Merge origin/main into lockfile

Author: tobiasraabe

.pre-commit-config.yaml

@@ -25,12 +25,12 @@ repos:
     -   id: python-no-log-warn
     -   id: text-unicode-replacement-char
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.14.14
+    rev: v0.15.1
     hooks:
       - id: ruff-format
       - id: ruff-check
 -   repo: https://github.com/astral-sh/uv-pre-commit
-    rev: 0.9.26
+    rev: 0.10.3
     hooks:
       - id: uv-lock
 -   repo: https://github.com/executablebooks/mdformat

CHANGELOG.md

@@ -10,9 +10,18 @@ releases are available on [PyPI](https://pypi.org/project/pytask) and
 - {issue}`735` adds the `pytask.lock` lockfile as the primary state backend with a
   portable format, documentation, and a one-run SQLite fallback when no lockfile
   exists.
+- {pull}`787` makes the `attributes` field mandatory on `PNode` and
+  `PProvisionalNode`, and preserves existing node attributes when loading entries from
+  the data catalog.
+- {pull}`744` Removed the direct dependency on attrs and migrated internal models to
+  dataclasses.
 - {pull}`766` moves runtime profiling persistence from SQLite to a JSON snapshot plus
   append-only journal in `.pytask/`, keeping runtime data resilient to crashes and
   compacted on normal build exits.
+- {pull}`776` clears decoration-time `annotation_locals` snapshots after collection so
+  task functions remain picklable in process-based parallel backends.
+- {pull}`788` updates debugger interaction tests to support Python-version-specific
+  `pdb` stepping behavior and avoid frame-specific assumptions on repeated breakpoints.
 
 ## 0.5.8 - 2025-12-30
 

docs/source/explanations/comparison_to_other_tools.md

@@ -95,6 +95,11 @@ pipefunc is a lightweight library for creating function pipelines as directed ac
 graphs (DAGs) in pure Python. It automatically handles execution order, supports
 map-reduce operations, parallel execution, and provides resource profiling.
 
+## [Hamilton](https://github.com/dagworks-inc/hamilton)
+
+Hamilton is a Python framework for building dataflows as function graphs, with a focus
+on modular transformations, lineage, and observability for data and ML pipelines.
+
 ## [Common Workflow Language (CWL)](https://www.commonwl.org/)
 
 CWL is an open standard for describing data analysis workflows in a portable,

docs/source/how_to_guides/writing_custom_nodes.md

@@ -127,11 +127,13 @@ databases. [^kedro]
 
 ## References
 
-[^structural-subtyping]: Structural subtyping is similar to ABCs an approach in Python to enforce interfaces,
-    but it can be considered more pythonic since it is closer to duck typing. Hynek
-    Schlawack wrote a comprehensive
-    [guide on subclassing](https://hynek.me/articles/python-subclassing-redux/) that
-    features protocols under "Type 2". Glyph wrote an introduction to protocols called
+[^structural-subtyping]: Structural subtyping is similar to ABCs an approach in Python
+    to enforce interfaces, but it can be considered more pythonic
+    since it is closer to duck typing. Hynek Schlawack wrote a
+    comprehensive
+    [guide on subclassing](https://hynek.me/articles/python-subclassing-redux/)
+    that features protocols under "Type 2". Glyph wrote an
+    introduction to protocols called
     [I want a new duck](https://glyph.twistedmatrix.com/2020/07/new-duck.html).
 
 [^kedro]: Kedro, another workflow system, provides many adapters to data sources:

docs/source/plugin_list.rst

@@ -25,7 +25,7 @@ This list contains 7 plugins.
    :pypi:`pytask-environment`  Detect changes in your pytask environment and abort a project build.  Feb 10, 2023    4 - Beta   pytask (>=0.2)
    :pypi:`pytask-julia`        A Pytask plugin for Julia                                             Oct 07, 2023    4 - Beta   pytask >=0.4.0
    :pypi:`pytask-latex`        Compile LaTeX documents with pytask.                                  Jun 01, 2025    4 - Beta   pytask>=0.4.0
-   :pypi:`pytask-parallel`     Parallelize the execution of tasks with pytask.                       Mar 09, 2025    4 - Beta   pytask>=0.5.2
+   :pypi:`pytask-parallel`     Parallelize the execution of tasks with pytask.                       Feb 06, 2026    4 - Beta   pytask>=0.5.2
    :pypi:`pytask-r`            Run R scripts with pytask.                                            Apr 20, 2024    4 - Beta   pytask>=0.4.5
    :pypi:`pytask-stata`        Execute do-files with Stata and pytask.                               Jul 26, 2025    4 - Beta   pytask>=0.5.2
    :pypi:`pytask-vscode`       Additional Logging for VS Code integration                            Nov 21, 2023    3 - Alpha  pytask >=0.4.2

docs/source/reference_guides/api.md

@@ -203,6 +203,7 @@ Nodes are the interface for different kinds of dependencies or products.
    :members:
 .. autoclass:: pytask.PickleNode
    :members:
+   :exclude-members: serializer, deserializer
 .. autoclass:: pytask.PythonNode
    :members:
 .. autoclass:: pytask.DirectoryNode

docs_src/how_to_guides/the_data_catalog.py

@@ -1,11 +1,11 @@
+from dataclasses import dataclass
 from pathlib import Path
 from typing import Any
 
 import cloudpickle
-from attrs import define
 
 
-@define
+@dataclass
 class PickleNode:
     """A node for pickle files.
 

pyproject.toml

@@ -21,7 +21,6 @@ classifiers = [
 ]
 dynamic = ["version"]
 dependencies = [
-    "attrs>=21.3.0",
     "click>=8.1.8,!=8.2.0",
     "click-default-group>=1.2.4",
     "msgspec>=0.18.6",

src/_pytask/cache.py

@@ -5,16 +5,15 @@
 import functools
 import hashlib
 import inspect
+from dataclasses import dataclass
+from dataclasses import field
 from inspect import FullArgSpec
 from typing import TYPE_CHECKING
 from typing import Any
 from typing import ParamSpec
 from typing import Protocol
 from typing import TypeVar
 
-from attrs import define
-from attrs import field
-
 from _pytask._hashlib import hash_value
 
 if TYPE_CHECKING:
@@ -35,17 +34,17 @@ class HasCache(Protocol):
     cache: Cache
 
 
-@define
+@dataclass
 class CacheInfo:
     hits: int = 0
     misses: int = 0
 
 
-@define
+@dataclass
 class Cache:
-    _cache: dict[str, Any] = field(factory=dict)
-    _sentinel: Any = field(factory=object)
-    cache_info: CacheInfo = field(factory=CacheInfo)
+    _cache: dict[str, Any] = field(default_factory=dict)
+    _sentinel: Any = field(default_factory=object)
+    cache_info: CacheInfo = field(default_factory=CacheInfo)
 
     def memoize(self, func: Callable[P, R]) -> Memoized[P, R]:
         func_module = getattr(func, "__module__", "")

src/_pytask/clean.py

@@ -6,11 +6,11 @@
 import itertools
 import shutil
 import sys
+from dataclasses import dataclass
 from typing import TYPE_CHECKING
 from typing import Any
 
 import click
-from attrs import define
 
 from _pytask.click import ColoredCommand
 from _pytask.click import EnumChoice
@@ -243,7 +243,7 @@ def _find_all_unknown_paths(
     )
 
 
-@define(repr=False)
+@dataclass(repr=False)
 class _RecursivePathNode:
     """A class for a path to a file or directory which recursively instantiates itself.