exe compiling changes, add rebuild option to dynamic data compilier, add columns example in read me

nick-gorman · nick-gorman · commit 0ffb74a82bdf · 2022-06-27T14:29:25.000+10:00
diff --git a/README.md b/README.md
@@ -163,6 +163,28 @@ price_data = dynamic_data_compiler(start_time, end_time, table, raw_data_cache,
 If the option `fformat='parquet'` is provided then no feather files will be created, and a parquet file will be used instead.
 While feather might have faster read/write, parquet has excellent compression characteristics and good compatability with packages for handling large on-memory/cluster datasets (e.g. Dask). This helps with local storage (especially for Causer Pays data) and file size for version control.
 
+#### Accessing additional table columns
+
+By default NEMOSIS only includes a subset of an AEMO table's columns, the full set of columns are listed in the 
+[MMS Data Model Reports](https://visualisations.aemo.com.au/aemo/di-help/Content/Data_Model/MMS_Data_Model.htm
+?TocPath=_____8), or can be seen by inspecting the CSVs in the raw data cache. Users of the python interface can add
+additional columns as shown below. If you using a feather or parquet based cache the rebuild option should be set to
+true so the additional columns are added to the cache files when they are rebuilt. This method of adding additional
+columns should also work with the `cache_compiler` function.
+
+```python
+from nemosis import defaults, dynamic_data_compiler
+
+defaults.table_columns['BIDPEROFFER_D'] += ['PASAAVAILABILITY']
+
+start_time = '2017/01/01 00:00:00'
+end_time = '2017/01/01 00:05:00'
+table = 'BIDPEROFFER_D'
+raw_data_cache = 'C:/Users/your_data_storage'
+
+volume_bid_data = dynamic_data_compiler(start_time, end_time, table, raw_data_cache, rebuild=True)
+```
+
 ##### Cache compiler
 
 This may be useful if you're using NEMOSIS to
diff --git a/nemosis/compiling to exe b/nemosis/compiling to exe
@@ -3,11 +3,12 @@
 3. Activate venv then comlpete the next steps
 4. Install NEMOSIS to venv: pip install e .
 5. Also install pyinstaller: pip install pyinstaller
-6. run pyinstaller: PyInstaller --onefile  nemosis\gui.py
+6. run pyinstaller: pyi-makespec --onefile --icon=favicon.ico nemosis/gui.py
 7. edit gui.spec by adding the following line after the 'a' variable is defined. The file path will need to be
 adjusted.
 a.datas += [('favicon.ico', 'C:\\Users\\user\\Documents\\GitHub\\nemosis\\nemosis\\favicon.ico',  'DATA')]
-8. rerun pyinstaller but use the spec file: Pyinstaller --onefile --icon=favicon.ico --clean  gui.spec
+8. Also add "icon='nemosis\\favicon.ico'" to EXE call in spec
+8. rerun pyinstaller but use the spec file: Pyinstaller --clean gui.spec
 
 Sample Spec:
 # -*- mode: python ; coding: utf-8 -*-
@@ -29,7 +30,7 @@ a = Analysis(['nemosis\\gui.py'],
              cipher=block_cipher,
              noarchive=False)
 
-a.datas += [('favicon.ico', 'C:\\Users\\nick\\Documents\\GitHub\\Abi_NEMOSIS\\nemosis\\favicon.ico',  'DATA')]
+a.datas += [('favicon.ico', 'C:\\Users\\nick\\Documents\\GitHub\\nemosis\\nemosis\\favicon.ico',  'DATA')]
 
