app/src/schema.py

from typing import Any, Dict, Tuple

METRIC_KEYS = [
    "hierarchy_strength",
    "focal_clarity",
    "typography_control",
    "font_conflict",
    "palette_control",
    "color_conflict",
    "visual_clutter",
    "template_genericity",
    "originality",
    "technical_quality",
    "visual_impact",
]

REQUIRED_KEYS = {
    "content_present": bool,
    "hierarchy_strength": int,
    "focal_clarity": int,
    "typography_control": int,
    "font_conflict": int,
    "palette_control": int,
    "color_conflict": int,
    "visual_clutter": int,
    "template_genericity": int,
    "originality": int,
    "technical_quality": int,
    "visual_impact": int,
    "confidence": (float, int),
    "comment": str,
}


PRESENCE_SCHEMA = {
    "type": "object",
    "properties": {
        "content_present": {"type": "boolean"},
    },
    "required": ["content_present"],
    "additionalProperties": False,
}


def metric_schema(metric_name: str) -> Dict[str, Any]:
    return {
        "type": "object",
        "properties": {
            metric_name: {"type": "integer", "minimum": 1, "maximum": 5},
        },
        "required": [metric_name],
        "additionalProperties": False,
    }


FINAL_JUDGE_SCHEMA = {
    "type": "object",
    "properties": {
        "label": {"type": "string", "enum": ["bad", "medium", "good"]},
        "confidence": {"type": "number", "minimum": 0.0, "maximum": 1.0},
        "rationale": {"type": "string"},
    },
    "required": ["label", "confidence", "rationale"],
    "additionalProperties": False,
}


OUTPUT_SCHEMA = {
    "type": "object",
    "properties": {
        "content_present": {"type": "boolean"},
        "hierarchy_strength": {"type": "integer", "minimum": 1, "maximum": 5},
        "focal_clarity": {"type": "integer", "minimum": 1, "maximum": 5},
        "typography_control": {"type": "integer", "minimum": 1, "maximum": 5},
        "font_conflict": {"type": "integer", "minimum": 1, "maximum": 5},
        "palette_control": {"type": "integer", "minimum": 1, "maximum": 5},
        "color_conflict": {"type": "integer", "minimum": 1, "maximum": 5},
        "visual_clutter": {"type": "integer", "minimum": 1, "maximum": 5},
        "template_genericity": {"type": "integer", "minimum": 1, "maximum": 5},
        "originality": {"type": "integer", "minimum": 1, "maximum": 5},
        "technical_quality": {"type": "integer", "minimum": 1, "maximum": 5},
        "visual_impact": {"type": "integer", "minimum": 1, "maximum": 5},
        "confidence": {"type": "number", "minimum": 0.0, "maximum": 1.0},
        "comment": {"type": "string"},
    },
    "required": [
        "content_present",
        "hierarchy_strength",
        "focal_clarity",
        "typography_control",
        "font_conflict",
        "palette_control",
        "color_conflict",
        "visual_clutter",
        "template_genericity",
        "originality",
        "technical_quality",
        "visual_impact",
        "confidence",
        "comment",
    ],
    "additionalProperties": False,
}


def validate_scores(data: Dict[str, Any]) -> Tuple[bool, str]:
    for key, expected_type in REQUIRED_KEYS.items():
        if key not in data:
            return False, f"Missing key: {key}"

        if not isinstance(data[key], expected_type):
            return False, f"Invalid type for key: {key}"

    for key in METRIC_KEYS:
        value = data[key]
        if not isinstance(value, int):
            return False, f"{key} must be int"
        if value < 1 or value > 5:
            return False, f"{key} must be in range 1..5"

    confidence = float(data["confidence"])
    if confidence < 0.0 or confidence > 1.0:
        return False, "confidence must be in range 0.0..1.0"

    comment = data["comment"].strip()
    if not comment:
        return False, "comment must not be empty"

    return True, "ok"