Add --use-standard-primitive-types option by koxudaxi · Pull Request #2736 · koxudaxi/datamodel-code-generator

koxudaxi · 2025-12-22T14:20:39Z

Summary by CodeRabbit

Release Notes

New Features
- Added --use-standard-primitive-types CLI option to use Python standard library types (UUID, IPv4Address, IPv6Address, Path) for string format fields instead of plain strings in dataclass, msgspec, and TypedDict model outputs. Pydantic models already use these types by default.
Documentation
- Updated CLI reference and typing customization documentation to describe the new --use-standard-primitive-types option and its effects.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

coderabbitai · 2025-12-22T14:20:49Z

Walkthrough

Adds a new CLI flag --use-standard-primitive-types to enable dataclass and msgspec models to use Python standard library types (UUID, IPv4Address, IPv6Address, Path) for string formats instead of plain str, achieving parity with Pydantic's type handling. The flag is threaded through CLI arguments, config, parsers, and type managers to enable conditional primitive-type mapping.

Changes

Cohort / File(s)	Change Summary
Documentation Updates `docs/cli-reference/index.md`, `docs/cli-reference/quick-reference.md`, `docs/cli-reference/typing-customization.md`	Added CLI option documentation and reference for `--use-standard-primitive-types` flag with descriptions, usage examples, and affected output formats.
CLI Infrastructure `src/datamodel_code_generator/arguments.py`, `src/datamodel_code_generator/cli_options.py`	Added `--use-standard-primitive-types` argument definition and CLI option metadata entry under Typing category.
Core Generate & Config `src/datamodel_code_generator/__init__.py`, `src/datamodel_code_generator/__main__.py`	Added `use_standard_primitive_types` parameter to `generate()` function and `Config` class, with propagation to parser instantiation and config flow.
Type System Foundation `src/datamodel_code_generator/imports.py`, `src/datamodel_code_generator/types.py`	Added Import constants for IPv4Address, IPv6Address, IPv4Network, IPv6Network; introduced `standard_primitive_type_map_factory()` function and threaded `use_standard_primitive_types` through base `DataTypeManager`.
Dataclass Model `src/datamodel_code_generator/model/dataclass.py`	Added `use_standard_primitive_types` parameter to `DataClass` and `DataTypeManager` constructors; integrated standard primitive type mappings into type map when flag is enabled.
Msgspec Model `src/datamodel_code_generator/model/msgspec.py`	Added `use_standard_primitive_types` parameter and conditional standard primitive type map integration into `DataTypeManager`.
Pydantic Type Managers `src/datamodel_code_generator/model/pydantic/types.py`, `src/datamodel_code_generator/model/pydantic_v2/types.py`	Added `use_standard_primitive_types` parameter to `DataTypeManager` constructors; refactored to keyword arguments in pydantic/types.py.
Parser Classes `src/datamodel_code_generator/parser/base.py`, `src/datamodel_code_generator/parser/graphql.py`, `src/datamodel_code_generator/parser/jsonschema.py`, `src/datamodel_code_generator/parser/openapi.py`	Added `use_standard_primitive_types` parameter to parser `__init__` signatures and propagated through superclass initialization.
Test Data & Tests `tests/data/jsonschema/use_standard_primitive_types.json`, `tests/data/expected/main/use_standard_primitive_types.py`, `tests/main/test_main_general.py`	Added JSON Schema input with uuid, ipv4, and path formats; generated dataclass output with standard library types; added test case validating end-to-end feature functionality.

Sequence Diagram

sequenceDiagram
    actor User
    participant CLI as CLI Arguments
    participant Config as Config
    participant Generator as generate()
    participant Parser as Parser
    participant DTM as DataTypeManager
    participant TypeMap as Type Map Factory
    
    User->>CLI: --use-standard-primitive-types
    CLI->>Config: use_standard_primitive_types=True
    Config->>Generator: run_generate_from_config(use_standard_primitive_types=True)
    Generator->>Parser: Parser(..., use_standard_primitive_types=True)
    Parser->>DTM: DataTypeManager(..., use_standard_primitive_types=True)
    DTM->>TypeMap: standard_primitive_type_map_factory()
    TypeMap-->>DTM: {UUID→Type, IPv4Address→Type, ...}
    DTM->>DTM: Merge primitive mappings into type_map
    DTM-->>Parser: Configured type mappings
    Parser-->>Generator: Generate models with standard lib types
    Generator-->>User: Output with UUID, IPv4Address, Path, etc.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Repetitive pattern: The same parameter addition and threading pattern is applied across ~10 files (parsers, type managers), which reduces cognitive load per file.
New logic: standard_primitive_type_map_factory() and integration into type map construction requires careful review to ensure correct type mappings.
File spread: Changes span documentation, CLI infrastructure, core generation, type system, and multiple model backends, requiring familiarity with multiple code areas.
Special cases: pydantic_v2/types.py adds the parameter but doesn't consume it—verify if this is intentional or incomplete.
Test coverage: Duplication in test_main_general.py (test function appears twice) should be consolidated.

