.times method

src/icartt/dataset.py

@@ -44,7 +44,10 @@ class DataStore1001:
         self.data = None
 
     def __getitem__(self, s=slice(None)):
-        return self.data[s]
+        # we can only slice if we have something, so
+        if self.data is not None:
+            return self.data[s]
+        # returns None implicitly if self.data is None
 
     def addBulk(self, raw):
         nlines, nvars = raw.shape
@@ -117,7 +120,10 @@ class DataStore2110(collections.UserDict):
         self.dvars = dvars
 
     def __getitem__(self, s=slice(None)):
-        return self.data[s]
+        # we can only slice if we have something, so
+        if self.data is not None:
+            return self.data[s]
+        # returns None implicitly if self.data is None
 
     def _addAuxline(self, auxline):
         newdata = {
@@ -421,13 +427,13 @@ class Dataset:
         :return: list of time steps
         :rtype: list
         """
-        # TODO: this method currently does not work at all.
-        #       suggest to also change to return a numpy array of numpy.datetime64
-        #       for consistency with other data output
-        return [
-            self.dateOfCollection + datetime.timedelta(seconds=x)
-            for x in self.independentVariable
-        ]
+        if self.data.data is None or self.independentVariable is None:
+            return np.datetime64('NaT')
+
+        ref_dt = np.datetime64(datetime.datetime(*self.dateOfCollection), 'ns')
+
+        # ivar unit is seconds as per standard; need to convert to ns to use timedelta64[ns] type.
+        return ref_dt + (self.data[self.independentVariable.shortname]*10**9).astype('timedelta64[ns]')
 
     @property
     def variables(self):
@@ -510,11 +516,8 @@ class Dataset:
         # line 7 - UTC date when data begin, UTC date of data reduction or revision
         # - comma delimited (yyyy, mm, dd, yyyy, mm, dd).
         dmp = f.readline()
-        self.dateOfCollection = datetime.datetime(*map(int, dmp[:3]))
-        self.dateOfRevision = datetime.datetime(*map(int, dmp[3:6]))
-        # TODO: we should either use aware datetime (UTC), date objects or
-        #       numpy.datetime64 here to avoid some 'rough edges' of Python's datetime library...
-
+        self.dateOfCollection = tuple(map(int, dmp[:3]))
+        self.dateOfRevision = tuple(map(int, dmp[3:6]))
 
         # line 8 - Data Interval (This value describes the time spacing (in seconds)
         # between consecutive data records. It is the (constant) interval between
@@ -696,7 +699,7 @@ class Dataset:
             + "_"
             + self.locationID
             + "_"
-            + datetime.datetime.strftime(self.dateOfCollection, dateFormat)
+            + datetime.datetime.strftime(datetime.datetime(*self.dateOfCollection), dateFormat)
         )
         fn += "_R" + str(self.revision) if not self.revision is None else ""
         fn += "_L" + str(self.launch) if not self.launch is None else ""
@@ -755,13 +758,7 @@ class Dataset:
         )
         # UTC date when data begin, UTC date of data reduction or revision - comma delimited (yyyy, mm, dd, yyyy, mm, dd).
         write_to_file(
-            delimiter.join(
-                [
-                    # TODO: if we use anything other than datetime.datetime, we'll have to ensure this still works...
-                    datetime.datetime.strftime(x, delimiter.join(["%Y", "%m", "%d"]))
-                    for x in [self.dateOfCollection, self.dateOfRevision]
-                ]
-            )
+            delimiter.join(f"{x:02d}" for x in (*self.dateOfCollection, *self.dateOfRevision))
         )
         # Data Interval (This value describes the time spacing (in seconds) between consecutive data records. It is the (constant) interval between values of the independent variable. For 1 Hz data the data interval value is 1 and for 10 Hz data the value is 0.1. All intervals longer than 1 second must be reported as Start and Stop times, and the Data Interval value is set to 0. The Mid-point time is required when it is not at the average of Start and Stop times. For additional information see Section 2.5 below.).
         write_to_file(delimiter.join([str(x) for x in self.dataIntervalCode]))
@@ -894,12 +891,9 @@ class Dataset:
         self.dataSourceDescription = "Musterdatenprodukt"
         self.missionName = "MUSTEREX"
 
-        # TODO: see also comment on _readData method.
-        #       we should either use aware datetime (UTC), date objects or
-        #       numpy.datetime64 here to avoid some 'rough edges' of Python's
-        #       datetime library...
-        self.dateOfCollection = datetime.datetime.today()
-        self.dateOfRevision = datetime.datetime.today()
+        # .utcnow() should not be used in general, but it is ok if you just need the timetuple.
+        self.dateOfCollection = datetime.datetime.utcnow().timetuple()[:3]
+        self.dateOfRevision = datetime.datetime.utcnow().timetuple()[:3]
         self.dataIntervalCode = [0.0]
 
         self.independentVariable = None

tests/test_dataset.py

@@ -255,14 +255,16 @@ class Simple1001TestCase(unittest.TestCase):
 
 class Create1001TestCase(unittest.TestCase):
     def testCreateDs(self):
+        now = datetime.datetime.today()
+
         ict = icartt.Dataset(format=icartt.Formats.FFI1001)
 
         ict.PIName = "Knote, Christoph"
         ict.PIAffiliation = "Faculty of Medicine, University Augsburg, Germany"
         ict.dataSourceDescription = "Example data"
         ict.missionName = "MBEES"
-        ict.dateOfCollection = datetime.datetime.today()
-        ict.dateOfRevision = datetime.datetime.today()
+        ict.dateOfCollection = now.timetuple()[:3]
+        ict.dateOfRevision = now.timetuple()[:3]
 
         ict.dataIntervalCode = [0]
 
@@ -295,6 +297,11 @@ class Create1001TestCase(unittest.TestCase):
 
         ict.endDefineMode()
 
+        # we have not added data yet, so data must be None
+        self.assertIsNone(ict.data[:])
+        # and times must be NaT
+        self.assertTrue(np.isnat(ict.times))
+
         ict.data.add(Time_Start=12.3, Time_Stop=12.5, Payload=23789423.2e5)
 
         mydict = {"Time_Start": 12.6, "Time_Stop": 13.1, "Payload": 324235644.1e5}
@@ -303,6 +310,11 @@ class Create1001TestCase(unittest.TestCase):
         data = np.array([(13.4, 14.0, 2348925e5), (14.1, 14.9, 23425634e5)])
         ict.data.addBulk(data)
 
+        # elements of the time array must be equal to our input
+        t0 = np.datetime64(datetime.datetime(*now.timetuple()[:3]), 'ns')
+        for have, want in zip(ict.times, (12.3, 12.6, 13.4, 14.1)):
+            self.assertEqual(int(have-t0), int(want*10**9))
+
         strOut = io.StringIO()
 
         ict.write(f=strOut)
@@ -310,6 +322,8 @@ class Create1001TestCase(unittest.TestCase):
         return True
 
 
+
+
 class BulkIOTestCase(unittest.TestCase):
     # TODO: need to test invalid input
     # def testInvalid(self):