import os
import re

from functools import wraps

import torch
import torch_npu
from torch_npu.utils._error_code import ErrCode, pta_error
from .unsupport_api import unsupported_Tensor_api, unsupported_nn_api
from .collect_env import get_cann_version


cann_pytorch_version_map = {
    "6.3.RC2": ["1.8.1.post2", "1.11.0.post1", "2.0.0.rc1"],
    "6.3.RC1": ["1.8.1.post1", "1.11.0"],
    "6.1.RC1": ["1.8.1.post1", "1.11.0"],
    "6.0.1": ["1.8.1", "1.11.0.rc2"],
    "6.0.RC1": ["1.8.1", "1.11.0.rc1"]
}


__all__ = []


def _cann_package_check():
    if "ASCEND_HOME_PATH" in os.environ:
        ascend_home_path = os.environ["ASCEND_HOME_PATH"]
        if not os.path.exists(ascend_home_path):
            raise Exception(f"ASCEND_HOME_PATH : {ascend_home_path} does not exist. "
                            "Please run 'source set_env.sh' in the CANN installation path." +
                            pta_error(ErrCode.NOT_FOUND))

        # check whether environment variables are correctly configured
        if "ASCEND_OPP_PATH" not in os.environ:
            raise Exception(f"ASCEND_OPP_PATH environment variable is not set. "
                            "Please check whether the opp package has been installed. If exist, please run "
                            "'source set_env.sh' in the CANN installation path." +
                            pta_error(ErrCode.NOT_FOUND))

        ascend_opp_path = os.environ["ASCEND_OPP_PATH"]
        if not os.path.exists(ascend_opp_path):
            raise Exception(f"ASCEND_OPP_PATH : {ascend_opp_path} does not exist. "
                            "Please check whether the opp package has been installed. If exist, please run "
                            "'source set_env.sh' in the CANN installation path." +
                            pta_error(ErrCode.NOT_FOUND))

        ascend_runtime_path = os.path.join(ascend_home_path, "runtime")
        if not os.path.exists(ascend_runtime_path):
            raise Exception(f"ASCEND_RUNTIME_PATH : {ascend_runtime_path} does not exist. "
                            "Please check whether the runtime package has been installed. If exist, please run "
                            "'source set_env.sh' in the CANN installation path." +
                            pta_error(ErrCode.NOT_FOUND))

        ascend_compiler_path = os.path.join(ascend_home_path, "compiler")
        if not os.path.exists(ascend_compiler_path):
            raise Exception(f"ASCEND_COMPILER_PATH : {ascend_compiler_path} does not exist. "
                            "Please check whether the compiler package has been installed. If exist, please run "
                            "'source set_env.sh' in the CANN installation path." +
                            pta_error(ErrCode.NOT_FOUND))

        # get the cann version
        cann_version = get_cann_version()

        # check whether the CANN package version matches the pytorch version
        if cann_version in cann_pytorch_version_map and \
                torch_npu.__version__ not in cann_pytorch_version_map[cann_version]:
            print(f"Warning : CANN package version {cann_version} and PyTorch version {torch_npu.__version__} "
                  "is not matched, please check the README of the ascend pytorch repo.")
    else:
        print(f"Warning : ASCEND_HOME_PATH environment variable is not set.")


def _create_wrap_func(check_func):
    def decorator(func):
        @wraps(func)
        def wrapper(*args, **kwargs):
            if check_func(*args, **kwargs):
                raise RuntimeError(f"{str(func)} is not supported in npu." + pta_error(ErrCode.NOT_SUPPORT))

            return func(*args, **kwargs)
        return wrapper
    return decorator


# Specific check functions
def _is_tensor_npu_supported(*args, **kwargs):
    return torch.is_tensor(args[0]) and args[0].is_npu


def _is_module_parameters_supported(*args, **kwargs):
    module_args = [m for m in args if isinstance(m, torch.nn.Module) and hasattr(m, "_modules")]
    module_parameters = [p for _, p in module_args[0].named_parameters()]
    return any(p.device is not None and p.device.type == "npu" for p in module_parameters)


def _apply_wrap_func_to_modules(wrap_func, unsupported_modules):
    for attr_name, parent_module in unsupported_modules.items():
        setattr(parent_module, attr_name, wrap_func(getattr(parent_module, attr_name)))


# Apply wrap functions to specific modules
def _add_intercept_methods():
    _apply_wrap_func_to_modules(_create_wrap_func(_is_tensor_npu_supported), unsupported_Tensor_api)
    _apply_wrap_func_to_modules(_create_wrap_func(_is_module_parameters_supported), unsupported_nn_api)