    types: Optional[list[str]] = None,
+    with_deps: bool = False,
+    recursive: bool = False,
+    **kwargs: Any,
+) -> "IndexView":
+    from dvc.stage.exceptions import StageFileDoesNotExistError, StageNotFound
+    from dvc.utils import parse_target
+
+    index: Optional[Index] = None
+    if targets and all(targets) and not with_deps and not recursive:
+        indexes: list[Index] = []
+        try:
+            for target in targets:
+                if not target:
+                    continue
+                file, name = parse_target(target)
+                if file and not name:
+                    index = Index.from_file(repo, file)
+                else:
+                    stages = repo.stage.collect(target)
+                    index = Index(repo, stages=list(stages))
+                indexes.append(index)
+        except (StageFileDoesNotExistError, StageNotFound):
+            pass
+        else:
+            index = Index.from_indexes(repo, indexes)
+            targets = None
+
+    if index is None:
+        index = repo.index
+    return index.targets_view(
+        targets,
+        stage_filter=stage_filter,
+        outs_filter=outs_filter,
+        max_size=max_size,
+        types=types,
+        recursive=recursive,
+        with_deps=with_deps,
+        **kwargs,
+    )

dvc/repo/init.py

@@ -63,7 +63,7 @@ def init(root_dir=os.curdir, no_scm=False, force=False, subdir=False):  # noqa:
 
         remove(dvc_dir)
 
-    os.mkdir(dvc_dir)
+    os.makedirs(dvc_dir, exist_ok=True)
 
     config = Config.init(dvc_dir)
 

dvc/repo/ls.py

@@ -99,7 +99,9 @@ def ls_tree(
         path = path or ""
         fs: DVCFileSystem = repo.dvcfs
         fs_path = fs.from_os_path(path)
-        return _ls_tree(fs, fs_path, dvc_only, maxdepth)
+        return _ls_tree(
+            fs, fs_path, maxdepth=maxdepth, dvc_only=dvc_only, dvcfiles=True
+        )
 
 
 def _ls(
@@ -145,27 +147,34 @@ def _ls(
     return ret_list
 
 
-def _ls_tree(
-    fs, path, dvc_only: bool = False, maxdepth: Optional[int] = None, _info=None
-):
-    ret = {}
+def _ls_tree(fs, path, maxdepth=None, _info=None, **fs_kwargs):
     info = _info or fs.info(path)
+    if _info is None:
+        # preserve the original path name
+        name = path
+        if not name:
+            name = os.curdir if fs.protocol == "local" else fs.root_marker
+        path = info["name"]
+    else:
+        name = path.rsplit(fs.sep, 1)[-1]
 
-    path = info["name"].rstrip(fs.sep) or os.curdir
-    name = path.rsplit("/", 1)[-1]
+    ret = {}
     ls_info = _adapt_info(info)
     ls_info["path"] = path
 
     recurse = maxdepth is None or maxdepth > 0
     if recurse and info["type"] == "directory":
-        infos = fs.ls(path, dvcfiles=True, dvc_only=dvc_only, detail=True)
+        try:
+            infos = fs.ls(path, detail=True, **fs_kwargs)
+        except FileNotFoundError:
+            # broken symlink?
+            infos = []
+
         infos.sort(key=lambda f: f["name"])
         maxdepth = maxdepth - 1 if maxdepth is not None else None
         contents = {}
         for info in infos:
-            d = _ls_tree(
-                fs, info["name"], dvc_only=dvc_only, maxdepth=maxdepth, _info=info
-            )
+            d = _ls_tree(fs, info["name"], maxdepth=maxdepth, _info=info, **fs_kwargs)
             contents.update(d)
         ls_info["contents"] = contents
 

dvc/repo/ls_url.py

@@ -1,19 +1,30 @@
+from fsspec.implementations.local import LocalFileSystem as _LocalFileSystem
+
 from dvc.exceptions import URLMissingError
-from dvc.fs import parse_external_url
+from dvc.fs import LocalFileSystem, parse_external_url
 
 
-def ls_url(url, *, fs_config=None, recursive=False, config=None):
+def ls_url(url, *, fs_config=None, recursive=False, maxdepth=None, config=None):
     fs, fs_path = parse_external_url(url, fs_config=fs_config, config=config)
     try:
         info = fs.info(fs_path)
     except FileNotFoundError as exc:
         raise URLMissingError(url) from exc
-    if info["type"] != "directory":
+    if maxdepth == 0 or info["type"] != "directory":
         return [{"path": info["name"], "isdir": False}]
 
+    if isinstance(fs, LocalFileSystem):
+        # dvc's LocalFileSystem does not support maxdepth yet
+        walk = _LocalFileSystem().walk
+    else:
+        walk = fs.walk
+
     ret = []
-    for _, dirs, files in fs.walk(fs_path, detail=True):
-        if not recursive:
+    for root, dirs, files in walk(fs_path, detail=True, maxdepth=maxdepth):
+        parts = fs.relparts(root, fs_path)
+        if parts == (".",):
+            parts = ()
+        if not recursive or (maxdepth and len(parts) >= maxdepth - 1):
             files.update(dirs)
 
         for info in files.values():

dvc/repo/open_repo.py

@@ -99,7 +99,7 @@ def clean_repos():
 
 def _get_remote_config(url):
     try:
-        repo = Repo(url)
+        repo = Repo(url, uninitialized=True)
     except NotDvcRepoError:
         return {}
 
