211 Python dependencies (#221)

* upgrade pandas and matplotlib

* added documentation

* corrected typo

Author: annette-lutz

pycode/memilio-epidata/memilio/epidata/getCaseData.py

@@ -182,7 +182,7 @@ def get_case_data(read_data=dd.defaultDict['read_data'],
                         "dataframe is empty for csv and geojson!")
             # drop columns that do not exist in data from github
             df = df.drop(["Altersgruppe2", "Datenstand", "OBJECTID",
-                          "Bundesland", "Landkreis"], 1)
+                          "Bundesland", "Landkreis"], axis=1)
     df = df.convert_dtypes()
 
     # output data to not always download it
@@ -237,7 +237,7 @@ def get_case_data(read_data=dd.defaultDict['read_data'],
 
     # get rid of unnecessary columns
     df = df.drop(['NeuerFall', 'NeuerTodesfall', 'NeuGenesen',
-                 "IstErkrankungsbeginn", "Meldedatum", "Refdatum"], 1)
+                  "IstErkrankungsbeginn", "Meldedatum", "Refdatum"], axis=1)
 
     print("Available columns:", df.columns)
 

pycode/memilio-epidata/memilio/epidata/getCaseDatawithEstimations.py

@@ -143,7 +143,7 @@ def get_case_data_with_estimations(read_data=dd.defaultDict['read_data'],
             df_cases.loc[(df_cases[dstr] == date_jh), deaths_estimated] = np.round(
                 fraction_deaths_conf * df_cases.loc[(df_cases[dstr] == date_jh), confirmed])
 
-        df_cases = df_cases.drop([dstr], 1)
+        df_cases = df_cases.drop([dstr], axis=1)
         gd.write_dataframe(df_cases, data_path,
                            file_to_change + "_estimated", file_format)
 

pycode/memilio-epidata/memilio/epidata/getPopulationData.py

@@ -286,7 +286,7 @@ def get_population_data(read_data=dd.defaultDict['read_data'],
         print('Information: Using new population data file ' + filename)
         df_pop_raw = gd.loadExcel(
             new_data_file, apiUrl='', extension='.xlsx',
-            param_dict={"engine": None, "sheet_name": filename, "header": 4})
+            param_dict={"engine": "openpyxl", "sheet_name": filename, "header": 4})
         column_names = list(df_pop_raw.columns)
         # rename columns
         rename_columns = {

pycode/memilio-epidata/memilio/epidata_test/test_epidata_getCaseData.py

@@ -43,9 +43,9 @@ class TestGetCaseData(fake_filesystem_unittest.TestCase):
 
     # load test data for read
     filename = os.path.join(here, 'test_epidata_getCaseData_data_read.json')
-    file_object = open(filename, 'r')
     # Load JSON file data to a python dict object.
-    dict_object = json.load(file_object)
+    with open(filename, 'r') as file_object:
+        dict_object = json.load(file_object)
     test_string_all_federal_states_and_counties_read = json.dumps(dict_object)[:-1] +\
         (""",{"Altersgruppe":"A60-A79","Geschlecht":"M","AnzahlFall":1,"AnzahlTodesfall":0,"Meldedatum":"2020-08-11",\
     "IdLandkreis":1002,"NeuerFall":0,"NeuerTodesfall":-9,"Refdatum":"2020-08-07","NeuGenesen":0,"AnzahlGenesen":1,\
@@ -109,9 +109,9 @@ class TestGetCaseData(fake_filesystem_unittest.TestCase):
     # (https://github.com/robert-koch-institut/SARS-CoV-2_Infektionen_in_Deutschland)
     filename = os.path.join(
         here, 'test_epidata_getCaseData_data_github.json')
-    file_object = open(filename, 'r')
     # Load JSON file data to a python dict object.
-    dict_object_github = json.load(file_object)
+    with open(filename, 'r') as file_object:
+        dict_object_github = json.load(file_object)
 
     test_string_all_federal_states_and_counties_github = json.dumps(
         dict_object_github)[:-1] + ("""]""")
@@ -120,9 +120,9 @@ class TestGetCaseData(fake_filesystem_unittest.TestCase):
     # (https://npgeo-corona-npgeo-de.hub.arcgis.com/datasets/66876b81065340a4a48710b062319336/about)
     filename = os.path.join(
         here, 'test_epidata_getCaseData_data_arcgis.json')
-    file_object = open(filename, 'r')
     # Load JSON file data to a python dict object.
-    dict_object_arcgis = json.load(file_object)
+    with open(filename, 'r') as file_object:
+        dict_object_arcgis = json.load(file_object)
 
     test_string_all_federal_states_and_counties_arcgis = json.dumps(
         dict_object_arcgis)[:-1] + ("""]""")

pycode/memilio-epidata/setup.py

@@ -68,10 +68,12 @@ def run(self):
     long_description='',
     test_suite='memilio.epidata_test',
     install_requires=[
-        'pandas<1.2.0',
-        'matplotlib<3.4',
+        # smaller pandas versions contain a bug that sometimes prevents reading
+        # some excel files (e.g. population or twitter data)
+        'pandas>=1.2.2',
+        'matplotlib',
         'tables',
-        'numpy>=1.21',
+        'numpy>=1.21',  # smaller numpy versions cause a security issue
         'openpyxl',
         'xlrd',
         'requests',
@@ -80,7 +82,8 @@ def run(self):
     ],
     extras_require={
         'dev': [
-            'pyfakefs==4.1.0',
+            # smaller pyfakefs versions use deprecated functions for matplotlib versions >=3.4
+            'pyfakefs>=4.2.1',
             'freezegun',
             'coverage',
             'pylint<=2.11.1',