def retrieve_model_point(data_code: str,
parameter: str,
start_time: pd.Timestamp,
forecast_time: Union[str, pd.Timedelta, Tuple],
level_type: Union[str, int],
level: Union[int, float],
point: Union[Tuple[float, float],
List[Tuple[float, float]]] = None,
station: Union[List[str], str] = None,
config_file: Optional[str] = None) -> pd.DataFrame:
interface_config = {
"name": "getNafpEle",
"point": None,
"time": "Time",
"level": "Level",
"valid_time": "Validtime",
"station": None,
}
params = {
"dataCode": data_code,
"fcstEle": parameter,
"levelType": str(level_type),
"fcstLevel": str(level)
}
time = _get_time_string(start_time)
params["time"] = time
def _get_valid_time(f):
if isinstance(f, str):
f = pd.to_timedelta(f)
return int(f / pd.Timedelta(hours=1))
if isinstance(forecast_time, str or pd.Timedelta):
valid_time = _get_valid_time(forecast_time)
params["validTime"] = str(valid_time)
interface_config["valid_time"] = "Validtime"
elif isinstance(forecast_time, Tuple):
interface_config["valid_time"] = "ValidtimeRange"
params["minVT"] = str(_get_valid_time(forecast_time[0]))
params["maxVT"] = str(_get_valid_time(forecast_time[1]))
if point is not None:
_get_point_params(point, params, interface_config)
if station is not None:
_get_station_params(station, params, interface_config)
interface_id = _get_interface_id(interface_config)
logger.info(f"interface_id: {interface_id}")
client = _get_client(config_file)
result = client.callAPI_to_array2D(interface_id, params)
if result.request.error_code != 0:
logger.warning(
f"request error {result.request.error_code}: {result.request.error_message}"
)
df = result.to_pandas()
return df
def _do_request(
self,
interface_id: str,
method: str,
params: Dict,
server_id: str,
success_handler: Callable[[bytes], Any],
failure_handler: Callable[[bytes], Any],
exception_handler: Callable[[Exception], Any],
):
fetch_url = self._get_fetch_url(
interface_id, method, params, server_id
)
logger.info(f"fetch url: {fetch_url}")
result = self._connection.make_request(
fetch_url,
success_handler,
failure_handler,
exception_handler
)
return result
def download_obs_file(
data_code: str,
elements: str = None,
time: typing.Union[pd.Interval, pd.Timestamp, typing.List,
pd.Timedelta] = None,
station: typing.Union[str, typing.List, typing.Tuple] = None,
region=None,
station_level: typing.Union[str, typing.List[str]] = None,
order: str = None,
count: int = None,
output_dir: str = "./",
interface_data_type="Surf",
config_file: typing.Union[str, Path] = None,
**kwargs,
):
"""
下载地面观测数据文件
默认对应 CMADaaS 中以 `getSurfFile` 开头的一系列数据接口
**区域筛选条件**
- 经纬度范围
.. code-block:: python
{
"type": "rect",
"start_longitude": 115.7,
"end_longitude": 117.4,
"start_latitude": 41.6,
"end_latitude": 39.4,
}
- 流域
.. code-block:: python
{
"type": "basin",
"basin_codes": "CJLY"
}
- 地区
.. code-block:: python
{
"type": "region",
"admin_codes": "110000"
}
Parameters
----------
data_code:
数据种类，即 CMADaaS 中的资料代码
elements:
要素字段代码，以逗号分隔
time:
时间筛选条件，支持单个时间，时间列表，时间段和时间间隔
- 时间对象：``pd.Timestamp`` 类型，单个时间点，对应接口的 times 参数
- 时间列表：``typing.List[pd.Timestamp]`` 类型，多个时间列表，对应接口的 times 参数
- 时间段：``pd.Interval`` 类型，起止时间，定义区间端点是否闭合，对应接口的 timeRange 参数
- 时间间隔：``pd.Timedelta`` 类型，用于获取地面资料最新时次 (getSurfLatestTime)，忽略其余筛选条件
station:
站点筛选条件，支持字符串，列表和元组
- 字符串：单个站点
- 列表：多个站点
- 元组：站点范围，二元组，第一个元素是起止站号 (minStaId)，第二个元素是终止站号 (maxStaId)
region:
区域筛选条件：
- 经纬度范围 (rect)
- 流域 (basin)
- 地区 (region)
station_level:
台站级别：
- 011: 国家基准气候站
- 012: 基本气象站
- 013: 一般气象站
order:
排序字段
count:
最大返回记录数，对应接口的 limitCnt 参数
output_dir:
保存文件的目录
interface_data_type:
资料类型
config_file:
配置文件路径
kwargs:
其他需要传递给 MUSIC 接口的参数，例如：
- eleValueRanges: 要素值范围
- hourSeparate: 小时取整条件
- minSeparate: 分钟取整条件
Returns
-------
