from __future__ import annotations

from typing import Any
from uuid import UUID

from geoalchemy2.shape import to_shape
from shapely.geometry import mapping
from sqlalchemy import or_
from sqlalchemy.orm import Session

from app.core.errors import AppError
from app.models import Detection, DetectionReview, QualityCheck, Segmentation, VectorFeature


class QualityEvidenceService:
    @staticmethod
    def evidence_geojson(db: Session, *, project_id: UUID, quality_check_id: UUID) -> dict[str, Any]:
        quality_check = db.get(QualityCheck, quality_check_id)
        if not quality_check or quality_check.project_id != project_id:
            raise AppError(code="QUALITY_CHECK_NOT_FOUND", message="Quality check not found", status_code=404)

        findings = quality_check.findings_json or {}
        features: list[dict[str, Any]] = []
        warnings: list[str] = []
        candidate_ids, reference_ids = QualityEvidenceService._evidence_identifiers(findings)
        candidate_index = QualityEvidenceService._candidate_feature_index(db, quality_check, candidate_ids)
        reference_index = QualityEvidenceService._reference_feature_index(db, quality_check, reference_ids)

        for evidence in QualityEvidenceService._evidence_items(findings.get("match_evidence")):
            candidate_id = QualityEvidenceService._string_value(evidence.get("candidate_feature_id"))
            reference_id = QualityEvidenceService._string_value(evidence.get("reference_feature_id"))
            iou = evidence.get("iou")
            if candidate_id:
                row = candidate_index.get(candidate_id)
                if row is not None:
                    features.append(
                        QualityEvidenceService._row_to_feature(
                            row,
                            role="match_candidate",
                            quality_check=quality_check,
                            evidence=evidence,
                        )
                    )
                else:
                    warnings.append(f"Candidate evidence feature not found: {candidate_id}")
            if reference_id:
                row = reference_index.get(reference_id)
                if row is not None:
                    features.append(
                        QualityEvidenceService._row_to_feature(
                            row,
                            role="match_reference",
                            quality_check=quality_check,
                            evidence={"candidate_feature_id": candidate_id, "reference_feature_id": reference_id, "iou": iou},
                        )
                    )
                else:
                    warnings.append(f"Reference evidence feature not found: {reference_id}")

        for evidence in QualityEvidenceService._evidence_items(findings.get("false_positive_evidence")):
            candidate_id = QualityEvidenceService._string_value(evidence.get("candidate_feature_id"))
            if not candidate_id:
                continue
            row = candidate_index.get(candidate_id)
            if row is not None:
                features.append(
                    QualityEvidenceService._row_to_feature(
                        row,
                        role="false_positive",
                        quality_check=quality_check,
                        evidence=evidence,
                    )
                )
            else:
                warnings.append(f"False-positive evidence feature not found: {candidate_id}")

        for evidence in QualityEvidenceService._evidence_items(findings.get("false_negative_evidence")):
            reference_id = QualityEvidenceService._string_value(evidence.get("reference_feature_id"))
            if not reference_id:
                continue
            row = reference_index.get(reference_id)
            if row is not None:
                features.append(
                    QualityEvidenceService._row_to_feature(
                        row,
                        role="false_negative",
                        quality_check=quality_check,
                        evidence=evidence,
                    )
                )
            else:
                warnings.append(f"False-negative evidence feature not found: {reference_id}")

        QualityEvidenceService._annotate_reviews(db, quality_check, features)

        return {
            "quality_check_id": str(quality_check.id),
            "project_id": str(quality_check.project_id),
            "candidate_dataset_id": str(quality_check.candidate_dataset_id) if quality_check.candidate_dataset_id else None,
            "reference_dataset_id": str(quality_check.reference_dataset_id),
            "analysis_run_id": str(quality_check.analysis_run_id) if quality_check.analysis_run_id else None,
            "feature_count": len(features),
            "warnings": warnings,
            "geojson": {
                "type": "FeatureCollection",
                "features": features,
            },
        }

    @staticmethod
    def _evidence_items(value: Any) -> list[dict[str, Any]]:
        if not isinstance(value, list):
            return []
        return [item for item in value if isinstance(item, dict)]

    @staticmethod
    def _string_value(value: Any) -> str | None:
        if value is None:
            return None
        text = str(value).strip()
        return text or None

