Keep generate string input inline (#3573)

* Keep generate string input inline

* Address generate string input review

Author: koxudaxi

docs/llms-full.txt

@@ -35736,14 +35736,32 @@ result = generate(json_schema, config=config)
 print(result)
 ```
 
+### 📝 Reading Input From a File
+
+Pass a `Path` to `input_` for file input. Plain strings are treated as schema
+text; if a string points to an existing file and parsing fails, `generate()`
+warns and recommends using `Path`.
+
+```python
+from pathlib import Path
+
+from datamodel_code_generator import InputFileType, generate
+
+result = generate(
+    input_=Path("example.json"),
+    input_file_type=InputFileType.JsonSchema,
+)
+print(result)
+```
+
 ### 📝 Multiple Module Output
 
 When the schema generates multiple modules, `generate()` returns a `GeneratedModules` dictionary mapping module path tuples to generated code:
 
 ```python
 from datamodel_code_generator import InputFileType, generate, GenerateConfig, GeneratedModules
 
-# Your OpenAPI specification (string, Path, or dict)
+# Your OpenAPI specification (string schema text, Path, or dict)
 openapi_spec: str = "..."  # Replace with your actual OpenAPI spec
 
 # Schema that generates multiple modules (e.g., with $ref to other files)

docs/using_as_module.md

@@ -42,14 +42,32 @@ result = generate(json_schema, config=config)
 print(result)
 ```
 
+### 📝 Reading Input From a File
+
+Pass a `Path` to `input_` for file input. Plain strings are treated as schema
+text; if a string points to an existing file and parsing fails, `generate()`
+warns and recommends using `Path`.
+
+```python
+from pathlib import Path
+
+from datamodel_code_generator import InputFileType, generate
+
+result = generate(
+    input_=Path("example.json"),
+    input_file_type=InputFileType.JsonSchema,
+)
+print(result)
+```
+
 ### 📝 Multiple Module Output
 
 When the schema generates multiple modules, `generate()` returns a `GeneratedModules` dictionary mapping module path tuples to generated code:
 
 ```python
 from datamodel_code_generator import InputFileType, generate, GenerateConfig, GeneratedModules
 
-# Your OpenAPI specification (string, Path, or dict)
+# Your OpenAPI specification (string schema text, Path, or dict)
 openapi_spec: str = "..."  # Replace with your actual OpenAPI spec
 
 # Schema that generates multiple modules (e.g., with $ref to other files)

src/datamodel_code_generator/__init__.py

@@ -106,8 +106,6 @@
 Returned by generate() when output=None and multiple modules are generated.
 """
 
-_SCHEMA_TEXT_START_CHARS = ("{", "[", "<")
-
 
 def _apply_generate_config_preset(config: GenerateConfig) -> GenerateConfig:
     """Return a generate config with preset defaults applied."""
@@ -761,21 +759,41 @@ def _generate_config_values(generate_config: GenerateConfig) -> dict[str, Any]:
     return values
 
 
-def _coerce_existing_path_input(
+def _warn_if_input_string_points_to_existing_path(
     input_: Path | str | ParseResult | Mapping[str, Any] | list[Any],
-) -> Path | str | ParseResult | Mapping[str, Any] | list[Any]:
-    if not isinstance(input_, str):
-        return input_
-    if not input_ or "\n" in input_ or "\r" in input_:
-        return input_
-    if (stripped := input_.lstrip()).startswith(_SCHEMA_TEXT_START_CHARS) or "://" in stripped:
-        return input_
+) -> None:
+    match input_:
+        case str() as input_text if input_text and "\n" not in input_text and "\r" not in input_text:
+            pass
+        case _:
+            return
     try:
-        path = Path(input_).expanduser()
+        path = Path(input_text).expanduser()
         path_exists = path.exists()
-    except (OSError, RuntimeError, ValueError):  # pragma: no cover
-        return input_
-    return path if path_exists else input_
+    except (OSError, RuntimeError, ValueError):
+        return
+    if not path_exists:
+        return
+
+    import warnings  # noqa: PLC0415
+
+    with contextlib.suppress(Warning):
+        warnings.warn(
+            "`input_` strings are treated as schema text. "
+            "The value also resolves to an existing path; pass a `Path` object to read it as a file.",
+            stacklevel=3,
+        )
+
+
+@contextlib.contextmanager
+def _warn_on_input_string_path_failure(
+    input_: Path | str | ParseResult | Mapping[str, Any] | list[Any],
+) -> Iterator[None]:
+    try:
+        yield
+    except Exception:
+        _warn_if_input_string_points_to_existing_path(input_)
+        raise
 
 
 def _create_parser_config(
@@ -1320,7 +1338,7 @@ def generate(  # noqa: PLR0912, PLR0914, PLR0915
     (JSON, YAML, Dict, CSV) as input.
 