 pyz = PYZ(a.pure, a.zipped_data,
              cipher=block_cipher)
diff --git a/nemosis/data_fetch_methods.py b/nemosis/data_fetch_methods.py
@@ -14,7 +14,7 @@ def dynamic_data_compiler(start_time, end_time, table_name, raw_data_location,
                           select_columns=None, filter_cols=None,
                           filter_values=None, fformat='feather',
                           keep_csv=True, parse_data_types=True,
-                          **kwargs):
+                          rebuild=False, **kwargs):
     """
     Downloads and compiles data for all dynamic tables. For non-CSV formats,
     will save data typed as strings/objects. To save typed data (e.g.
@@ -42,6 +42,8 @@ def dynamic_data_compiler(start_time, end_time, table_name, raw_data_location,
                            If False, will not return any data.
         parse_data_types (bool): infers data types of columns when reading
                                  data. default True for API use.
+        rebuild (bool): If True then cache files are rebuilt
+                        even if they exist already. False by default.
         **kwargs: additional arguments passed to the pd.to_{fformat}() function
 
     Returns:
@@ -77,6 +79,7 @@ def dynamic_data_compiler(start_time, end_time, table_name, raw_data_location,
                                            select_columns, date_filter,
                                            fformat=fformat,
                                            keep_csv=keep_csv,
+                                           rebuild=rebuild,
                                            write_kwargs=kwargs)
     if data_tables:
         all_data = _pd.concat(data_tables, sort=False)
diff --git a/nemosis/date_generators.py b/nemosis/date_generators.py
@@ -1,6 +1,6 @@
 from nemosis import defaults
 from calendar import monthrange
-
+from datetime import timedelta
 
 def year_and_month_gen(start_time, end_time):
 
@@ -60,21 +60,14 @@ def year_month_day_index_gen(start_time, end_time):
 
 def bid_table_gen(start_time, end_time):
 
-
     # Test for if we are after the date that aemo changes the datafiles to a daily format.
     if (start_time.year >= 2021 and start_time.month >= 4) or start_time.year >= 2022:
-        if (start_time.day == 2 and start_time.hour <= 4) or start_time.day == 1:
-            # If its a daily format only push the start buffer back by a day.
-            if start_time.month == 1:
-                start_time = start_time.replace(month=12)
-                start_time = start_time.replace(day=31)
-                start_time = start_time.replace(year=start_time.year - 1)
-            else:
-                start_time = start_time.replace(month=start_time.month - 1)
-                last_day_previous_month = monthrange(start_time.year, start_time.month)[1]
-                start_time = start_time.replace(day=last_day_previous_month)
+        if start_time.day == 2 and start_time.hour <= 4:
+            start_time = start_time - timedelta(days=2)
+        elif start_time.hour <= 4 or start_time.day == 1:
+            start_time = start_time - timedelta(days=1)
     else:
-        if start_time.day == 1 and start_time.hour == 0 or start_time.minute == 0:
+        if start_time.day == 1 and start_time.hour == 0 and start_time.minute == 0:
             # If its a monthly format push the buffer back by a month.
             if start_time.month == 1:
                 start_time = start_time.replace(month=12)
diff --git a/nemosis/test_data_fetch_methods.py b/nemosis/test_data_fetch_methods.py
@@ -384,8 +384,8 @@ def test_dispatch_tables_start_of_month(self):
             print('Passed')
 
     def test_dispatch_tables_start_of_month_previous_market_day_but_not_start_calendar_month(self):
-        start_time = '2021/06/01 03:00:00'
-        end_time = '2021/06/01 03:15:00'
+        start_time = '2021/06/05 03:00:00'
+        end_time = '2021/06/05 03:15:00'
         for table in self.table_names:
             print(f'Testing {table} returning values at start of month two.')
             dat_col = defaults.primary_date_columns[table]
@@ -398,11 +398,11 @@ def test_dispatch_tables_start_of_month_previous_market_day_but_not_start_calend
             expected_last_time = pd.to_datetime(end_time, format='%Y/%m/%d %H:%M:%S')
             if table == 'BIDDAYOFFER_D':
                 expected_length = 1 * 4
-                expected_last_time = '2021/05/31 00:00:00'
+                expected_last_time = '2021/06/04 00:00:00'
                 expected_last_time = \
                     pd.to_datetime(expected_last_time,
                                    format='%Y/%m/%d %H:%M:%S')
-                expected_first_time = '2021/05/31 00:00:00'
+                expected_first_time = '2021/06/04 00:00:00'
                 expected_first_time =\
                     pd.to_datetime(expected_first_time,
                                    format='%Y/%m/%d %H:%M:%S')
diff --git a/nemosis/test_date_generators.py b/nemosis/test_date_generators.py
@@ -89,6 +89,196 @@ def test_two_times_in_middle_of_jan_and_march_return_3_months(self):
         self.assertEqual(len(times), 3)
 