@@ -109,10 +109,16 @@ def _get_remote_config(url):
             # Fill the empty upstream entry with a new remote pointing to the
             # original repo's cache location.
             name = "auto-generated-upstream"
-            return {
-                "core": {"remote": name},
-                "remote": {name: {"url": repo.cache.local_cache_dir}},
-            }
+            try:
+                local_cache_dir = repo.cache.local_cache_dir
+            except AttributeError:
+                # if the `.dvc` dir is missing, we get an AttributeError
+                return {}
+            else:
+                return {
+                    "core": {"remote": name},
+                    "remote": {name: {"url": local_cache_dir}},
+                }
 
         # Use original remote to make sure that we are using correct url,
         # credential paths, etc if they are relative to the config location.

dvc/repo/plots/__init__.py

@@ -32,7 +32,7 @@ logger = logger.getChild(__name__)
 
 
 def onerror_collect(result: dict, exception: Exception, *args, **kwargs):
-    logger.debug("", exc_info=True)
+    logger.debug("", exc_info=True)  # noqa: LOG014
     result["error"] = exception
 
 
@@ -346,14 +346,14 @@ def _get_data_targets(definitions: dict):
 
 
 def infer_data_sources(plot_id, config=None):
-    y = config.get("y", None)
+    y = config.get("y", None) if config else None
 
     if isinstance(y, dict):
         sources = list(y.keys())
     else:
         sources = [plot_id]
 
-    x = config.get("x", None)
+    x = config.get("x", None) if config else None
     if isinstance(x, dict):
         sources.append(first(x.keys()))
 
@@ -507,7 +507,7 @@ def _collect_pipeline_files(repo, targets: list[str], props, onerror=None):
 @error_handler
 def _collect_definitions(
     repo: "Repo",
-    targets=None,
+    targets: list[str],
     props: Optional[dict] = None,
     onerror: Optional[Callable] = None,
     **kwargs,

dvc/repo/pull.py

@@ -39,7 +39,7 @@ def pull(  # noqa: PLR0913
         run_cache=run_cache,
     )
     try:
-        stats = self.checkout(
+        result = self.checkout(
             targets=expanded_targets,
             with_deps=with_deps,
             force=force,
@@ -47,8 +47,10 @@ def pull(  # noqa: PLR0913
             allow_missing=allow_missing,
         )
     except CheckoutError as exc:
-        exc.stats["fetched"] = processed_files_count
+        # put fetched counts first
+        exc.result["stats"] = {"fetched": processed_files_count} | exc.result["stats"]
         raise
-
-    stats["fetched"] = processed_files_count
-    return stats
+    else:
+        # put fetched counts first
+        result["stats"] = {"fetched": processed_files_count} | result["stats"]
+    return result

dvc/repo/push.py

@@ -152,10 +152,12 @@ def push(  # noqa: PLR0913
     finally:
         ws_idx = indexes.get("workspace")
         if ws_idx is not None:
+            from dvc.repo.index import IndexView
+
+            _index = ws_idx.index if isinstance(ws_idx, IndexView) else ws_idx
             _update_meta(
-                self.index,
+                _index,
                 targets=glob_targets(targets, glob=glob),
-                remote=remote,
                 with_deps=with_deps,
                 recursive=recursive,
             )

dvc/repo/scm_context.py

@@ -41,8 +41,10 @@ class SCMContext:
     def add(self, paths: Union[str, Iterable[str]]) -> None:
         from scmrepo.exceptions import UnsupportedIndexFormat
 
+        from dvc.scm import add_no_submodules
+
         try:
-            return self.scm.add(paths)
+            add_no_submodules(self.scm, paths)
         except UnsupportedIndexFormat:
             link = "https://github.com/iterative/dvc/issues/610"
             add_cmd = self._make_git_add_cmd([relpath(path) for path in paths])

dvc/repo/status.py

@@ -130,7 +130,7 @@ def status(  # noqa: PLR0913
             remote=remote,
             all_tags=all_tags,
             all_commits=all_commits,
-            recursive=True,
+            recursive=recursive,
         )
 
     ignored = list(

dvc/repo/worktree.py

@@ -152,7 +152,7 @@ def _merge_push_meta(  # noqa: C901
     else:
         if entry.hash_info:
             out.hash_info = entry.hash_info
-        if out.meta.version_id is None:
+        if out.meta is None or out.meta.version_id is None:
             out.meta = entry.meta
     if out.meta:
         out.meta.remote = remote
@@ -181,7 +181,7 @@ def _update_worktree_out(
 ):
     from dvc_data.index import build
 
-    remote_name = out.remote or out.meta.remote
+    remote_name = out.remote or (out.meta.remote if out.meta is not None else None)
     if not remote_name:
         logger.warning("Could not update '%s', it was never pushed to a remote", out)
         return
@@ -212,11 +212,12 @@ def _update_worktree_out(
         logger.warning("Could not update '%s', directory is empty in the remote", out)
         return
 
