Add --validators option for Pydantic v2 field validators

docs/cli-reference/index.md

@@ -12,7 +12,7 @@ This documentation is auto-generated from test cases.
 | 🔧 [Typing Customization](typing-customization.md) | 27 | Type annotation and import behavior |
 | 🏷️ [Field Customization](field-customization.md) | 24 | Field naming and docstring behavior |
 | 🏗️ [Model Customization](model-customization.md) | 39 | Model generation behavior |
-| 🎨 [Template Customization](template-customization.md) | 18 | Output formatting and custom rendering |
+| 🎨 [Template Customization](template-customization.md) | 19 | Output formatting and custom rendering |
 | 📘 [OpenAPI-only Options](openapi-only-options.md) | 7 | OpenAPI-specific features |
 | 📋 [GraphQL-only Options](graphql-only-options.md) | 1 |  |
 | ⚙️ [General Options](general-options.md) | 15 | Utilities and meta options |
@@ -219,6 +219,7 @@ This documentation is auto-generated from test cases.
 ### V {#v}
 
 - [`--validation`](openapi-only-options.md#validation)
+- [`--validators`](template-customization.md#validators)
 - [`--version`](utility-options.md#version)
 
 ### W {#w}

docs/cli-reference/quick-reference.md

@@ -150,6 +150,7 @@ datamodel-codegen [OPTIONS]
 | [`--treat-dot-as-module`](template-customization.md#treat-dot-as-module) | Treat dots in schema names as module separators. |
 | [`--use-double-quotes`](template-customization.md#use-double-quotes) | Use double quotes for string literals in generated code. |
 | [`--use-exact-imports`](template-customization.md#use-exact-imports) | Import exact types instead of modules. |
+| [`--validators`](template-customization.md#validators) | Add custom field validators to generated Pydantic v2 models. |
 | [`--wrap-string-literal`](template-customization.md#wrap-string-literal) | Wrap long string literals across multiple lines. |
 
 ### 📘 OpenAPI-only Options
@@ -347,6 +348,7 @@ All options sorted alphabetically:
 - [`--use-union-operator`](typing-customization.md#use-union-operator) - Use | operator for Union types (PEP 604).
 - [`--use-unique-items-as-set`](typing-customization.md#use-unique-items-as-set) - Generate set types for arrays with uniqueItems constraint.
 - [`--validation`](openapi-only-options.md#validation) - Enable validation constraints (deprecated, use --field-const...
+- [`--validators`](template-customization.md#validators) - Add custom field validators to generated Pydantic v2 models.
 - [`--version`](utility-options.md#version) - Show program version and exit
 - [`--watch`](general-options.md#watch) - Watch input file(s) for changes and regenerate output automa...
 - [`--watch-delay`](general-options.md#watch-delay) - Set debounce delay in seconds for watch mode.

docs/cli-reference/template-customization.md

@@ -21,6 +21,7 @@
 | [`--treat-dot-as-module`](#treat-dot-as-module) | Treat dots in schema names as module separators. |
 | [`--use-double-quotes`](#use-double-quotes) | Use double quotes for string literals in generated code. |
 | [`--use-exact-imports`](#use-exact-imports) | Import exact types instead of modules. |
+| [`--validators`](#validators) | Add custom field validators to generated Pydantic v2 models.... |
 | [`--wrap-string-literal`](#wrap-string-literal) | Wrap long string literals across multiple lines. |
 
 ---
@@ -2692,6 +2693,83 @@ modules are generated. For single-file output, the difference is minimal.
 
 ---
 
+## `--validators` {#validators}
+
+Add custom field validators to generated Pydantic v2 models.
+
+The `--validators` option takes a JSON file defining validators per model.
+Each validator specifies the field(s) to validate, the validation function
+to import, and optionally the mode (before/after/wrap/plain).
+This allows injecting custom validation logic into generated models.
+
+**See also:** [Field Validators](../validators.md)
+
+!!! tip "Usage"
+
+    ```bash
+    datamodel-codegen --input schema.json --validators tests/data/jsonschema/field_validators_config.json --output-model-type pydantic_v2.BaseModel --disable-timestamp # (1)!
+    ```
+
+    1. :material-arrow-left: `--validators` - the option documented here
+
+??? example "Examples"
+
+    **Input Schema:**
+
+    ```json
+    {
+      "$schema": "http://json-schema.org/draft-07/schema#",
+      "title": "User",
+      "type": "object",
+      "properties": {
+        "name": {
+          "type": "string"
+        },
+        "email": {
+          "type": "string",
+          "format": "email"
+        },
+        "age": {
+          "type": "integer",
+          "minimum": 0
+        }
+      },
+      "required": ["name", "email"]
+    }
+    ```
+
+    **Output:**
+
+    ```python
+    # generated by datamodel-codegen:
+    #   filename:  field_validators.json
+
+    from __future__ import annotations
+
+    from typing import Any
+
+    from myapp.validators import validate_email, validate_name
+    from pydantic import BaseModel, EmailStr, ValidationInfo, conint, field_validator
+
+
+    class User(BaseModel):
+        name: str
+        email: EmailStr
+        age: conint(ge=0) | None = None
+
+        @field_validator('name', mode='before')
+        @classmethod
+        def validate_name_validator(cls, v: Any, info: ValidationInfo) -> Any:
+            return validate_name(v, info)
+
+        @field_validator('email', mode='after')
+        @classmethod
+        def validate_email_validator(cls, v: Any, info: ValidationInfo) -> Any:
+            return validate_email(v, info)
+    ```
+
+---
+
 ## `--wrap-string-literal` {#wrap-string-literal}
 
