improved error handling when reading files

sample/dk2_plant.py

@@ -34,6 +34,8 @@ def plant_dk(ffargs):
 
     # Gather default data on plants
     plant_dir = r'W:\PROJECTS\MNP2020\MNP_2020\c_fases\f7_draagkracht\a_source_data\plant'
+    if not os.path.isdir(plant_dir):
+        raise AssertionError('Please connect to W:\\projects\MNP2020')
 
     # Vertaling associatie <-> beheertype. Add generalized syntaxon code and repair beheertypen ("01.02" -> "N01.02.00")
     plant_associaties_src = 'vertaling_IPO_ASSO_2009_ps.xls'  # Vertaling Associatie --> Beheertype
@@ -48,19 +50,24 @@ def plant_dk(ffargs):
 
     # Get master list of beheertypen
     bt_master_src = r'c:\apps\proj_code\benb_utils\resources\snl_beheertypen.csv'
-    bt_master = pd.read_csv(bt_master_src, sep=',', comment='#', quotechar='"')
+    try:
+        bt_master = pd.read_csv(bt_master_src, sep=',', comment='#', quotechar='"')
+    except FileNotFoundError:
+        raise AssertionError("please instal BenB repository and update local path to \\resources\snl_beheertypen.csv")
     btcode2omschrijving = dict(zip(bt_master.LAND_TYPE_CODE.apply(fix_bt, as_mnp=True), bt_master.LST_NAME))
 
     # Get default mnp draagkracht values
     mnp_dk_src = r'w:\PROJECTS\qmar\MNP-SNL-ParameterSet\Parameters_v06_2019_12_09\03_MNP_versie6_par_density_factors_BT2019_v2.csv'
-    mnp_dk = pd.read_csv(mnp_dk_src, sep=',', usecols=['Species_code', 'Land_type_code', 'Land_type_quality'])
+    try:
+        mnp_dk = pd.read_csv(mnp_dk_src, sep=',', usecols=['Species_code', 'Land_type_code', 'Land_type_quality'])
+    except FileNotFoundError:
+        raise AssertionError('Please connect to w:\\projects\qmar')
 
     # Read list of equested beheertypen to analyse, add mnp-style code
     try:
         bts = pd.read_csv(args.bt_src, sep=',', usecols=['Land_type_code'])
-    except ValueError as e:
-        print('Error: beheertypen list must contain column "Land_type_code"')
-        raise
+    except (ValueError, FileNotFoundError) as e:
+        raise AssertionError('cannot read beheertypen source: {}. Check for column Land_type_code!'.format(e))
     bts['bt_code'] = bts.Land_type_code.apply(fix_bt, as_mnp=True)  # Add MNP-style Beheertype codes
     bts.drop_duplicates(subset='bt_code', inplace=True)
 
@@ -72,10 +79,11 @@ def plant_dk(ffargs):
     # Read list of equested species, add syntax group and assert all plants
     try:
         species = pd.read_csv(args.sp_src, sep=',', usecols=['Species_code', 'Local_name', 'Scientific_name'])
-    except ValueError:
-        print('Error: species list must contain columns "Species_code" (S09xxxxxx) "Local_name" "Scientific_name"')
+    except (ValueError, FileNotFoundError) as e:
+        raise AssertionError('Cannot find species list, or species list is missing "Species_code" (S09xxxxxx) '
+                             '"Local_name" "Scientific_name" columns: {}'.format(e))
     species.drop_duplicates(subset='Species_code', inplace=True)
-    assert all(species.Species_code.apply(code2tax) == 'P'), "Error: non plants found in requested species list."
+    assert all(species.Species_code.apply(code2tax) == 'P'), "non plants found in requested species list."
 