-    _fetch_out_changes(out, local_index, remote_index, remote)
+    _fetch_out_changes(repo, out, local_index, remote_index, remote)
     _update_out_meta(repo, out, local_index, remote_index, remote)
 
 
 def _fetch_out_changes(
+    repo: "Repo",
     out: "Output",
     local_index: Union["DataIndex", "DataIndexView"],
     remote_index: Union["DataIndex", "DataIndexView"],
@@ -242,7 +243,7 @@ def _fetch_out_changes(
         cb.set_size(total)
         apply(
             diff,
-            out.repo.root_dir,
+            repo.root_dir,
             out.fs,
             update_meta=False,
             storage="data",

dvc/scm.py

@@ -1,17 +1,18 @@
 """Manages source control systems (e.g. Git)."""
 
 import os
-from collections.abc import Iterator, Mapping
+from collections.abc import Iterable, Iterator, Mapping
 from contextlib import contextmanager
 from functools import partial
 from typing import TYPE_CHECKING, Literal, Optional, Union, overload
 
 from funcy import group_by
-from scmrepo.base import Base  # noqa: F401
+from scmrepo.base import Base  # noqa: TC002
 from scmrepo.git import Git
 from scmrepo.noscm import NoSCM
 
 from dvc.exceptions import DvcException
+from dvc.log import logger
 from dvc.progress import Tqdm
 
 if TYPE_CHECKING:
@@ -19,6 +20,8 @@ if TYPE_CHECKING:
 
     from dvc.fs import FileSystem
 
+logger = logger.getChild(__name__)
+
 
 class SCMError(DvcException):
     """Base class for source control management errors."""
@@ -283,3 +286,35 @@ def lfs_prefetch(fs: "FileSystem", paths: list[str]):
             include=[(path if path.startswith("/") else f"/{path}") for path in paths],
             progress=pbar.update_git,
         )
+
+
+def add_no_submodules(
+    scm: "Base",
+    paths: Union[str, Iterable[str]],
+    **kwargs,
+) -> None:
+    """Stage paths to Git, excluding those inside submodules."""
+
+    if isinstance(paths, str):
+        paths = [paths]
+
+    submodule_roots = {os.path.join(scm.root_dir, sub) for sub in scm.list_submodules()}
+
+    repo_paths: list[str] = []
+    skipped_paths: list[str] = []
+
+    for p in paths:
+        abs_path = os.path.abspath(p)
+        if any(
+            abs_path == root or abs_path.startswith(root + os.sep)
+            for root in submodule_roots
+        ):
+            skipped_paths.append(p)
+        else:
+            repo_paths.append(p)
+
+    if skipped_paths:
+        msg = "Skipping staging for path(s) inside submodules: %s"
+        logger.debug(msg, ", ".join(skipped_paths))
+
+    scm.add(repo_paths, **kwargs)

dvc/stage/__init__.py

@@ -509,34 +509,20 @@ class Stage(params.StageParams):
                 if not allow_missing:
                     raise
 
-    def get_versioned_outs(self) -> dict[str, "Output"]:
-        from .exceptions import StageFileDoesNotExistError, StageNotFound
-
-        try:
-            old = self.reload()
-        except (StageFileDoesNotExistError, StageNotFound):
-            return {}
-
-        return {
-            out.def_path: out
-            for out in old.outs
-            if out.files is not None
-            or (out.meta is not None and out.meta.version_id is not None)
-        }
-
     def save_outs(self, allow_missing: bool = False):
         from dvc.output import OutputDoesNotExistError
 
-        old_versioned_outs = self.get_versioned_outs()
         for out in self.outs:
+            # old state just before saving so to merge them later
+            old_state = out._get_versioned_meta()
             try:
                 out.save()
             except OutputDoesNotExistError:
                 if not allow_missing:
                     raise
 
-            if old_out := old_versioned_outs.get(out.def_path):
-                out.merge_version_meta(old_out)
+            if old_state:
+                out.merge_version_meta(*old_state)
 
     def ignore_outs(self) -> None:
         for out in self.outs:
@@ -579,8 +565,9 @@ class Stage(params.StageParams):
         from dvc.output import OutputDoesNotExistError
 
         link_failures = []
-        old_versioned_outs = self.get_versioned_outs()
         for out in self.filter_outs(filter_info):
+            # old state just before saving so to merge them later
+            old_state = out._get_versioned_meta()
             try:
                 out.add(filter_info, **kwargs)
             except (FileNotFoundError, OutputDoesNotExistError):
@@ -589,8 +576,8 @@ class Stage(params.StageParams):
             except CacheLinkError:
                 link_failures.append(filter_info or out.fs_path)
 
-            if old_out := old_versioned_outs.get(out.def_path):
-                out.merge_version_meta(old_out)
+            if old_state:
+                out.merge_version_meta(*old_state)
 
         if link_failures:
             raise CacheLinkError(link_failures)

dvc/stage/serialize.py

@@ -168,6 +168,7 @@ def to_single_stage_lockfile(stage: "Stage", **kwargs) -> dict:
                     for f in _serialize_tree_obj_to_files(obj)
                 ]
         else:
+            assert item.meta is not None
             meta_d = item.meta.to_dict()
             meta_d.pop("isdir", None)
             ret.update(_serialize_hi_to_dict(item.hash_info))