     Args:
-        input_: The input source (file path, string content, URL, dict,
+        input_: The input source (Path file input, string content, URL, dict,
             list of file paths, or MCP tools list when input_file_type is
             InputFileType.MCPTools).
         config: A GenerateConfig object with all options. Cannot be used together with **options.
@@ -1361,8 +1379,6 @@ def generate(  # noqa: PLR0912, PLR0914, PLR0915
     skip_root_model = config.skip_root_model
     source_override: Mapping[str, Any] | None = None
 
-    input_ = _coerce_existing_path_input(input_)
-
     if (
         isinstance(input_, list)
         and input_file_type != InputFileType.MCPTools
@@ -1446,8 +1462,9 @@ def generate(  # noqa: PLR0912, PLR0914, PLR0915
             raise Error(msg) from exc
 
         try:
-            assert isinstance(input_text_, str)
-            input_file_type = infer_input_type(input_text_)
+            with _warn_on_input_string_path_failure(input_):
+                assert isinstance(input_text_, str)
+                input_file_type = infer_input_type(input_text_)
         except Exception as exc:
             raise InvalidFileFormatError(exc) from exc
         else:
@@ -1460,15 +1477,17 @@ def generate(  # noqa: PLR0912, PLR0914, PLR0915
             if isinstance(input_, Path) and input_.is_file() and input_file_type not in RAW_DATA_TYPES:
                 input_text = input_text_
 
-    input_text = _normalize_raw_input(input_, input_text, input_file_type, config)
+    with _warn_on_input_string_path_failure(input_):
+        input_text = _normalize_raw_input(input_, input_text, input_file_type, config)
 
     if input_file_type == InputFileType.MCPTools:
-        source_override, input_file_type, skip_root_model = _convert_mcp_tools(
-            input_,
-            input_text,
-            config,
-            remote_text_cache,
-        )
+        with _warn_on_input_string_path_failure(input_):
+            source_override, input_file_type, skip_root_model = _convert_mcp_tools(
+                input_,
+                input_text,
+                config,
+                remote_text_cache,
+            )
 
     if isinstance(input_, ParseResult) and input_file_type not in RAW_DATA_TYPES:
         input_text = None
@@ -1489,29 +1508,35 @@ def generate(  # noqa: PLR0912, PLR0914, PLR0915
         _resolve_schema_versions(input_file_type, config.schema_version)
     )
 
-    parser = _build_parser(
-        input_file_type,
-        source,
-        config,
-        additional_options,
-        data_model_types,
-        jsonschema_version=jsonschema_version,
-        openapi_version=openapi_version,
-        asyncapi_version=asyncapi_version,
-        xmlschema_version=xmlschema_version,
-        protobuf_version=protobuf_version,
-    )
+    with _warn_on_input_string_path_failure(input_):
+        parser = _build_parser(
+            input_file_type,
+            source,
+            config,
+            additional_options,
+            data_model_types,
+            jsonschema_version=jsonschema_version,
+            openapi_version=openapi_version,
+            asyncapi_version=asyncapi_version,
+            xmlschema_version=xmlschema_version,
+            protobuf_version=protobuf_version,
+        )
 
     with chdir(config.output):
         try:
-            results = parser.parse(
-                settings_path=config.settings_path,
-                disable_future_imports=config.disable_future_imports,
-                all_exports_scope=config.all_exports_scope,
-                all_exports_collision_strategy=config.all_exports_collision_strategy,
-                module_split_mode=config.module_split_mode,
-                collect_model_metadata=config.emit_model_metadata is not None,
-            )
+            with _warn_on_input_string_path_failure(input_):
+                results = parser.parse(
+                    settings_path=config.settings_path,
+                    disable_future_imports=config.disable_future_imports,
+                    all_exports_scope=config.all_exports_scope,
+                    all_exports_collision_strategy=config.all_exports_collision_strategy,
+                    module_split_mode=config.module_split_mode,
+                    collect_model_metadata=config.emit_model_metadata is not None,
+                )
+        except Exception:
+            with contextlib.suppress(BaseException):
+                parser._dispose()  # noqa: SLF001
+            raise
         except BaseException:
             with contextlib.suppress(BaseException):
                 parser._dispose()  # noqa: SLF001

…ted/main/generate_accepts_string_path.py …cted/main/generate_accepts_path_input.pytests/data/expected/main/generate_accepts_string_path.py renamed to tests/data/expected/main/generate_accepts_path_input.py tests/data/expected/main/generate_accepts_string_path.py renamed to tests/data/expected/main/generate_accepts_path_input.py

@@ -2,13 +2,17 @@
 #   filename:  person.json
 
 from __future__ import annotations
-from pydantic import BaseModel, Field, conint
+
 from typing import Any
 
+from pydantic import BaseModel, Field, conint
+
 
 class Person(BaseModel):
     firstName: str | None = Field(None, description="The person's first name.")
     lastName: str | None = Field(None, description="The person's last name.")
-    age: conint(ge=0) | None = Field(None, description='Age in years which must be equal to or greater than zero.')
+    age: conint(ge=0) | None = Field(
+        None, description='Age in years which must be equal to or greater than zero.'
+    )
     friends: list[Any] | None = None
     comment: None = Field(None)

…eps_unresolved_user_path_string_input.py …rate_keeps_existing_path_string_input.pytests/data/expected/main/generate_keeps_unresolved_user_path_string_input.py renamed to tests/data/expected/main/generate_keeps_existing_path_string_input.py tests/data/expected/main/generate_keeps_unresolved_user_path_string_input.py renamed to tests/data/expected/main/generate_keeps_existing_path_string_input.py

@@ -2,8 +2,9 @@
 #   filename:  inline.yaml
 
 from __future__ import annotations
+
 from pydantic import RootModel
 
 
 class Model(RootModel[str]):
-    root: str
+    root: str

tests/data/expected/main/generate_keeps_non_path_string_input.py

@@ -2,8 +2,9 @@
 #   filename:  inline.yaml
 
 from __future__ import annotations
+
 from pydantic import BaseModel
 
 
 class Model(BaseModel):
-    name: str
+    name: str