    @staticmethod
    def _evidence_identifiers(findings: dict[str, Any]) -> tuple[set[str], set[str]]:
        candidate_ids: set[str] = set()
        reference_ids: set[str] = set()
        for evidence in QualityEvidenceService._evidence_items(findings.get("match_evidence")):
            candidate_id = QualityEvidenceService._string_value(evidence.get("candidate_feature_id"))
            reference_id = QualityEvidenceService._string_value(evidence.get("reference_feature_id"))
            if candidate_id:
                candidate_ids.add(candidate_id)
            if reference_id:
                reference_ids.add(reference_id)
        for evidence in QualityEvidenceService._evidence_items(findings.get("false_positive_evidence")):
            candidate_id = QualityEvidenceService._string_value(evidence.get("candidate_feature_id"))
            if candidate_id:
                candidate_ids.add(candidate_id)
        for evidence in QualityEvidenceService._evidence_items(findings.get("false_negative_evidence")):
            reference_id = QualityEvidenceService._string_value(evidence.get("reference_feature_id"))
            if reference_id:
                reference_ids.add(reference_id)
        return candidate_ids, reference_ids

    @staticmethod
    def _uuid_identifiers(identifiers: set[str]) -> list[UUID]:
        values: list[UUID] = []
        for identifier in identifiers:
            try:
                values.append(UUID(identifier))
            except (TypeError, ValueError):
                continue
        return values

    @staticmethod
    def _vector_feature_rows(db: Session, dataset_id: UUID, identifiers: set[str]) -> list[VectorFeature]:
        if not identifiers:
            return []
        conditions = [VectorFeature.source_feature_id.in_(identifiers)]
        uuid_identifiers = QualityEvidenceService._uuid_identifiers(identifiers)
        if uuid_identifiers:
            conditions.append(VectorFeature.id.in_(uuid_identifiers))
        return (
            db.query(VectorFeature)
            .filter(VectorFeature.dataset_id == dataset_id, or_(*conditions))
            .all()
        )

    @staticmethod
    def _candidate_feature_index(
        db: Session,
        quality_check: QualityCheck,
        identifiers: set[str],
    ) -> dict[str, Any]:
        index: dict[str, Any] = {}
        if not identifiers:
            return index
        uuid_identifiers = QualityEvidenceService._uuid_identifiers(identifiers)
        if quality_check.candidate_dataset_id:
            for row in QualityEvidenceService._vector_feature_rows(db, quality_check.candidate_dataset_id, identifiers):
                QualityEvidenceService._add_index_keys(index, row)
            if uuid_identifiers:
                for row in db.query(Detection).filter(
                    Detection.dataset_id == quality_check.candidate_dataset_id,
                    Detection.id.in_(uuid_identifiers),
                ).all():
                    QualityEvidenceService._add_index_keys(index, row)
                for row in db.query(Segmentation).filter(
                    Segmentation.dataset_id == quality_check.candidate_dataset_id,
                    Segmentation.id.in_(uuid_identifiers),
                ).all():
                    QualityEvidenceService._add_index_keys(index, row)
        if quality_check.analysis_run_id and uuid_identifiers:
            for row in db.query(Detection).filter(
                Detection.analysis_run_id == quality_check.analysis_run_id,
                Detection.id.in_(uuid_identifiers),
            ).all():
                QualityEvidenceService._add_index_keys(index, row)
            for row in db.query(Segmentation).filter(
                Segmentation.analysis_run_id == quality_check.analysis_run_id,
                Segmentation.id.in_(uuid_identifiers),
            ).all():
                QualityEvidenceService._add_index_keys(index, row)
        return index

    @staticmethod
    def _reference_feature_index(
        db: Session,
        quality_check: QualityCheck,
        identifiers: set[str],
    ) -> dict[str, Any]:
        index: dict[str, Any] = {}
        for row in QualityEvidenceService._vector_feature_rows(db, quality_check.reference_dataset_id, identifiers):
            QualityEvidenceService._add_index_keys(index, row)
        return index