 
+class TestBidTableGen(unittest.TestCase):
+    def setUp(self):
+        pass
+
+    def test_two_times_not_at_edge_of_month_return_one_month(self):
+        start_time = datetime.strptime('2017/01/02 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2017/01/03 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2017')
+        self.assertEqual(times[0][1], '01')
+        self.assertEqual(times[0][2], None)
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(len(times), 1)
+
+
+    def test_two_times_first_at_edge_of_month_return_month_before_and_month_of_times(self):
+        start_time = datetime.strptime('2017/02/01 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2017/02/03 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2017')
+        self.assertEqual(times[0][1], '01')
+        self.assertEqual(times[0][2], None)
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2017')
+        self.assertEqual(times[1][1], '02')
+        self.assertEqual(times[1][2], None)
+        self.assertEqual(times[1][3], None)
+
+    def test_two_times_first_at_edge_of_year_return_month_before_and_month_of_times(self):
+        start_time = datetime.strptime('2017/01/01 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2017/01/03 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2016')
+        self.assertEqual(times[0][1], '12')
+        self.assertEqual(times[0][2], None)
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2017')
+        self.assertEqual(times[1][1], '01')
+        self.assertEqual(times[1][2], None)
+        self.assertEqual(times[1][3], None)
+
+    def test_two_times_second_at_edge_of_month_returns_one_month(self):
+        start_time = datetime.strptime('2017/01/05 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2017/01/31 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2017')
+        self.assertEqual(times[0][1], '01')
+        self.assertEqual(times[0][2], None)
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(len(times), 1)
+
+    def test_two_times_second_at_edge_of_year_returns_one_month(self):
+        start_time = datetime.strptime('2017/12/02 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2017/12/31 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2017')
+        self.assertEqual(times[0][1], '12')
+        self.assertEqual(times[0][2], None)
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(len(times), 1)
+
+    def test_two_times_in_middle_of_jan_and_march_return_3_months(self):
+        start_time = datetime.strptime('2017/01/05 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2017/03/24 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2017')
+        self.assertEqual(times[0][1], '01')
+        self.assertEqual(times[0][2], None)
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2017')
+        self.assertEqual(times[1][1], '02')
+        self.assertEqual(times[1][2], None)
+        self.assertEqual(times[1][3], None)
+        self.assertEqual(times[2][0], '2017')
+        self.assertEqual(times[2][1], '03')
+        self.assertEqual(times[2][2], None)
+        self.assertEqual(times[2][3], None)
+        self.assertEqual(len(times), 3)
+
+    def test_change_from_months_to_days(self):
+        start_time = datetime.strptime('2021/02/01 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2021/04/03 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        # Note we expect the 1st of april to be skipped
+        self.assertEqual(times[0][0], '2021')
+        self.assertEqual(times[0][1], '01')
+        self.assertEqual(times[0][2], None)
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2021')
+        self.assertEqual(times[1][1], '02')
+        self.assertEqual(times[1][2], None)
+        self.assertEqual(times[1][3], None)
+        # Data for march and the first of april is missing from the AEMO website so we don't generate the dates
+        # for these times.
+        self.assertEqual(times[2][0], '2021')
+        self.assertEqual(times[2][1], '04')
+        self.assertEqual(times[2][2], '02')