 Wrap long string literals across multiple lines.

docs/validators.md

@@ -0,0 +1,199 @@
+<!-- related-cli-options: --validators -->
+
+# Field Validators
+
+The `--validators` option allows you to add custom field validators to generated Pydantic v2 models. This enables you to inject validation logic into generated code without manually editing it.
+
+## Basic Usage
+
+```bash
+datamodel-codegen --input schema.json --output model.py \
+  --validators validators.json \
+  --output-model-type pydantic_v2.BaseModel
+```
+
+## Validators File Format
+
+The validators file is a JSON file that maps model names to their validator definitions.
+
+### Structure
+
+```json
+{
+  "ModelName": {
+    "validators": [
+      {
+        "field": "field_name",
+        "function": "module.path.to.validator_function",
+        "mode": "after"
+      }
+    ]
+  }
+}
+```
+
+### Fields
+
+| Field | Description | Required |
+|-------|-------------|----------|
+| `field` | Single field name to validate | One of `field` or `fields` |
+| `fields` | List of field names (for multi-field validators) | One of `field` or `fields` |
+| `function` | Fully qualified path to the validator function | Yes |
+| `mode` | Validator mode: `before`, `after`, `wrap`, or `plain` | No (default: `after`) |
+
+## Validator Modes
+
+Pydantic v2 supports different validator modes, each with its own signature:
+
+### `before` / `after` Mode
+
+Standard validators that run before or after Pydantic's own validation:
+
+```python
+def validate_name(v: Any, info: ValidationInfo) -> Any:
+    if not v:
+        raise ValueError("Name cannot be empty")
+    return v.strip()
+```
+
+### `wrap` Mode
+
+Wrap validators receive a handler to call the next validator in the chain:
+
+```python
+from pydantic import ValidationInfo, ValidatorFunctionWrapHandler
+
+def wrap_validate_name(
+    v: Any,
+    handler: ValidatorFunctionWrapHandler,
+    info: ValidationInfo
+) -> Any:
+    # Pre-processing
+    v = v.strip() if isinstance(v, str) else v
+    # Call next validator
+    result = handler(v)
+    # Post-processing
+    return result.upper()
+```
+
+### `plain` Mode
+
+Plain validators replace Pydantic's validation entirely:
+
+```python
+def plain_validate_name(v: Any) -> str:
+    if not isinstance(v, str):
+        raise TypeError("Expected string")
+    return v
+```
+
+## Example
+
+### Input Schema
+
+```json
+{
+  "type": "object",
+  "title": "User",
+  "properties": {
+    "name": {"type": "string"},
+    "email": {"type": "string", "format": "email"},
+    "age": {"type": "integer", "minimum": 0}
+  },
+  "required": ["name", "email"]
+}
+```
+
+### Validators File
+
+```json
+{
+  "User": {
+    "validators": [
+      {
+        "field": "name",
+        "function": "myapp.validators.validate_name",
+        "mode": "before"
+      },
+      {
+        "field": "email",
+        "function": "myapp.validators.validate_email",
+        "mode": "after"
+      },
+      {
+        "fields": ["name", "email"],
+        "function": "myapp.validators.validate_contact_info",
+        "mode": "after"
+      }
+    ]
+  }
+}
+```
+
+### Validator Functions (myapp/validators.py)
+
+```python
+from typing import Any
+from pydantic import ValidationInfo
+
+def validate_name(v: Any, info: ValidationInfo) -> Any:
+    if isinstance(v, str):
+        return v.strip()
+    return v
+
+def validate_email(v: Any, info: ValidationInfo) -> Any:
+    if isinstance(v, str) and not v.endswith("@example.com"):
+        # Custom email domain validation
+        pass
+    return v
+
+def validate_contact_info(v: Any, info: ValidationInfo) -> Any:
+    # This runs for both name and email fields
+    return v
+```
+
+### Generated Output
+
+```python
+from __future__ import annotations
+
+from typing import Any
+
+from myapp.validators import validate_contact_info, validate_email, validate_name
+from pydantic import BaseModel, EmailStr, ValidationInfo, conint, field_validator
+
+
+class User(BaseModel):
+    name: str
+    email: EmailStr
+    age: conint(ge=0) | None = None
+
+    @field_validator('name', mode='before')
+    @classmethod
+    def validate_name_validator(cls, v: Any, info: ValidationInfo) -> Any:
+        return validate_name(v, info)
+
+    @field_validator('email', mode='after')
+    @classmethod
+    def validate_email_validator(cls, v: Any, info: ValidationInfo) -> Any:
+        return validate_email(v, info)
+
+    @field_validator('name', 'email', mode='after')
+    @classmethod
+    def validate_contact_info_validator(cls, v: Any, info: ValidationInfo) -> Any:
+        return validate_contact_info(v, info)
+```
+
+## Notes
+
+- This feature only supports Pydantic v2 (`--output-model-type pydantic_v2.BaseModel`)
+- The `ModelName` in the validators file must match the generated Python class name
+- Validator functions are imported automatically based on the `function` path
+- When the same validator function is used multiple times, an incrementing suffix (`_1`, `_2`, etc.) is added to ensure method name uniqueness
+
+---
+
+## See Also
+
+- [CLI Reference: `--validators`](cli-reference/general-options.md) - CLI option documentation
+- [Pydantic v2 Validators Documentation](https://docs.pydantic.dev/latest/concepts/validators/) - Official Pydantic documentation