    @staticmethod
    def _annotate_reviews(
        db: Session,
        quality_check: QualityCheck,
        features: list[dict[str, Any]],
    ) -> None:
        if quality_check.check_type != "detections_vs_reference":
            return
        reviews = db.query(DetectionReview).filter(DetectionReview.quality_check_id == quality_check.id).all()
        review_index = {(row.evidence_role, row.evidence_feature_id): row for row in reviews}
        for feature in features:
            properties = feature.get("properties")
            if not isinstance(properties, dict):
                continue
            role = QualityEvidenceService._string_value(properties.get("qa_evidence_role"))
            if role == "false_positive":
                evidence_id = QualityEvidenceService._string_value(properties.get("candidate_feature_id"))
            elif role == "false_negative":
                evidence_id = QualityEvidenceService._string_value(properties.get("reference_feature_id"))
            else:
                continue
            review = review_index.get((role, evidence_id or ""))
            properties.update(
                {
                    "review_decision": review.decision if review else "unreviewed",
                    "review_notes": review.notes if review else None,
                    "reviewed_by": review.reviewed_by if review else None,
                    "reviewed_at": review.updated_at.isoformat() if review and review.updated_at else None,
                }
            )

    @staticmethod
    def _add_index_keys(index: dict[str, Any], row: Any) -> None:
        for key in QualityEvidenceService._row_identifiers(row):
            index.setdefault(key, row)

    @staticmethod
    def _row_identifiers(row: Any) -> set[str]:
        identifiers = {str(row.id)}
        source_feature_id = getattr(row, "source_feature_id", None)
        if source_feature_id:
            identifiers.add(str(source_feature_id))
        properties = getattr(row, "properties_json", None) or {}
        if isinstance(properties, dict):
            for property_key in ("vector_feature_id", "source_feature_id", "detection_id", "segmentation_id", "id", "name"):
                value = properties.get(property_key)
                if value is not None:
                    identifiers.add(str(value))
        return identifiers

    @staticmethod
    def _row_to_feature(row: Any, *, role: str, quality_check: QualityCheck, evidence: dict[str, Any]) -> dict[str, Any]:
        try:
            geometry = to_shape(row.geometry)
        except Exception as exc:
            raise AppError(
                code="INVALID_QA_EVIDENCE_GEOMETRY",
                message="Persisted QA evidence geometry could not be converted to GeoJSON",
                details={"feature_id": str(getattr(row, "id", ""))},
                status_code=500,
            ) from exc

        properties = dict(getattr(row, "properties_json", None) or {})
        properties.update(
            {
                "qa_evidence_role": role,
                "quality_check_id": str(quality_check.id),
                "project_id": str(quality_check.project_id),
                "candidate_dataset_id": str(quality_check.candidate_dataset_id) if quality_check.candidate_dataset_id else None,
                "reference_dataset_id": str(quality_check.reference_dataset_id),
                "analysis_run_id": str(quality_check.analysis_run_id) if quality_check.analysis_run_id else None,
                "feature_id": str(row.id),
                "dataset_id": str(getattr(row, "dataset_id", "")) if getattr(row, "dataset_id", None) else None,
                "source_feature_id": getattr(row, "source_feature_id", None),
                "feature_class": getattr(row, "feature_class", None) or getattr(row, "class_name", None),
                "candidate_feature_id": QualityEvidenceService._string_value(evidence.get("candidate_feature_id")),
                "reference_feature_id": QualityEvidenceService._string_value(evidence.get("reference_feature_id")),
                "iou": evidence.get("iou"),
            }
        )
        properties.update(QualityEvidenceService._row_provenance(row))

        return {
            "type": "Feature",
            "id": f"{role}:{row.id}",
            "geometry": mapping(geometry),
            "properties": properties,
        }

    @staticmethod
    def _row_provenance(row: Any) -> dict[str, Any]:
        if isinstance(row, Detection):
            return {
                "detection_id": str(row.id),
                "job_id": str(row.job_id) if row.job_id else None,
                "confidence": row.confidence,
                "model_name": row.model_name,
                "model_version": row.model_version,
                "source_tile_path": row.source_tile_path,
                "bbox_json": row.bbox_json,
            }
        if isinstance(row, Segmentation):
            return {
                "segmentation_id": str(row.id),
                "job_id": str(row.job_id) if row.job_id else None,
                "confidence": row.confidence,
                "model_name": row.model_name,
                "model_version": row.model_version,
                "source_tile_path": row.source_tile_path,
                "bbox_json": row.bbox_json,
                "mask_path": row.mask_path,
                "area_m2": row.area_m2,
            }
        return {}