+        self.assertEqual(times[2][3], None)
+        self.assertEqual(times[3][0], '2021')
+        self.assertEqual(times[3][1], '04')
+        self.assertEqual(times[3][2], '03')
+        self.assertEqual(times[3][3], None)
+        self.assertEqual(len(times), 4)
+
+    def test_day_given_in_april_2021(self):
+        start_time = datetime.strptime('2021/04/01 00:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2021/04/03 00:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        # Note we expect the 1st of april to be skipped
+        self.assertEqual(times[0][0], '2021')
+        self.assertEqual(times[0][1], '04')
+        self.assertEqual(times[0][2], '02')
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2021')
+        self.assertEqual(times[1][1], '04')
+        self.assertEqual(times[1][2], '03')
+        self.assertEqual(times[1][3], None)
+        self.assertEqual(len(times), 2)
+
+    def test_include_previous_market_day(self):
+        start_time = datetime.strptime('2021/05/10 01:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2021/05/10 05:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        # Note we expect the 1st of april to be skipped
+        self.assertEqual(times[0][0], '2021')
+        self.assertEqual(times[0][1], '05')
+        self.assertEqual(times[0][2], '09')
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2021')
+        self.assertEqual(times[1][1], '05')
+        self.assertEqual(times[1][2], '10')
+        self.assertEqual(times[1][3], None)
+        self.assertEqual(len(times), 2)
+
+    def test_include_previous_month_if_1st_market_day_of_month(self):
+        start_time = datetime.strptime('2021/05/01 05:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2021/05/03 05:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2021')
+        self.assertEqual(times[0][1], '04')
+        self.assertEqual(times[0][2], '30')
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2021')
+        self.assertEqual(times[1][1], '05')
+        self.assertEqual(times[1][2], '01')
+        self.assertEqual(times[1][3], None)
+        self.assertEqual(times[2][0], '2021')
+        self.assertEqual(times[2][1], '05')
+        self.assertEqual(times[2][2], '02')
+        self.assertEqual(times[2][3], None)
+        self.assertEqual(times[3][0], '2021')
+        self.assertEqual(times[3][1], '05')
+        self.assertEqual(times[3][2], '03')
+        self.assertEqual(times[3][3], None)
+        self.assertEqual(len(times), 4)
+
+    def test_include_previous_month_if_1st_market_day_of_month_but_2nd_calendar_day(self):
+        start_time = datetime.strptime('2021/05/02 04:00:00', '%Y/%m/%d %H:%M:%S')
+        end_time = datetime.strptime('2021/05/03 05:00:00', '%Y/%m/%d %H:%M:%S')
+        gen = date_generators.bid_table_gen(start_time, end_time)
+        times = [(year, month, day, index) for year, month, day, index in gen]
+        self.assertEqual(times[0][0], '2021')
+        self.assertEqual(times[0][1], '04')
+        self.assertEqual(times[0][2], '30')
+        self.assertEqual(times[0][3], None)
+        self.assertEqual(times[1][0], '2021')
+        self.assertEqual(times[1][1], '05')
+        self.assertEqual(times[1][2], '01')
+        self.assertEqual(times[1][3], None)
+        self.assertEqual(times[2][0], '2021')
+        self.assertEqual(times[2][1], '05')
+        self.assertEqual(times[2][2], '02')
+        self.assertEqual(times[2][3], None)
+        self.assertEqual(times[3][0], '2021')
+        self.assertEqual(times[3][1], '05')
+        self.assertEqual(times[3][2], '03')
+        self.assertEqual(times[3][3], None)
+        self.assertEqual(len(times), 4)
+
+
 class TestYearMonthDayIndexGen(unittest.TestCase):
     def setUp(self):
         pass
diff --git a/setup.py b/setup.py
@@ -5,7 +5,7 @@
 
 setuptools.setup(
     name="nemosis",
-    version="3.0.0",
+    version="3.0.1",
     author="Nicholas Gorman, Abhijith Prakash",
     author_email="n.gorman305@gmail.com",
     description="A tool for accessing AEMO data.",
