from typing import Tuple, List
import numpy as np
from anndata import AnnData


class AnnDataValidator:
    """Validate AnnData objects for spatial visualization requirements"""

    MAX_OBS = 500_000  # Max number of observations (cells/spots)
    MAX_VARS = 50_000  # Max number of variables (genes)

    @staticmethod
    def validate(adata: AnnData) -> Tuple[bool, List[str]]:
        """
        Validate AnnData object for spatial visualization

        Args:
            adata: AnnData object to validate

        Returns:
            Tuple of (is_valid, error_messages)
        """
        errors = []

        # Check spatial coordinates exist
        if "spatial" not in adata.obsm:
            errors.append(
                "Missing spatial coordinates. adata.obsm['spatial'] is required."
            )

        # Validate spatial coordinates format
        if "spatial" in adata.obsm:
            spatial = adata.obsm["spatial"]
            if spatial.shape[1] != 2:
                errors.append(
                    f"Spatial coordinates must be 2D (x, y). Got shape: {spatial.shape}"
                )

        # Check number of observations
        if adata.n_obs > AnnDataValidator.MAX_OBS:
            errors.append(
                f"Too many observations: {adata.n_obs:,} (max: {AnnDataValidator.MAX_OBS:,})"
            )

        # Check number of variables
        if adata.n_vars > AnnDataValidator.MAX_VARS:
            errors.append(
                f"Too many variables: {adata.n_vars:,} (max: {AnnDataValidator.MAX_VARS:,})"
            )

        # Check if data is accessible
        try:
            _ = adata.var_names
        except Exception as e:
            errors.append(f"Cannot access variable names: {str(e)}")

        return (len(errors) == 0, errors)

    @staticmethod
    def validate_gene(adata: AnnData, gene_name: str) -> Tuple[bool, str]:
        """
        Validate if a gene exists in the dataset

        Args:
            adata: AnnData object
            gene_name: Gene name to check

        Returns:
            Tuple of (exists, message)
        """
        if gene_name not in adata.var_names:
            # Try to find similar gene names
            var_names = list(adata.var_names)
            similar = [g for g in var_names if gene_name.lower() in g.lower()][:5]

            if similar:
                return (
                    False,
                    f"Gene '{gene_name}' not found. Similar genes: {', '.join(similar)}",
                )
            else:
                return (False, f"Gene '{gene_name}' not found in dataset.")

        return (True, f"Gene '{gene_name}' found.")

    @staticmethod
    def get_gene_expression(adata: AnnData, gene_name: str) -> np.ndarray:
        """
        Extract gene expression for a specific gene

        Args:
            adata: AnnData object
            gene_name: Gene name to extract

        Returns:
            Expression vector as numpy array

        Raises:
            ValueError: If gene not found
        """
        is_valid, message = AnnDataValidator.validate_gene(adata, gene_name)
        if not is_valid:
            raise ValueError(message)

        # Extract gene expression (works with backed mode)
        gene_data = adata[:, gene_name].X

        # Convert to dense array if sparse
        if hasattr(gene_data, "toarray"):
            gene_data = gene_data.toarray()

        # Flatten if needed
        if gene_data.ndim > 1:
            gene_data = gene_data.flatten()

        return gene_data

    @staticmethod
    def get_gene_list(adata: AnnData, limit: int = 1000) -> List[str]:
        """
        Get list of available genes (limited for performance)

        Args:
            adata: AnnData object
            limit: Maximum number of genes to return

        Returns:
            List of gene names
        """
        var_names = list(adata.var_names)
        return var_names[:limit]