dvc/testing/api_tests.py

@@ -32,12 +32,13 @@ class TestAPI:
         "fs_kwargs",
         [
             {},
-            {"url": "{path}"},
-            {"url": "{path}", "rev": "{default_branch}"},
-            {"url": "file://{posixpath}"},
-            {"url": "file://{posixpath}", "rev": "{default_branch}"},
+            {"repo": "{path}"},
+            {"repo": "{path}", "rev": "{default_branch}"},
+            {"repo": "file://{posixpath}"},
+            {"repo": "file://{posixpath}", "rev": "{default_branch}"},
+            {"url": "{path}"},  # test for backward compatibility
         ],
-        ids=["current", "local", "local_rev", "git", "git_rev"],
+        ids=["current", "local", "local_rev", "git", "git_rev", "local_url"],
     )
     def test_filesystem(
         self,
@@ -59,6 +60,8 @@ class TestAPI:
         if clear_cache:
             remove(dvc.cache.repo.path)
 
+        if repo := fs_kwargs.get("repo"):
+            fs_kwargs["repo"] = repo.format(path=tmp_dir, posixpath=tmp_dir.as_posix())
         if url := fs_kwargs.get("url"):
             fs_kwargs["url"] = url.format(path=tmp_dir, posixpath=tmp_dir.as_posix())
         if rev := fs_kwargs.get("rev"):

dvc/testing/benchmarks/cli/commands/test_data_status.py

@@ -4,12 +4,15 @@ from shutil import rmtree
 def test_data_status(bench_dvc, tmp_dir, scm, dvc, make_dataset):
     args = ("data", "status")
     dataset = make_dataset(cache=True, files=True, dvcfile=True, commit=False)
+    scm.ignore(dataset)
     rmtree(dvc.tmp_dir)
 
     bench_dvc(*args, name="new")
     bench_dvc(*args, name="noop")
 
-    tmp_dir.scm_add(dataset.with_suffix(".dvc").name, commit="add dataset")
+    tmp_dir.scm_add(
+        [dataset.with_suffix(".dvc").name, ".gitignore"], commit="add dataset"
+    )
 
     (dataset / "new").write_text("new")
     bench_dvc(*args, name="changed")
@@ -26,12 +29,15 @@ def test_data_status_all_flags(bench_dvc, tmp_dir, scm, dvc, make_dataset):
         "--json",
     )
     dataset = make_dataset(cache=True, files=True, dvcfile=True, commit=False)
+    scm.ignore(dataset)
     rmtree(dvc.tmp_dir)
 
     bench_dvc(*args, name="new")
     bench_dvc(*args, name="noop")
 
-    tmp_dir.scm_add(dataset.with_suffix(".dvc").name, commit="add dataset")
+    tmp_dir.scm_add(
+        [dataset.with_suffix(".dvc").name, ".gitignore"], commit="add dataset"
+    )
 
     (dataset / "new").write_text("new")
     bench_dvc(*args, name="changed")

dvc/testing/benchmarks/cli/stories/use_cases/test_sharing.py

@@ -1,7 +1,17 @@
 import shutil
 
+from packaging import version
+
+
+def test_sharing(bench_dvc, tmp_dir, dvc, make_dataset, remote, dvc_bin):
+    is_gs = dvc.config["remote"]["upstream"]["url"].startswith("gs://")
+    if is_gs and version.Version(dvc_bin.version) < version.Version("3.59.2"):
+        # support for allow_anonymous_login in gs was introduced in dvc==3.60
+        # This should not impact the test, just that it will make it slower if used
+        # with fake-gcs-server.
+        with dvc.config.edit() as d:
+            d["remote"]["upstream"].pop("allow_anonymous_login", None)
 
-def test_sharing(bench_dvc, tmp_dir, dvc, make_dataset, remote):
     dataset = make_dataset(cache=True, dvcfile=True)
 
     bench_dvc("push")

dvc/testing/benchmarks/fixtures.py

@@ -1,3 +1,4 @@
+import inspect
 import os
 import shutil
 import sys
@@ -76,7 +77,10 @@ def dvc_repo(tmp_path_factory, bench_config):
 def dvc_bench_repo(tmp_path_factory, bench_config):
     url = bench_config.dvc_bench_repo
     if url is None:
-        pytest.skip("--dvc-bench-repo is not set")
+        pytest.skip(
+            "--dvc-bench-repo is not set, "
+            "clone https://github.com/iterative/dvc-bench repository and set its path"
+        )
 
     if os.path.isdir(url):
         return Path(url)
@@ -138,8 +142,7 @@ def dvc_bin(request, make_dvc_bin):
         version = Version(parse(dvc_version).base_version)
         if version not in spec:
             pytest.skip(
-                f"Version {dvc_version} "
-                f"does not satisfy requirement {spec!r}: {reason}"
+                f"Version {dvc_version} does not satisfy requirement {spec!r}: {reason}"
             )
     return make_dvc_bin
 
@@ -150,8 +153,18 @@ def make_bench(request):
         import pytest_benchmark.plugin
 
