spec_validator.py

"""
spec_validator_v2.py

Strict JSON schema validator + generator for Django backend specs.
"""

import json
import re
from copy import deepcopy
from typing import Dict, Any, List, Tuple

from jsonschema import Draft202012Validator, ValidationError
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.output_parsers import JsonOutputParser


# =====================================================
# FILE LOADERS
# =====================================================

def load_schema(path: str) -> Dict[str, Any]:
    with open(path, "r", encoding="utf-8") as f:
        return json.load(f)


def load_json(path: str) -> Dict[str, Any]:
    with open(path, "r", encoding="utf-8") as f:
        return json.load(f)


# =====================================================
# VALIDATION HELPERS
# =====================================================

def _format_error(err: ValidationError) -> Dict[str, str]:
    path = "/" + "/".join(map(str, err.absolute_path)) if err.absolute_path else "/"
    return {"path": path, "message": err.message}


def validate_schema(
    spec: Dict[str, Any],
    schema: Dict[str, Any]
) -> List[Dict[str, str]]:
    validator = Draft202012Validator(schema)
    errors = sorted(validator.iter_errors(spec), key=lambda e: e.path)
    return [_format_error(e) for e in errors]


# =====================================================
# SAFE NORMALIZATIONS (SCHEMA-COMPATIBLE ONLY)
# =====================================================

def _to_pascal_case(name: str) -> str:
    return "".join(part.capitalize() for part in re.split(r"[_\\s-]+", name) if part)


def normalize_spec(spec: Dict[str, Any]) -> List[str]:
    """
    Normalize names without introducing new keys.
    """
    warnings = []

    # Normalize project name
    if "project_name" in spec:
        normalized = re.sub(r"[^a-z0-9_]", "_", spec["project_name"].lower())
        if normalized != spec["project_name"]:
            warnings.append("Normalized project_name")
            spec["project_name"] = normalized

    # Normalize model names
    models = spec.get("apps", {}).get("core", {}).get("models", {})
    new_models = {}

    for model_name, model_def in models.items():
        new_name = _to_pascal_case(model_name)
        if new_name != model_name:
            warnings.append(f"Renamed model '{model_name}' → '{new_name}'")
        new_models[new_name] = model_def

    if new_models:
        spec["apps"]["core"]["models"] = new_models

    return warnings


# =====================================================
# MAIN VALIDATION FUNCTION
# =====================================================

def validate_and_clean_spec(
    spec: Dict[str, Any],
    schema: Dict[str, Any],
    auto_fix: bool = True
) -> Tuple[bool, Dict[str, Any], List[Dict[str, str]], List[str]]:

    cleaned = deepcopy(spec)
    warnings: List[str] = []

    if auto_fix:
        warnings.extend(normalize_spec(cleaned))

    errors = validate_schema(cleaned, schema)

    if errors:
        return False, cleaned, errors, warnings

    return True, cleaned, [], warnings


# =====================================================
# JSON SPEC GENERATION (LLM)
# =====================================================

def generate_valid_json_spec(
    *,
    user_prompt: str,
    llm,
    retries: int = 3,
) -> Dict[str, Any]:
    """
    Generates schema-valid JSON using LLM + strict validation.
    """

    system_prompt = """
You are a JSON compiler.

Return ONLY a JSON object that EXACTLY matches this schema.
Do NOT add, remove, or rename keys.
Do NOT include explanations or formatting.

Schema:
{{
  "$schema": "http://json-schema.org/draft-07/schema#",
  "title": "Minimal Django Backend Generator Spec",
  "type": "object",

  "required": [
    "project_name",
    "database",
    "auth",
    "apps",
    "api_config"
  ],

  "properties": {{
    "project_name": {{
      "type": "string"
    }},

    "database": {{
      "type": "object",
      "required": ["engine", "name"],
      "properties": {{
        "engine": {{
          "type": "string",
          "enum": ["sqlite", "postgresql"]
        }},
        "name": {{
          "type": "string"
        }}
      }}
    }},

    "auth": {{
      "type": "object",
      "required": ["type"],
      "properties": {{
        "type": {{
          "type": "string",
          "enum": ["jwt", "session"]
        }}
      }}
    }},

    "apps": {{
      "type": "object",
      "required": ["core"],
      "properties": {{
        "core": {{
          "type": "object",
          "required": ["models", "apis"],
          "properties": {{
            "models": {{
              "type": "object",
              "patternProperties": {{
                "^[A-Z][a-zA-Z0-9]*$": {{
                  "type": "object",
                  "required": ["fields"],
                  "properties": {{
                    "fields": {{
                      "type": "object",
                      "patternProperties": {{
                        "^[a-z_][a-z0-9_]*$": {{
                          "type": "object",
                          "required": ["type"],
                          "properties": {{
                            "type": {{
                              "type": "string",
                              "enum": [
                                "CharField",
                                "TextField",
                                "EmailField",
                                "IntegerField",
                                "BooleanField",
                                "DateField",
                                "DateTimeField",
                                "ForeignKey"
                              ]
                            }},
                            "to": {{
                              "type": "string"
                            }}
                          }}
                        }}
                      }}
                    }}
                  }}
                }}
              }}
            }},

            "apis": {{
              "type": "object",
              "patternProperties": {{
                "^[A-Z][a-zA-Z0-9]*$": {{
                  "type": "array",
                  "items": {{
                    "type": "string",
                    "enum": ["list", "create", "retrieve", "update", "delete"]
                  }}
                }}
              }}
            }}
          }}
        }}
      }}
    }},

    "api_config": {{
      "type": "object",
      "required": ["base_url"],
      "properties": {{
        "base_url": {{
          "type": "string"
        }}
      }}
    }}
  }}
}}

"""

    prompt = ChatPromptTemplate.from_messages([
        ("system", system_prompt),
        ("user", "{user_prompt}")
    ])

# usage
    schema = load_schema('spec_schema.json')


    chain = prompt | llm | JsonOutputParser()

    for attempt in range(1, retries + 1):
        try:
            result = chain.invoke({"user_prompt": user_prompt})

            valid, cleaned, errors, _ = validate_and_clean_spec(
                result,
                schema,
                auto_fix=False
            )

            if valid:
                return cleaned

            raise ValueError(errors)

        except Exception as e:
            if attempt == retries:
                raise RuntimeError(f"Failed after {retries} attempts: {e}")

    raise RuntimeError("Unreachable")


# =====================================================
# CLI (OPTIONAL)
# =====================================================

if __name__ == "__main__":
    import argparse

    parser = argparse.ArgumentParser(description="Validate Django backend JSON spec")
    parser.add_argument("spec", help="Path to spec.json")
    parser.add_argument("schema", help="Path to schema.json")
    args = parser.parse_args()

    schema = load_schema(args.schema)
    spec = load_json(args.spec)

    valid, cleaned, errors, warnings = validate_and_clean_spec(spec, schema)

    print("VALID:", valid)
    if warnings:
        print("WARNINGS:")
        for w in warnings:
            print("-", w)

    if errors:
        print("ERRORS:")
        for e in errors:
            print(f"{e['path']}: {e['message']}")
    else:
        print("Spec is valid.")