Possibly related PRs

PR #2673: Extends the CLI documentation infrastructure and CLI_OPTION_META structure used by the new --use-standard-primitive-types flag registration.
PR #2733: Adds another CLI/config option and modifies the same parser/DataTypeManager constructor call sites and threading pattern as this PR.
PR #2688: Introduces a related CLI flag and modifies the same code paths (generate/init, main, arguments, cli_options, and OpenAPIParser.init).

Suggested labels

safe-to-fix

Suggested reviewers

ilovelinux

Poem

🐰 Hops with joy at types so clear,
No more strings for formats dear!
UUIDs bloom and paths take flight,
Dataclasses now shine so bright!
Standard lib types, here at last,
Parity with Pydantic—cast! ✨

Pre-merge checks and finishing touches

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The pull request title directly and specifically describes the main change: adding a new CLI option --use-standard-primitive-types to use standard library types for formatted strings.
Linked Issues check	✅ Passed	The PR successfully implements the core objective from issue #1784: enabling dataclass generation to use standard library types (UUID, IPv4Address, Path) for formatted string fields instead of plain str, matching Pydantic behavior.
Out of Scope Changes check	✅ Passed	All changes are within scope and directly support the --use-standard-primitive-types feature, from CLI arguments to parser initialization to type mapping and test cases.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

✨ Finishing touches

📝 Generate docstrings

🧪 Generate unit tests (beta)

Create PR with unit tests
Post copyable unit tests in a comment
Commit unit tests in branch feature/use-standard-primitive-types

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

🤖 Generated by GitHub Actions

codecov · 2025-12-22T14:25:41Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.36%. Comparing base (a231d46) to head (568c21e).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2736   +/-   ##
=======================================
  Coverage   99.36%   99.36%           
=======================================
  Files          83       83           
  Lines       12107    12123   +16     
  Branches     1458     1458           
=======================================
+ Hits        12030    12046   +16     
  Misses         45       45           
  Partials       32       32

Flag	Coverage Δ
unittests	`99.36% <100.00%> (+<0.01%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

codspeed-hq · 2025-12-22T14:27:25Z

CodSpeed Performance Report

Merging #2736 will not alter performance

_{Comparing feature/use-standard-primitive-types (568c21e) with main (a231d46)}

Summary

✅ 70 untouched
⏩ 10 skipped¹

10 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

coderabbitai

Actionable comments posted: 0

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

src/datamodel_code_generator/model/pydantic/types.py (1)
173-201: Clarify intentional no-op flag and clean up unused noqa directives

DataTypeManager.__init__ now accepts use_standard_primitive_types but doesn’t forward it to super().__init__, effectively making the flag a no-op for Pydantic v1. That’s reasonable given the docs explicitly say Pydantic already uses these standard types by default, and the parameter is needed mainly so this class can be constructed with the shared kwargs set.

Ruff is flagging the newly added # noqa entries here (FBT001, FBT002) as unused. To satisfy the linter while keeping the intentional “unused-arg” suppression, consider trimming the comments as follows:
Suggested change to align with Ruff RUF100
-        use_standard_primitive_types: bool = False,  # noqa: FBT001, FBT002, ARG002
-        treat_dot_as_module: bool | None = None,  # noqa: FBT001
-        use_serialize_as_any: bool = False,  # noqa: FBT001, FBT002
+        use_standard_primitive_types: bool = False,  # noqa: ARG002
+        treat_dot_as_module: bool | None = None,
+        use_serialize_as_any: bool = False,
This keeps the constructor signature compatible with the shared parser kwargs, avoids new RUF100 warnings, and leaves behavior unchanged.

🧹 Nitpick comments (4)

docs/cli-reference/typing-customization.md (1)

22-23: --use-standard-primitive-types docs and example are accurate; minor markdownlint nit

The option description, “Related” links, and dataclass example (UUID / IPv4Address / Path) accurately reflect the new behavior and match the test/golden output wiring.

markdownlint (MD050) flags the **Related:** line here for strong-style; if that rule is enforced, consider switching to __Related:__ (or relaxing the rule) to keep the linter happy.

