Add get_hist_obs_daily function.

nmc_met_io/retrieve_cmadaas.py

@@ -547,7 +547,7 @@ def cmadaas_obs_in_rect_by_time(times, limit, data_code="SURF_CHN_MUL_HOR_N",
 
 
 def cmadaas_obs_in_rect_by_time_range(time_range, limit, data_code="SURF_CHN_MUL_HOR_N",
-                                      sta_levels=None, ranges=None, order=None, 
+                                      ranges=None, order=None, 
                                       count=None, trans_type=True,
                                       elements="Station_Id_C,Datetime,Lat,Lon,TEM"):
     """
@@ -557,8 +557,6 @@ def cmadaas_obs_in_rect_by_time_range(time_range, limit, data_code="SURF_CHN_MUL
         time_range (str): time for retrieve, "[YYYYMMDDHHMISS,YYYYMMDDHHMISS]"
         limit (list):  map limits, [min_lat, min_lon, max_lat, max_lon]
         data_code (str, optional): dataset code. Defaults to "SURF_CHN_MUL_HOR_N".
-        sta_levels (str, optional): station levels, seperated by ',',
-             like "011,012,013" for standard, base and general stations. Defaults to None.
         ranges (str, optional): elements value ranges, seperated by ';'
             range: (a,) is >a, [a,) is >=a, (,a) is <a, (,a] is <=a, (a,b) is >a & <b, 
                    [a,b) is >=a & <b, (a,b] is >a & <=b, [a,b] is >=a & <=b
@@ -593,7 +591,7 @@ def cmadaas_obs_in_rect_by_time_range(time_range, limit, data_code="SURF_CHN_MUL
               'maxLat': '{:.10f}'.format(limit[2]),
               'maxLon': '{:.10f}'.format(limit[3]),
               'orderby': order if order is not None else "Datetime:ASC"}
-    if sta_levels is not None: params['staLevels'] = sta_levels
+    #if sta_levels is not None: params['staLevels'] = sta_levels    # getSurfEleInRectByTimeRange not support this parameter.
     if ranges is not None: params['eleValueRanges'] = ranges
     if count is not None: params['limitCnt'] = str(count)
 

nmc_met_io/retrieve_cmadaas_history.py

@@ -21,7 +21,7 @@ def get_hist_obs_id(years=np.arange(2000, 2011, 1),
                     elements=None, sta_ids="54511"):
     """
     Retrieve hitory observations for sta_ids.
-    从大数据云平台上获取指定站点的地面观测数据. 由于大数据云平台对一次性检索有数量限制,
+    从CMADaaS上获取指定站点的地面观测数据. 由于大数据云平台对一次性检索有数量限制,
     因此先逐年下载, 然后再联接成一张观测记录表.
 
     注:
@@ -64,7 +64,7 @@ def get_hist_obs_id(years=np.arange(2000, 2011, 1),
 def get_accumulated_rainfall(time_range, data_code="SURF_CHN_MUL_HOR", 
                              accumulated=True, limit=None):
     """
-    从大数据云平台上“中国地面逐小时资料”下载站点的逐小时降水观测, 并累加为一段时间的累积降水.
+    从CMADaaS上“中国地面逐小时资料”下载站点的逐小时降水观测, 并累加为一段时间的累积降水.
     例如, 需要24h的累积降水, 用站点观测的PRE_24h有误差, 需要直接用PRE_1h来进行累积.
 
     Args:
@@ -75,7 +75,7 @@ def get_accumulated_rainfall(time_range, data_code="SURF_CHN_MUL_HOR",
     """
     
     # 读入数据
-    elements = "Station_Name,Province,Station_Id_C,Lon,Lat,PRE_1h"
+    elements = "Station_Name,Province,Station_Id_C,Lon,Lat,Datetime,PRE_1h"
     if limit is None:
         df = cmadaas_obs_by_time_range(time_range, data_code=data_code, elements=elements)
     else:
@@ -88,4 +88,52 @@ def get_accumulated_rainfall(time_range, data_code="SURF_CHN_MUL_HOR",
         df.rename(columns={"PRE_1h":"PRE"}, inplace=True)
         
     # 返回计算值
-    return df
+    return df
+
+
+def get_hist_obs_daily(years=np.arange(2000, 2011, 1), 
+                       data_code='SURF_CHN_MUL_DAY_N', 
+                       elements=None, limit=None,
+                       sta_levels="011,012,013"):
+    """
+    从CMADaaS上逐日的观测数据.
+
+    Args:
+        years (np.array, optional): years for historical data. Defaults to np.arange(2000, 2011, 1).
+        data_code (str, optional): dataset code. Defaults to 'SURF_CHN_MUL_DAY_N'.
+        elements ([type], optional): elements for retrieve, 'ele1, ele2, ...'. Defaults to None.
+        limit (tuple, optional): 指定返回数据的范围, [min_lat, min_lon, max_lat, max_lon]
+        sta_levels(str, optional): 指定返回的站点级别, 默认"011,012,013", 即国家基准气候站, 国家基本气象站, 国家一般气象站
+                                   若设为None值, 则返回全部级别的站点. 如果设置了limit, 该参数则不起作用.
+    Returns:
+        dataframe: station obervation records.
+    """
+    
+    # check elements
+    if elements is None:
+        elements = 'Station_Id_d,Station_levl,Station_Name,Province,Lat,Lon,Alti,Datetime,PRE_Time_0808,Q_PRE_Time_0808'
+        
+    # loop every yeas
+    data_list = []
+    tqdm_years = tqdm(years, desc="Years: ")
+    for year in tqdm_years:
+        start_time = str(year) + '0101000000'
+        end_time = str(year) + '1231230000'
+        time_range = "[" + start_time + "," + end_time + "]"
+        if limit is None:
+            df = cmadaas_obs_by_time_range(
+                time_range, data_code=data_code, elements=elements, 
+                sta_levels=sta_levels)
+        else:
+            df = cmadaas_obs_in_rect_by_time_range(
+                time_range, limit, data_code=data_code, elements=elements)
+        if df is not None:
+            df = df.drop_duplicates()
+            data_list.append(df)
+    
+    # concentrate dataframes
+    if len(data_list) == 0:
+        return None
+    else:
+        return pd.concat(data_list, axis=0, ignore_index=True)
+