         # hack from https://github.com/ionelmc/pytest-benchmark/issues/166
-        bench_gen = pytest_benchmark.plugin.benchmark.__pytest_wrapped__.obj(request)
-        bench = next(bench_gen)
+        fixture_function = pytest_benchmark.plugin.benchmark
+        try:
+            # pytest >= 8.4.0
+            wrapped_func = fixture_function._get_wrapped_function()
+        except AttributeError:
+            wrapped_func = fixture_function.__pytest_wrapped__.obj  # ty: ignore[unresolved-attribute]
+        assert inspect.isgeneratorfunction(wrapped_func)
+
+        generator = wrapped_func(request)
+        bench = next(generator)
+        assert isinstance(bench, pytest_benchmark.plugin.BenchmarkFixture)
+        request.addfinalizer(lambda: next(generator, None))
 
         suffix = f"-{name}"
 

dvc/testing/path_info.py

@@ -265,7 +265,7 @@ class CloudURLInfo(URLInfo):
 
 
 class HTTPURLInfo(URLInfo):
-    __hash__: Callable[["HTTPURLInfo"], int] = URLInfo.__hash__
+    __hash__: Callable[["HTTPURLInfo"], int] = URLInfo.__hash__  # type: ignore[assignment]
 
     def __init__(self, url):
         p = urlparse(url)

dvc/testing/remote_tests.py

@@ -137,8 +137,12 @@ class TestRemote:
         remove(tmp_dir / "bar")
 
         stats = dvc.pull()
-        assert stats["fetched"] == 2
-        assert set(stats["added"]) == {"foo", "bar"}
+        assert stats == {
+            "added": ["bar", "foo"],
+            "deleted": [],
+            "modified": [],
+            "stats": {"fetched": 2, "added": 2, "deleted": 0, "modified": 0},
+        }
 
     @pytest.mark.xfail(raises=NotImplementedError, strict=False)
     def test_pull_no_00_prefix(self, tmp_dir, dvc, remote, monkeypatch):
@@ -163,8 +167,12 @@ class TestRemote:
         remove(tmp_dir / "bar")
 
         stats = dvc.pull()
-        assert stats["fetched"] == 2
-        assert set(stats["added"]) == {"foo", "bar"}
+        assert stats == {
+            "added": ["bar", "foo"],
+            "deleted": [],
+            "modified": [],
+            "stats": {"fetched": 2, "added": 2, "deleted": 0, "modified": 0},
+        }
 
 
 class TestRemoteVersionAware:

dvc/testing/workspace_tests.py

@@ -163,7 +163,12 @@ class TestImportURLVersionAware:
 
         (remote_version_aware / "data_dir" / "foo").write_text("foo")
         dvc.update(no_download=True)
-        assert dvc.pull()["fetched"] == 2
+        assert dvc.pull() == {
+            "modified": [],
+            "added": ["data_dir" + os.sep],
+            "deleted": [],
+            "stats": {"fetched": 2, "modified": 0, "added": 2, "deleted": 0},
+        }
         assert (tmp_dir / "data_dir").read_text() == {
             "foo": "foo",
             "subdir": {"file": "file"},
@@ -174,7 +179,12 @@ class TestImportURLVersionAware:
         scm.checkout("v1")
         dvc.cache.local.clear()
         remove(tmp_dir / "data_dir")
-        assert dvc.pull()["fetched"] == 1
+        assert dvc.pull() == {
+            "modified": [],
+            "added": ["data_dir" + os.sep],
+            "deleted": [],
+            "stats": {"fetched": 1, "modified": 0, "added": 1, "deleted": 0},
+        }
         assert (tmp_dir / "data_dir").read_text() == {"subdir": {"file": "file"}}
 
         dvc.commit(force=True)
@@ -225,6 +235,39 @@ class TestLsUrl:
             ],
         )
 
+        result = ls_url(
+            str(cloud / "dir"), fs_config=cloud.config, recursive=True, maxdepth=0
+        )
+        match_files(
+            fs,
+            result,
+            [{"path": (cloud / "dir").fs_path, "isdir": False}],
+        )
+
+        result = ls_url(
+            str(cloud / "dir"), fs_config=cloud.config, recursive=True, maxdepth=1
+        )
+        match_files(
+            fs,
+            result,
+            [
+                {"path": "foo", "isdir": False},
+                {"path": "subdir", "isdir": True},
+            ],
+        )
+
+        result = ls_url(
+            str(cloud / "dir"), fs_config=cloud.config, recursive=True, maxdepth=2
+        )
+        match_files(
+            fs,
+            result,
+            [
+                {"path": "foo", "isdir": False},
+                {"path": "subdir/bar", "isdir": False},
+            ],
+        )
+
     def test_nonexistent(self, cloud):
         with pytest.raises(URLMissingError):
             ls_url(str(cloud / "dir"), fs_config=cloud.config)

dvc/ui/pager.py

@@ -44,6 +44,7 @@ def prepare_default_pager(
 
 def make_pager(cmd=None):
     def _pager(text):
+        assert cmd
         return pydoc.tempfilepager(pydoc.plain(text), cmd)
 
     return _pager if cmd else pydoc.plainpager

dvc/updater.py

@@ -99,13 +99,14 @@ class Updater:
         import json
 
         import requests
+        from requests.exceptions import RequestException
 
         url = os.environ.get(DVC_UPDATER_ENDPOINT, self.URL)
         logger.debug("Checking updates in %s", url)
         try:
             resp = requests.get(url, timeout=self.TIMEOUT_GET)
             info = resp.json()
-        except requests.exceptions.RequestException as exc:
+        except RequestException as exc:
             logger.trace("", exc_info=True)
             logger.debug("Failed to retrieve latest version: %s", exc)
             return

dvc/utils/__init__.py

@@ -8,6 +8,7 @@ import sys
 from typing import TYPE_CHECKING, Optional
 
 import colorama
+from colorama import AnsiToWin32
 
 if TYPE_CHECKING:
     from typing import TextIO
@@ -18,7 +19,7 @@ TARGET_REGEX = re.compile(r"(?P<path>.*?)(:(?P<name>[^\\/:]*))??$")
 
 
 def bytes_hash(byts, typ):
-    hasher = getattr(hashlib, typ)()
+    hasher = getattr(hashlib, typ)(usedforsecurity=False)
     hasher.update(byts)
     return hasher.hexdigest()
 