"""
# if elements is None:
# elements = STATION_DATASETS[data_code]["elements"]
interface_config = {
"name": f"get{interface_data_type}File",
"region": None,
"time": None,
"station": None,
}
params = {
"dataCode": data_code,
}
if order is not None:
params["orderby"] = order
if elements is not None:
params["elements"] = elements
if count is not None:
params["limitCnt"] = count
if isinstance(time, pd.Interval):
interface_config["time"] = "TimeRange"
params["timeRange"] = _get_time_range_string(time)
elif isinstance(time, pd.Timestamp):
interface_config["time"] = "Time"
params["times"] = _get_time_string(time)
elif isinstance(time, typing.List):
interface_config["time"] = "Time"
params["times"] = ",".join([_get_time_string(t) for t in time])
elif isinstance(time, pd.Timedelta):
interface_config["name"] = "getSurfLatestTime"
params["latestTime"] = str(int(time / pd.to_timedelta("1h")))
del params["orderby"]
del params["elements"]
if isinstance(station, str or int):
interface_config["station"] = "StaID"
params["staIds"] = station
elif isinstance(station, typing.List):
interface_config["station"] = "StaID"
params["staIds"] = ",".join(station)
elif isinstance(station, typing.Tuple):
interface_config["station"] = "StaIdRange"
params["minStaId"] = station[0]
params["maxStaId"] = station[1]
if region is not None:
_get_region_params(region, params, interface_config)
if station_level is not None:
del params["orderby"]
if isinstance(station_level, str):
params["staLevels"] = station_level
if interface_config["station"] is None:
interface_config["station"] = "StaLevels"
elif isinstance(station_level, typing.List):
params["staLevels"] = ",".join(station_level)
if interface_config["station"] is None:
interface_config["station"] = "StaLevels"
params.update(**kwargs)
interface_id = _get_interface_id(interface_config)
logger.info(f"interface_id: {interface_id}")
client = _get_client(config_file)
result = client.callAPI_to_downFile(interface_id,
params,
file_dir=output_dir)
if result.request.error_code != 0:
logger.warning(
f"request error {result.request.error_code}: {result.request.error_message}"
)
files_info = result.files_info
file_list = []
for f in files_info:
file_list.append(Path(output_dir, f.file_name))
return file_list
def retrieve_obs_station(
data_code: str = "SURF_CHN_MUL_HOR",
elements: str = None,
time: typing.Union[pd.Interval, pd.Timestamp, typing.List, pd.Timedelta] = None,
station: typing.Union[str, typing.List, typing.Tuple] = None,
region=None,
station_level: typing.Union[str, typing.List[str]] = None,
order: str = "Station_ID_d:asc",
count: int = None,
config_file: typing.Union[str, Path] = None,
**kwargs,
) -> pd.DataFrame:
"""
检索地面站点观测数据资料。
对应 CMADaaS 中以 `getSurfEle` 开头的一系列数据接口
**区域筛选条件**
- 经纬度范围
.. code-block:: python
{
"type": "rect",
"start_longitude": 115.7,
"end_longitude": 117.4,
"start_latitude": 41.6,
"end_latitude": 39.4,
}
- 流域
.. code-block:: python
{
"type": "basin",
"basin_codes": "CJLY"
}
- 地区
.. code-block:: python
{
"type": "region",
"admin_codes": "110000"
}
Parameters
----------
data_code:
数据种类，即 CMADaaS 中的资料代码
elements:
要素字段代码，以逗号分隔
time:
时间筛选条件，支持单个时间，时间列表，时间段和时间间隔
- 时间对象：``pd.Timestamp`` 类型，单个时间点，对应接口的 times 参数
- 时间列表：``typing.List[pd.Timestamp]`` 类型，多个时间列表，对应接口的 times 参数
- 时间段：``pd.Interval`` 类型，起止时间，定义区间端点是否闭合，对应接口的 timeRange 参数
- 时间间隔：``pd.Timedelta`` 类型，用于获取地面资料最新时次 (getSurfLatestTime)，忽略其余筛选条件
station:
站点筛选条件，支持字符串，列表和元组
- 字符串：单个站点
- 列表：多个站点
- 元组：站点范围，二元组，第一个元素是起止站号 (minStaId)，第二个元素是终止站号 (maxStaId)
region:
区域筛选条件：
- 经纬度范围 (rect)
- 流域 (basin)
- 地区 (region)
station_level:
台站级别：
- 011: 国家基准气候站
- 012: 基本气象站
- 013: 一般气象站
order:
排序字段
count:
最大返回记录数，对应接口的 limitCnt 参数
config_file:
配置文件路径