Also applies to: 2959-3024
src/datamodel_code_generator/model/pydantic_v2/types.py (1)
72-100: Tidy up unused noqa codes on use_standard_primitive_types

The extra ctor parameter is fine as an API compatibility shim (Parser can always pass it) and leaving it unused here preserves existing Pydantic v2 behavior. Given Ruff’s RUF100 hint, you can drop the FBT001, FBT002 parts from the noqa and keep only ARG002:
Proposed tweak
-        use_standard_primitive_types: bool = False,  # noqa: FBT001, FBT002, ARG002
+        use_standard_primitive_types: bool = False,  # noqa: ARG002
src/datamodel_code_generator/model/types.py (1)

10-21: Standard primitive mapping is well factored; consider trimming unused noqa

The new standard_primitive_type_map_factory() and the use_standard_primitive_types switch in DataTypeManager.__init__ cleanly layer stdlib types (UUID, IPv4Address, IPv6Address, networks, Path) on top of the existing type_map_factory while preserving previous defaults when the flag is False.

Minor lint nit: Ruff reports unused noqa directives for FBT001/FBT002 on the use_standard_primitive_types, treat_dot_as_module, and use_serialize_as_any parameters. If those codes aren’t enabled in your config, you can safely drop them and keep only the ones you actually need.

Also applies to: 77-141
src/datamodel_code_generator/model/dataclass.py (1)
220-220: Remove unused noqa directive.

The noqa directive for FBT001 and FBT002 is no longer needed according to Ruff.
🔎 Proposed fix
-        use_standard_primitive_types: bool = False,  # noqa: FBT001, FBT002
+        use_standard_primitive_types: bool = False,

📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between a231d46 and 568c21e.

📒 Files selected for processing (21)

docs/cli-reference/index.md
docs/cli-reference/quick-reference.md
docs/cli-reference/typing-customization.md
src/datamodel_code_generator/__init__.py
src/datamodel_code_generator/__main__.py
src/datamodel_code_generator/arguments.py
src/datamodel_code_generator/cli_options.py
src/datamodel_code_generator/imports.py
src/datamodel_code_generator/model/dataclass.py
src/datamodel_code_generator/model/msgspec.py
src/datamodel_code_generator/model/pydantic/types.py
src/datamodel_code_generator/model/pydantic_v2/types.py
src/datamodel_code_generator/model/types.py
src/datamodel_code_generator/parser/base.py
src/datamodel_code_generator/parser/graphql.py
src/datamodel_code_generator/parser/jsonschema.py
src/datamodel_code_generator/parser/openapi.py
src/datamodel_code_generator/types.py
tests/data/expected/main/use_standard_primitive_types.py
tests/data/jsonschema/use_standard_primitive_types.json
tests/main/test_main_general.py

🧰 Additional context used

🧬 Code graph analysis (4)

src/datamodel_code_generator/model/msgspec.py (1)

src/datamodel_code_generator/model/types.py (2)

standard_primitive_type_map_factory (77-96)

type_map_factory (34-74)

src/datamodel_code_generator/model/types.py (1)

src/datamodel_code_generator/types.py (3)

DataType (285-757)

Types (767-803)

from_import (344-368)

src/datamodel_code_generator/model/dataclass.py (3)

src/datamodel_code_generator/model/types.py (2)

standard_primitive_type_map_factory (77-96)

type_map_factory (34-74)

src/datamodel_code_generator/model/pydantic_v2/types.py (1)

type_map_factory (116-150)

src/datamodel_code_generator/types.py (2)

Types (767-803)

DataType (285-757)

tests/main/test_main_general.py (3)

tests/conftest.py (1)

freeze_time (285-288)

tests/main/conftest.py (2)

output_file (98-100)

run_main_and_assert (244-408)

tests/test_main_kr.py (1)

output_file (44-46)

🪛 GitHub Check: CodeQL

src/datamodel_code_generator/model/msgspec.py

[notice] 37-37: Cyclic import
Import of module datamodel_code_generator.model.types begins an import cycle.

src/datamodel_code_generator/model/dataclass.py

[notice] 23-23: Cyclic import
Import of module datamodel_code_generator.model.types begins an import cycle.

🪛 markdownlint-cli2 (0.18.1)

docs/cli-reference/typing-customization.md

2968-2968: Strong style
Expected: underscore; Actual: asterisk

(MD050, strong-style)

2968-2968: Strong style
Expected: underscore; Actual: asterisk

(MD050, strong-style)

🪛 Ruff (0.14.8)

src/datamodel_code_generator/types.py

827-827: Unused noqa directive (non-enabled: FBT001, FBT002)