@@ -205,7 +206,7 @@ def boxify(message, border_color=None):
 def _visual_width(line):
     """Get the number of columns required to display a string"""
 
-    return len(re.sub(colorama.ansitowin32.AnsiToWin32.ANSI_CSI_RE, "", line))
+    return len(re.sub(AnsiToWin32.ANSI_CSI_RE, "", line))
 
 
 def _visual_center(line, width):

dvc/utils/hydra.py

@@ -115,6 +115,7 @@ def apply_overrides(path: "StrPath", overrides: list[str]) -> None:
         except hydra_errors as e:
             raise InvalidArgumentError("Invalid `--set-param` value") from e
 
+        assert isinstance(new_data, dict)
         merge_dicts(original_data, new_data)
         remove_missing_keys(original_data, new_data)
 

dvc/utils/strictyaml.py

@@ -20,6 +20,7 @@ from dvc.utils.serialize import (
     parse_yaml,
     parse_yaml_for_update,
 )
+from dvc_objects.fs.local import LocalFileSystem
 
 if TYPE_CHECKING:
     from rich.syntax import Syntax
@@ -34,15 +35,24 @@ _T = TypeVar("_T")
 merge_conflict_marker = re.compile("^([<=>]{7}) .*$", re.MULTILINE)
 
 
-def make_relpath(path: str) -> str:
-    import os
+def make_relpath(fs_path: str, fs: "Optional[FileSystem]" = None) -> str:
+    from os import curdir, pardir, sep
 
     from dvc.utils import relpath
 
-    rel = relpath(path)
-    prefix = ""
-    if not rel.startswith(".."):
-        prefix = "./" if os.name == "posix" else ".\\"
+    if not fs_path:
+        return fs_path
+
+    if fs and not isinstance(fs, LocalFileSystem):
+        rel = fs.relpath(fs_path).replace(fs.sep, sep)
+    else:
+        rel = relpath(fs_path)
+
+    if rel.startswith(sep):
+        # if it is an absolute path, return as it is.
+        # This can happen on Windows when we are on a different drive.
+        return rel
+    prefix = curdir + sep if not rel.startswith(pardir) else ""
     return prefix + rel
 
 
@@ -77,14 +87,14 @@ class YAMLSyntaxError(PrettyDvcException, YAMLFileCorruptedError):
         exc: Exception,
         rev: Optional[str] = None,
     ) -> None:
-        self.path: str = path
         self.yaml_text: str = yaml_text
         self.exc: Exception = exc
 
         merge_conflicts = merge_conflict_marker.search(self.yaml_text)
         self.hint = " (possible merge conflicts)" if merge_conflicts else ""
         self.rev: Optional[str] = rev
-        super().__init__(self.path)
+        super().__init__(path)
+        self.relpath: str = path
 
     def __pretty_exc__(self, **kwargs: Any) -> None:  # noqa: C901
         from ruamel.yaml.error import MarkedYAMLError
@@ -134,9 +144,8 @@ class YAMLSyntaxError(PrettyDvcException, YAMLFileCorruptedError):
             # if there are no other outputs
             lines.insert(0, "")
 
-        rel = make_relpath(self.path)
         rev_msg = f" in revision '{self.rev[:7]}'" if self.rev else ""
-        msg_fmt = f"'{rel}' is invalid{self.hint}{rev_msg}."
+        msg_fmt = f"'{self.relpath}' is invalid{self.hint}{rev_msg}."
         lines.insert(0, _prepare_message(msg_fmt))
         for line in lines:
             ui.error_write(line, styled=True)
@@ -198,10 +207,9 @@ class YAMLValidationError(PrettyDvcException):
         self.text = text or ""
         self.exc = exc
 
-        rel = make_relpath(path) if path else ""
         self.path = path or ""
 
-        message = f"'{rel}' validation failed"
+        message = f"'{self.path}' validation failed"
         message += f" in revision '{rev[:7]}'" if rev else ""
         if len(self.exc.errors) > 1:
             message += f": {len(self.exc.errors)} errors"
@@ -288,10 +296,12 @@ def load(
         raise EncodingError(path, encoding) from exc
     except YAMLFileCorruptedError as exc:
         cause = exc.__cause__
-        raise YAMLSyntaxError(path, text, exc, rev=rev) from cause
+        relpath = make_relpath(path, fs)
+        raise YAMLSyntaxError(relpath, text, exc, rev=rev) from cause
 
     if schema:
+        relpath = make_relpath(path, fs)
         # not returning validated data, as it may remove
         # details from CommentedMap that we get from roundtrip parser
-        validate(data, schema, text=text, path=path, rev=rev)
+        validate(data, schema, text=text, path=relpath, rev=rev)
     return data, text

dvc/utils/studio.py

@@ -23,7 +23,7 @@ if TYPE_CHECKING:
 
 logger = logger.getChild(__name__)
 
