WIP

README.md

@@ -33,6 +33,11 @@ The MSC-W database contains the EBAS database for 1990-2021 and the EEA_Airquip
 contain already hourly data if enough hours have been measured. Therefore, `resolution` is a
 required parameter.
 
+### nilupmfebas: EBAS format (Nasa-Ames)
+Reader for random EBAS data in NASA-AMES format. This reader is tested only with PMF data provided by
+NILU, but should in principle able to read any random text file in EBAS NASA-AMES.
+The variables provided contain in EBAS terms a combination of matrix and component with a number sign (#)
+as seperator (e.g. `pm10_pm25#total_carbon` or `pm10#organic_carbon` or `pm10#galactosan`)
 
 ## Usage
 ### aeronetsunreader
@@ -106,29 +111,38 @@ with pyaro.open_timeseries(
     data.values
 
 ```
-
-
-### geocoder_reverse_natural_earth
-geocoder_reverse_natural_earth is small helper to identify country codes for obs networks that don't mention the
-countrycode of a station in their location data
+### nilupmfebas
 ```python
-from geocoder_reverse_natural_earth import (
-    Geocoder_Reverse_NE,
-    Geocoder_Reverse_Exception,
-)
-geo = Geocoder_Reverse_NE()
-print(geo.lookup(60, 10)["ISO_A2_EH"])
-lat = 78.2361926
-lon = 15.3692614
-try:
-    geo.lookup(lat, lon)
-except Geocoder_Reverse_Exception as grex:
-    dummy = geo.lookup_nearest(lat, lon)
-    if dummy is None:
-        print(f"error: {lat},{lon}")
-    else:
-        print(dummy["ISO_A2_EH"])
-
-
+import pyaro
+TEST_URL = "/home/jang/data/Python3/pyaro-readers/tests/testdata/PMF_EBAS/NO0042G.20171109070000.20220406124026.high_vol_sampler..pm10.4mo.1w.NO01L_hvs_week_no42_pm10.NO01L_NILU_sunset_002.lev2.nas"
+TEST_URL = "/home/jang/data/Python3/pyaro-readers/tests/testdata/PMF_EBAS/NO0042G.20171109070000.20220406124026.high_vol_sampler..pm10.4mo.1w.NO01L_hvs_week_no42_pm10.NO01L_NILU_sunset_002.lev2.nas"
+
+def main():
+    with pyaro.open_timeseries(
+        'nilupmfebas', TEST_URL, filters=[]
+    ) as ts:
+        variables = ts.variables()
+        for var in variables:
+
+            data = ts.data(var)
+            print(f"var:{var} ; unit:{data.units}")
+            # stations
+            print(set(data.stations))
+            # start_times
+            print(data.start_times)
+            for idx, time in enumerate(data.start_times):
+                print(f"{time}_{data.values[idx]}")
+
+            # stop_times
+            data.end_times
+            # latitudes
+            data.latitudes
+            # longitudes
+            data.longitudes
+            # altitudes
+            data.altitudes
+            # values
+            data.values
 
 ```
+

src/pyaro_readers/nilupmfebas/EbasPmfReader.py

@@ -11,6 +11,7 @@
 from tqdm import tqdm
 
 from pathlib import Path
+import cf_units
 
 logger = logging.getLogger(__name__)
 
@@ -32,6 +33,7 @@ def __init__(
         self._set_filters(filters)
         self._header = []
         self._opts = {"default": ReadEbasOptions()}
+        self._variables = {}
 
         realpath = Path(filename).resolve()
 
@@ -41,12 +43,12 @@ def __init__(
             bar = tqdm(desc=tqdm_desc, total=len(files))
 
             for _ridx, file in enumerate(files):
-                # print(file)
                 bar.update(1)
                 self.read_file(file)
             bar.close()
         elif Path(realpath).is_file():
             self.read_file(realpath)
+            # print(realpath)
 
         else:
             # filename is something else
@@ -97,8 +99,21 @@ def read_file(self, filename):
                     pass
 
                 var_name = f"{matrix}#{self._file_dummy.var_defs[var_idx].name}"
-                if add_meta_flag:
-                    stat_name = self._file_dummy.meta["station_code"]
+                if var_name not in self._variables:
+                    self._variables[var_name] = (
+                        var_name,
+                        self._file_dummy.meta["unit"],
+                    )
+                else:
+                    var_name_ext = f"{self._variables[var_name]}0"
+                    self._variables[var_name] = (
+                        var_name_ext,
+                        self._file_dummy.meta["unit"],
+                    )
+
+                var_unit = self._file_dummy.var_defs[var_idx].unit
+                stat_name = self._file_dummy.meta["station_code"]
+                if stat_name not in self._stations:
                     country = self._file_dummy.meta["station_code"][0:2]
 
                     lat = float(self._file_dummy.meta["station_latitude"])
@@ -119,9 +134,8 @@ def read_file(self, filename):
                     add_meta_flag = False
 
                 # we might want to put a CF compliant unit here
-                self._data[var_name] = NpStructuredData(
-                    var_name, self._file_dummy.meta["unit"]
-                )
+                self._data[var_name] = NpStructuredData(var_name, var_unit)
+                # self._data[var_name] = []
                 # now add ts after ts
                 for t_idx, ts in enumerate(self._file_dummy.start_meas):
                     self._data[var_name].append(