     # species dicts
     code2scientific = dict(zip(species.Species_code, species.Scientific_name))
@@ -88,8 +96,8 @@ def plant_dk(ffargs):
         print('Doing {0}-{1}'.format(sp_code, bt))
 
         # Dictionary to hold results
-        d = dict.fromkeys(['Scientific_name', 'Land_type_code', 'mean_trouw', 'mean_freq', 'hits', 'hits_gte_th', 'dk_ww'],
-                          np.nan)
+        d = dict.fromkeys(['Scientific_name', 'Land_type_code', 'mean_trouw', 'mean_freq', 'hits', 'hits_gte_th',
+                           'dk_ww'], np.nan)
 
         # Get initial selection from plant response data
         d['hits'] = responses.loc[(responses.Soortnaam == code2scientific[sp_code]) &
@@ -107,10 +115,13 @@ def plant_dk(ffargs):
         d['dk_ww'] = dk_ww(frequentie=d['mean_freq'], trouw=d['mean_trouw'], th=args.th_dk, tuning=args.A)
 
         d['Species_code'] = sp_code
-        d['Scientific_name'] = code2scientific[sp_code]
-        d['Local_name'] = code2local[sp_code]
         d['Land_type_code'] = bt
-        d['Land_type_omschrijving'] = btcode2omschrijving[bt]
+        try:
+            d['Scientific_name'] = code2scientific[sp_code]
+            d['Local_name'] = code2local[sp_code]
+            d['Land_type_omschrijving'] = btcode2omschrijving[bt]
+        except KeyError as e:
+            raise AssertionError("{} is not recognized".format( e))
 
         # Wrap up results and append to list
         dk_lst.append(pd.DataFrame(d, index=pd.RangeIndex(1)))
@@ -197,10 +208,12 @@ def plant_dk(ffargs):
                       dkww_df.to_csv(index=False, sep='\t', line_terminator='\n'),
                       pd.pivot_table(dk.assign(Soort=dk.Scientific_name.str.cat(dk.Local_name, sep="_"),
                                                BT=dk.Land_type_code.str.cat(dk.Land_type_omschrijving, sep='_')),
-                                     index='Soort', columns='BT', values='dk_ww', aggfunc='mean')\
+                                     index='Soort', columns='BT', values='dk_ww', aggfunc='mean')
                         .to_csv(sep=',', line_terminator='\n'))
 
+    # Prepare output
     basename = '03PlantDensityFactors{0}'.format(ts_full)
+    assert os.path.isdir(args.out_dir), "{} is not a valid output directory".format(args.out_dir)
 
     # Write report
     with open(os.path.join(args.out_dir, '{}.txt'.format(basename)), 'w') as f:
@@ -208,11 +221,11 @@ def plant_dk(ffargs):
 
     # MNP style CSV output
     dk.dropna(axis=0, subset=['dk_ww'])\
-        .rename(mapper={'dk_ww': 'Land_type_quality'}, axis=1)\
-        .loc[:, ['Species_code', 'Land_type_code', 'Land_type_quality', 'Local_name', 'Scientific_name',
-                 'Land_type_omschrijving']]\
-        .round({'Land_type_quality': 2})\
-        .to_csv(os.path.join(args.out_dir, '{}.csv'.format(basename)), sep=',', header=True, index=False)
+      .rename(mapper={'dk_ww': 'Land_type_quality'}, axis=1)\
+      .loc[:, ['Species_code', 'Land_type_code', 'Land_type_quality', 'Local_name', 'Scientific_name',
+               'Land_type_omschrijving']]\
+      .round({'Land_type_quality': 2})\
+      .to_csv(os.path.join(args.out_dir, '{}.csv'.format(basename)), sep=',', header=True, index=False)
 
     print('...done')
 
@@ -232,7 +245,7 @@ if __name__ == '__main__':
         try:
             plant_dk(fargs)
         except AssertionError as e:
-            print(e)
+            print('Error: {}'.format(e))
             sys.exit(1)
 
     def full(fargs):
@@ -240,7 +253,7 @@ if __name__ == '__main__':
         try:
             plant_dk(fargs)
         except AssertionError as e:
-            print(e)
+            print('Error: {}'.format(e))
             sys.exit(1)
 
     model_desc = \