-STUDIO_URL = "https://studio.dvc.ai"
+STUDIO_URL = "https://studio.datachain.ai"
 
 
 def post(

pyproject.toml

@@ -1,6 +1,6 @@
 [build-system]
 build-backend = "setuptools.build_meta"
-requires = ["setuptools>=61", "setuptools_scm[toml]>=7"]
+requires = ["setuptools>=77", "setuptools_scm[toml]>=8"]
 
 [project]
 name = "dvc"
@@ -16,7 +16,8 @@ keywords = [
     "machine-learning",
     "reproducibility",
 ]
-license = { text = "Apache License 2.0" }
+license = "Apache-2.0"
+license-files = ["LICENSE"]
 maintainers = [{ name = "Iterative", email = "[email protected]" }]
 authors = [{ name = "Dmitry Petrov", email = "[email protected]" }]
 requires-python = ">=3.9"
@@ -27,6 +28,7 @@ classifiers = [
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
 ]
 dynamic = ["version"]
 
@@ -65,7 +67,7 @@ dependencies = [
     "requests>=2.22",
     "rich>=12",
     "ruamel.yaml>=0.17.11",
-    "scmrepo>=3.3.8,<4",
+    "scmrepo>=3.5.2,<4",
     "shortuuid>=0.5",
     "shtab<2,>=1.3.4",
     "tabulate>=0.8.7",
@@ -80,10 +82,10 @@ all = ["dvc[azure,gdrive,gs,hdfs,oss,s3,ssh,webdav,webhdfs]"]
 azure = ["dvc-azure>=3.1.0,<4"]
 dev = ["dvc[azure,gdrive,gs,hdfs,lint,oss,s3,ssh,tests,webdav,webhdfs]"]
 gdrive = ["dvc-gdrive>=3,<4"]
-gs = ["dvc-gs>=3,<4"]
+gs = ["dvc-gs>=3.0.2,<4"]
 hdfs = ["dvc-hdfs>=3,<4"]
 lint = [
-    "mypy==1.13.0",
+    "mypy==1.17.1",
     "pandas-stubs",
     "types-colorama",
     "types-psutil",
@@ -95,7 +97,7 @@ lint = [
     "typing-extensions",
 ]
 oss = ["dvc-oss>=3,<4"]
-s3 = ["dvc-s3>=3.2.0,<4"]
+s3 = ["dvc-s3>=3.2.1,<4"]
 ssh = ["dvc-ssh>=4,<5"]
 ssh_gssapi = ["dvc-ssh[gssapi]>=4,<5"]
 testing = [
@@ -109,7 +111,7 @@ tests = [
     "dvc-ssh",
     "filelock",
     "pytest>=7,<9",
-    "pytest-rerunfailures",
+    "pytest-rerunfailures<16.0",
     "pytest-cov>=4.1.0",
     "pytest-docker>=1,<4",
     "pytest-mock",
@@ -148,9 +150,6 @@ dvc = "dvc.fs.dvc_path:DVCPath"
 hook-dirs = "dvc.__pyinstaller:get_hook_dirs"
 tests = "dvc.__pyinstaller:get_PyInstaller_tests"
 
-[tool.setuptools]
-license-files = ["LICENSE"]
-
 [tool.setuptools.packages.find]
 exclude = ["tests", "tests.*"]
 namespaces = false
@@ -172,8 +171,10 @@ filterwarnings = [
     "ignore:Deprecated call to `pkg_resources.declare_namespace:DeprecationWarning",
     # see https://github.com/networkx/networkx/issues/5723.
     "ignore:nx.nx_pydot.* depends on the pydot package, which has.*known issues and is not actively maintained:DeprecationWarning",
+    # TODO: investigate where we are not closing sqlite3.Connection
+    "ignore:unclosed database.*sqlite3.Connection:ResourceWarning",
+    "ignore:unclosed.*<socket.socket:ResourceWarning",
 ]
-log_level = "debug"
 markers = [
     "needs_internet: Might need network access for the tests",
     "studio: Tests verifying contract between DVC and Studio",
@@ -264,14 +265,14 @@ show-fixes = true
 
 [tool.ruff.lint]
 ignore = [
-    "N818", "S101", "ISC001", "PT007", "RET502", "RET503", "SIM105", "SIM108", "SIM117",
-    "TRY003", "TRY300", "PERF203", "PLR2004", "PLW2901", "LOG007",
+    "N818", "S101", "A005", "PT007", "RET502", "RET503", "SIM105", "SIM108", "SIM117",
+    "TRY003", "TRY300", "PERF203", "PLC0415", "PLR2004",  "PLW2901", "LOG007",
 ]
 select = [
     "F", "E", "W", "C90", "I", "N", "UP", "YTT", "ASYNC", "S", "BLE", "B", "A", "C4", "DTZ", "T10",
     "EXE", "ISC", "ICN", "LOG", "G", "INP", "PIE", "T20", "PYI", "PT", "Q", "RSE", "RET",
     "SLOT", "SIM", "TID", "TC", "ARG", "PGH", "PLC", "PLE", "PLR", "PLW", "TRY",
-    "FLY", "PERF", "FURB", "RUF", "RUF027", "RUF029", "RUF034", "RUF036",
+    "FLY", "PERF", "FURB", "RUF", "RUF027", "RUF029", "RUF036",
 ]
 preview = true
 explicit-preview-rules = true