kwargs:
其他需要传递给 MUSIC 接口的参数，例如：
- eleValueRanges: 要素值范围
- hourSeparate: 小时取整条件
- minSeparate: 分钟取整条件
Returns
-------
pd.DataFrame
站点观测资料表格数据，列名为 elements 中的值
"""
station_dataset_config = load_dataset_config("station")
if elements is None:
elements = station_dataset_config[data_code]["elements"]
interface_config = {
"name": "getSurfEle",
"region": None,
"time": None,
"station": None,
}
params = {
"dataCode": data_code,
"elements": elements,
"orderby": order,
}
if count is not None:
params["limitCnt"] = count
if isinstance(time, pd.Interval):
interface_config["time"] = "TimeRange"
params["timeRange"] = _get_time_range_string(time)
elif isinstance(time, pd.Timestamp):
interface_config["time"] = "Time"
params["times"] = _get_time_string(time)
elif isinstance(time, typing.List):
interface_config["time"] = "Time"
params["times"] = ",".join([_get_time_string(t) for t in time])
elif isinstance(time, pd.Timedelta):
interface_config["name"] = "getSurfLatestTime"
params["latestTime"] = str(int(time / pd.to_timedelta("1h")))
del params["orderby"]
del params["elements"]
if isinstance(station, str or int):
interface_config["station"] = "StaID"
params["staIds"] = station
elif isinstance(station, typing.List):
interface_config["station"] = "StaID"
params["staIds"] = ",".join(station)
elif isinstance(station, typing.Tuple):
interface_config["station"] = "StaIdRange"
params["minStaId"] = station[0]
params["maxStaId"] = station[1]
if region is not None:
_get_region_params(region, params, interface_config)
if station_level is not None:
del params["orderby"]
if isinstance(station_level, str):
params["staLevels"] = station_level
if interface_config["station"] is None:
interface_config["station"] = "StaLevels"
elif isinstance(station_level, typing.List):
params["staLevels"] = ",".join(station_level)
if interface_config["station"] is None:
interface_config["station"] = "StaLevels"
params.update(**kwargs)
interface_id = _get_interface_id(interface_config)
logger.info(f"interface_id: {interface_id}")
client = _get_client(config_file)
result = client.callAPI_to_array2D(interface_id, params)
if result.request.error_code != 0:
logger.warning(f"request error {result.request.error_code}: {result.request.error_message}")
df = result.to_pandas()
return df
def retrieve_obs_upper_air(
data_code: str = "UPAR_GLB_MUL_FTM",
elements: str = None,
time: typing.Union[pd.Interval, pd.Timestamp, typing.List,
pd.Timedelta] = None,
level_type: typing.Union[str, typing.Tuple[str]] = None,
level: typing.Union[float, int, typing.List[typing.Union[float, int]],
typing.Tuple] = None,
station: typing.Union[str, int, typing.List, typing.Tuple] = None,
region=None,
station_level: typing.Union[str, typing.List[str]] = None,
order: str = None,
count: int = None,
interface_data_type: str = None,
config_file: typing.Union[str, Path] = None,
**kwargs,
) -> pd.DataFrame:
"""
检索高空观测数据资料。
对应 CMADaaS 中以 `getUparEle` 开头的一系列数据接口
**区域筛选条件**
- 经纬度范围
.. code-block:: python
{
"type": "rect",
"start_longitude": 115.7,
"end_longitude": 117.4,
"start_latitude": 41.6,
"end_latitude": 39.4,
}
- 流域
.. code-block:: python
{
"type": "basin",
"basin_codes": "CJLY"
}
- 地区
.. code-block:: python
{
"type": "region",
"admin_codes": "110000"
}
Parameters
----------
data_code:
数据种类，即 CMADaaS 中的资料代码
elements:
要素字段代码，以逗号分隔
time:
时间筛选条件，支持单个时间，时间列表，时间段和时间间隔
- 时间对象：``pd.Timestamp`` 类型，单个时间点，对应接口的 times 参数
- 时间列表：``typing.List[pd.Timestamp]`` 类型，多个时间列表，对应接口的 times 参数
- 时间段：``pd.Interval`` 类型，起止时间，定义区间端点是否闭合，对应接口的 timeRange 参数
- 时间间隔：``pd.Timedelta`` 类型，用于获取地面资料最新时次 (getSurfLatestTime)，忽略其余筛选条件
level_type:
层次类型
- pl: 气压层
- gh/hgt: 位势高度
- vertical: 垂直探测仪一，例如 4096 表示湿度特征层
- fl/flight_level: 飞行高度
level:
层次
station:
站点筛选条件，支持字符串，列表和元组
- 字符串：单个站点
- 列表：多个站点
- 元组：站点范围，二元组，第一个元素是起止站号 (minStaId)，第二个元素是终止站号 (maxStaId)
region:
区域筛选条件：
- 经纬度范围 (rect)
- 流域 (basin)
- 地区 (region)
station_level:
台站级别：
- 011: 国家基准气候站
- 012: 基本气象站
- 013: 一般气象站
order:
排序字段
count:
最大返回记录数，对应接口的 limitCnt 参数
interface_data_type:
资料类型，默认自动生成，或使用 _dataset.py 文件数据集配置的 interface_data_type 字段
config_file:
配置文件路径
kwargs:
其他需要传递给 MUSIC 接口的参数，例如：
- eleValueRanges: 要素值范围
- hourSeparate: 小时取整条件
- minSeparate: 分钟取整条件
Returns
-------
pd.DataFrame
高空观测资料表格数据，列名为 elements 中的值
"""
if elements is None:
elements = UPPER_AIR_DATASETS[data_code]["elements"]
interface_config = {
"name": "getUparEle",
"region": None,
"time": None,
"station": None,
"level": None,
}
if (interface_data_type is None and data_code in UPPER_AIR_DATASETS
and "interface_data_type" in UPPER_AIR_DATASETS[data_code]):
interface_data_type = UPPER_AIR_DATASETS[data_code][
"interface_data_type"]
if interface_data_type is not None:
interface_config["name"] = f"get{interface_data_type}Ele"
params = {
"dataCode": data_code,
"elements": elements,
}
if order is not None:
params["orderby"] = order
if data_code in UPPER_AIR_DATASETS and "order_by" in UPPER_AIR_DATASETS[
data_code]:
params["orderby"] = UPPER_AIR_DATASETS[data_code]["order_by"]
if count is not None:
params["limitCnt"] = count
if isinstance(time, pd.Interval):
interface_config["time"] = "TimeRange"
params["timeRange"] = _get_time_range_string(time)
elif isinstance(time, pd.Timestamp):
interface_config["time"] = "Time"
params["times"] = _get_time_string(time)
elif isinstance(time, typing.List):
interface_config["time"] = "Time"
params["times"] = ",".join([_get_time_string(t) for t in time])
elif isinstance(time, pd.Timedelta):
interface_config["name"] = "getSurfLatestTime"
params["latestTime"] = str(int(time / pd.to_timedelta("1h")))
del params["orderby"]
del params["elements"]
_get_level_params(level=level,
level_type=level_type,
interface_config=interface_config,
params=params)
if isinstance(station, str) or isinstance(station, int):
interface_config["station"] = "StaID"
params["staIds"] = str(station)
elif isinstance(station, typing.List):
interface_config["station"] = "StaID"
params["staIds"] = ",".join(station)
elif isinstance(station, typing.Tuple):
interface_config["station"] = "StaIdRange"
params["minStaId"] = station[0]
params["maxStaId"] = station[1]
if region is not None:
_get_region_params(region, params, interface_config)
if station_level is not None:
del params["orderby"]
if isinstance(station_level, str):
params["staLevels"] = station_level
if interface_config["station"] is None:
interface_config["station"] = "StaLevels"
elif isinstance(station_level, typing.List):
params["staLevels"] = ",".join(station_level)
if interface_config["station"] is None:
interface_config["station"] = "StaLevels"
params.update(**kwargs)
interface_id = _get_interface_id(interface_config)
logger.info(f"interface_id: {interface_id}")
params = _fix_params(interface_id, params)
client = _get_client(config_file)
result = client.callAPI_to_array2D(interface_id, params)
if result.request.error_code != 0:
logger.warning(
f"request error {result.request.error_code}: {result.request.error_message}"
)
df = result.to_pandas()
return df
def download_model_file(
data_code: str,
parameter: Union[str, List[str]] = None,
start_time: Union[pd.Interval, pd.Timestamp, List, pd.Timedelta] = None,
forecast_time: Union[str, pd.Timedelta] = None,
level_type: Union[str, int] = None,
level: Union[int, float] = None,
region: Dict = None,
data_type: str = None,
output_dir: Union[pathlib.Path, str] = None,
config_file: Optional[str] = None
):
interface_config = {
"name": "getNafpFile",