tests/conftest.py

@@ -1,7 +1,7 @@
 import json
 import os
 import sys
-from contextlib import suppress
+from contextlib import ExitStack, suppress
 
 import pytest
 
@@ -46,10 +46,10 @@ def reset_loglevel(request, caplog):
         ini_opt = request.config.getini("log_level")
 
     level = request.config.getoption("--log-level") or ini_opt
-    if level:
-        with caplog.at_level(level.upper(), logger="dvc"):
-            yield
-    else:
+    with ExitStack() as stack:
+        if level:
+            for name in ["dvc", "dvc_data", "dvc_objects"]:
+                stack.enter_context(caplog.at_level(level.upper(), logger=name))
         yield
 
 

tests/dir_helpers.py

@@ -46,6 +46,7 @@ from global repo template to creating everything inplace, which:
 import os
 
 import pytest
+from dulwich.porcelain import remote_add as git_remote_add
 
 __all__ = [
     "erepo_dir",
@@ -96,7 +97,7 @@ class GitRemote:
 def git_upstream(tmp_dir, erepo_dir, git_dir, request):
     remote = erepo_dir if "dvc" in request.fixturenames else git_dir
     url = f"file://{remote.resolve().as_posix()}"
-    tmp_dir.scm.gitpython.repo.create_remote("upstream", url)
+    git_remote_add(tmp_dir, "upstream", url)
     return GitRemote(remote, "upstream", url)
 
 
@@ -104,5 +105,5 @@ def git_upstream(tmp_dir, erepo_dir, git_dir, request):
 def git_downstream(tmp_dir, erepo_dir, git_dir, request):
     remote = erepo_dir if "dvc" in request.fixturenames else git_dir
     url = f"file://{tmp_dir.resolve().as_posix()}"
-    remote.scm.gitpython.repo.create_remote("upstream", url)
+    git_remote_add(remote, "upstream", url)
     return GitRemote(remote, "upstream", url)

tests/func/api/test_data.py

@@ -5,6 +5,7 @@ from funcy import first, get_in
 
 from dvc import api
 from dvc.exceptions import OutputNotFoundError, PathMissingError
+from dvc.scm import CloneError
 from dvc.testing.api_tests import TestAPI  # noqa: F401
 from dvc.testing.tmp_dir import make_subrepo
 from dvc.utils.fs import remove
@@ -65,6 +66,34 @@ def test_get_url_from_remote(tmp_dir, erepo_dir, cloud, local_cloud):
     )
 
 
+def test_get_url_ignore_scm(tmp_dir, dvc, cloud, scm):
+    tmp_dir.add_remote(config=cloud.config)
+    tmp_dir.dvc_gen("foo", "foo", commit="add foo")
+
+    repo_posix = tmp_dir.as_posix()
+    expected_url = (cloud / "files" / "md5" / "ac/bd18db4cc2f85cedef654fccc4a4d8").url
+
+    # Test baseline with scm
+    assert api.get_url("foo", repo=repo_posix) == expected_url
+
+    # Simulate gitless environment (e.g. deployed container)
+    (tmp_dir / ".git").rename(tmp_dir / "gitless_environment")
+
+    assert api.get_url("foo", repo=repo_posix) == expected_url
+    assert (
+        api.get_url("foo", repo=repo_posix, config={"core": {"no_scm": True}})
+        == expected_url
+    )
+
+    # Addressing repos with `file://` triggers git, so it fails in a gitless environment
+    repo_url = f"file://{repo_posix}"
+    with pytest.raises(
+        CloneError,
+        match="SCM error",
+    ):
+        api.get_url("foo", repo=repo_url, config={"core": {"no_scm": True}})
+
+
 def test_open_external(tmp_dir, erepo_dir, cloud):
     erepo_dir.add_remote(config=cloud.config)
 

tests/func/artifacts/test_artifacts.py

@@ -4,6 +4,7 @@ from copy import deepcopy
 
 import pytest
 
+from dvc import env
 from dvc.annotations import Artifact
 from dvc.exceptions import ArtifactNotFoundError, InvalidArgumentError
 from dvc.repo.artifacts import Artifacts, check_name_format
@@ -283,7 +284,13 @@ def test_artifacts_download_subrepo(tmp_dir, scm, sub):
         assert Artifacts.get(".", name, force=True) == result
 
 
-def test_artifacts_download_studio(tmp_dir, dvc, mocker):
+def test_artifacts_download_studio(monkeypatch, tmp_dir, dvc, mocker, tmp_path_factory):
+    global_config_dir = tmp_path_factory.mktemp("global_config")
+    monkeypatch.setenv(env.DVC_GLOBAL_CONFIG_DIR, str(global_config_dir))
+    # reset config to ensure it picks up the new global config
+    dvc.__dict__.pop("config")
+    assert dvc.config.files["global"] == os.fspath(global_config_dir / "config")
+
     with dvc.config.edit("global") as conf:
         conf["studio"]["token"] = "mytoken"