"element": None,
"region": None,
"time": None,
"level": None,
"valid_time": None
}
if output_dir is None:
output_dir = "./"
if data_type is None:
data_type = "forecast"
# data_type_mapper = {
# "forecast": "getNafpEleGrid",
# "analysis": "getNafpAnaEleGrid",
# }
# interface_config["name"] = data_type_mapper.get(data_type)
params = {
"dataCode": data_code,
}
if isinstance(parameter, str):
params["fcstEle"] = parameter
interface_config["element"] = "Element"
elif isinstance(parameter, List):
params["elements"] = ".".join(parameter)
if level_type is not None:
params["levelType"] = str(level_type)
interface_config["level"] = "Level"
if level is not None:
params["fcstLevel"] = str(level)
if isinstance(start_time, pd.Interval):
interface_config["time"] = "TimeRange"
params["timeRange"] = _get_time_range_string(start_time)
elif isinstance(start_time, pd.Timestamp):
interface_config["time"] = "Time"
params["time"] = _get_time_string(start_time)
elif isinstance(start_time, List):
interface_config["time"] = "Time"
params["times"] = ",".join([_get_time_string(t) for t in start_time])
elif isinstance(start_time, pd.Timedelta):
interface_config["name"] = "getNafpLatestTime"
params["latestTime"] = str(int(start_time / pd.to_timedelta("1h")))
# del params["orderby"]
del params["elements"]
if forecast_time is not None:
interface_config["valid_time"] = "Validtime"
if isinstance(forecast_time, str):
forecast_time = pd.to_timedelta(forecast_time)
valid_time = int(forecast_time / pd.Timedelta(hours=1))
params["validTime"] = str(valid_time)
if region is not None:
_get_region_params(region, params, interface_config)
interface_id = _get_interface_id(interface_config)
logger.info(f"interface_id: {interface_id}")
client = _get_client(config_file)
result = client.callAPI_to_downFile(interface_id, params, file_dir=output_dir)
if result.request.error_code != 0:
logger.warning(f"request error {result.request.error_code}: {result.request.error_message}")
files_info = result.files_info
for file_info in files_info:
logger.info(file_info.file_name)
return
def retrieve_model_grid(data_code: str,
parameter: str,
start_time: pd.Timestamp = None,
forecast_time: Union[str, pd.Timedelta] = None,
level_type: Union[str, int] = None,
level: Union[int, float] = None,
region: Dict = None,
data_type: str = None,
config_file: Optional[str] = None) -> xr.DataArray:
interface_config = {
"name": "getNafpEleGrid",
"region": None,
"time": None,
"level": None,
"valid_time": None
}
if data_type is None:
data_type = "forecast"
data_type_mapper = {
"forecast": "getNafpEleGrid",
"analysis": "getNafpAnaEleGrid",
}
interface_config["name"] = data_type_mapper.get(data_type)
params = {
"dataCode": data_code,
"fcstEle": parameter,
}
if level_type is not None:
params["levelType"] = str(level_type)
interface_config["level"] = "Level"
if level is not None:
params["fcstLevel"] = str(level)
if start_time is not None:
interface_config["time"] = "Time"
time = _get_time_string(start_time)
params["time"] = time
if forecast_time is not None:
interface_config["valid_time"] = "Validtime"
if isinstance(forecast_time, str):
forecast_time = pd.to_timedelta(forecast_time)
valid_time = int(forecast_time / pd.Timedelta(hours=1))
params["validTime"] = str(valid_time)
if region is not None:
_get_region_params(region, params, interface_config)
interface_id = _get_interface_id(interface_config)
logger.info(f"interface_id: {interface_id}")
client = _get_client(config_file)
result = client.callAPI_to_gridArray2D(interface_id, params)
if result.request.error_code != 0:
logger.warning(
f"request error {result.request.error_code}: {result.request.error_message}"
)
field = result.to_xarray()
return field