Refactor redundant overlay code

visualizer.py

@@ -11,6 +11,7 @@ import datetime
 from io import BytesIO
 import itertools
 import json
+from types import SimpleNamespace
 from wsgiref.handlers import CGIHandler
 from urllib.request import urlopen
 
@@ -158,69 +159,57 @@ def plot():
 
 @app.route('/plot/overlay')
 def plot_overlay():
-    name1 = request.args.get('name1')
-    year1 = int(request.args.get('year1'))
-    name2 = request.args.get('name2')
-    year2 = int(request.args.get('year2'))
+    num_datasets = 2
+    datasets = tuple(SimpleNamespace() for _ in range(num_datasets))
+    for n, dset in enumerate(datasets, start=1):
+        dset.name = request.args.get(f'name{n}')
+        dset.year = int(request.args.get(f'year{n}'))
     field = int(request.args.get('field'))
     units = field_units[field]
-    json_link1 = get_link(name1, year1) + '.jsonld'
-    json_link2 = get_link(name2, year2) + '.jsonld'
-    data1 = read_data(name1, year1, json_link1)
-    data2 = read_data(name2, year2, json_link2)
 
     def ignore_feb_29(rows):
         return [row for row in rows if (row[0].month, row[0].day) != (2, 29)]
 
-    data1 = np.array(ignore_feb_29(data1))
-    data2 = np.array(ignore_feb_29(data2))
+    for dset in datasets:
+        dset.json_link = get_link(dset.name, dset.year) + '.jsonld'
+        raw_data = read_data(dset.name, dset.year, dset.json_link)
+        dset.data = np.array(ignore_feb_29(raw_data))
 
     fig, axes = plt.subplots()
     fig.set_figheight(6)
     fig.set_figwidth(12)
-    axes.plot(data1[:, 0],
-              data1[:, field],
-              label=f'{name1} {data1[0][0].year}')
-    axes.plot(data1[:, 0],
-              data2[:, field],
-              alpha=0.6,
-              label=f'{name2} {data2[0][0].year}')
-    avg1 = np.nanmean(data1[:, field], dtype='float32')
-    avg2 = np.nanmean(data2[:, field], dtype='float32')
-    axes.hlines(y=avg1,
-                xmin=data1[:, 0][0],
-                xmax=data1[:, 0][-1],
-                alpha=0.7,
-                label=f'Avg {name1} {data1[0][0].year}')
-    axes.hlines(y=avg2,
-                xmin=data1[:, 0][0],
-                xmax=data1[:, 0][-1],
+    for i, dset in enumerate(datasets):
+        alpha_kw = {'alpha': 0.6} if i else {}
+        axes.plot(datasets[0].data[:, 0],
+                  dset.data[:, field],
+                  **alpha_kw,
+                  label=f'{dset.name} {dset.data[0][0].year}')
+    for dset in datasets:
+        dset.avg = np.nanmean(dset.data[:, field], dtype='float32')
+        axes.hlines(y=dset.avg,
+                    xmin=datasets[0].data[:, 0][0],
+                    xmax=datasets[0].data[:, 0][-1],
                     alpha=0.7,
-                label=f'Avg {name2} {data2[0][0].year}')
+                    label=f'Avg {dset.name} {dset.data[0][0].year}')
     axes.set_ylabel(units)
 
-    avg1 = np.nanmean(data1[:, field], dtype='float32')
-    avg2 = np.nanmean(data2[:, field], dtype='float32')
-    max1 = max(data1, key=lambda row: row[field])
-    max2 = max(data2, key=lambda row: row[field])
-    maximum, max_name = ((max1, name1) if max1[field] > max2[field]
-                         else (max2, name2))
-    min1 = min(data1, key=lambda row: row[field])
-    min2 = min(data2, key=lambda row: row[field])
-    minimum, min_name = ((min1, name1) if min1[field] <= min2[field]
-                         else (min2, name2))
+    for dset in datasets:
+        dset.max = max(dset.data, key=lambda row: row[field])
+        dset.min = min(dset.data, key=lambda row: row[field])
+    max_dset = max(datasets, key=lambda dset: dset.max[field])
+    min_dset = min(datasets, key=lambda dset: dset.min[field])
 
     axes.grid(True)
-    axes.set_title((f'Max {units}: {maximum[field]}, {max_name} Station, '
-                    f'Date: ({maximum[0]}). '
-                    f'Min {units}: {minimum[field]}, {min_name} Station, '
-                    f'Date: ({minimum[0]}).'),
+    axes.set_title((f'Max {units}: {max_dset.max[field]}, '
+                    f'{max_dset.name} Station, Date: ({max_dset.max[0]}). '
+                    f'Min {units}: {min_dset.min[field]}, '
+                    f'{min_dset.name} Station, Date: ({min_dset.min[0]}).'),
                    fontsize='small')
     axes.legend()
     axes.tick_params(labelbottom=False)
-    plt.suptitle(f'{units} measurements, '
-                 f'{name1} Station, {data1[0][0].year} / '
-                 f'{name2} Station, {data2[0][0].year}')
+    title_dsets = ' / '.join(f'{dset.name} Station, {dset.data[0][0].year}'
+                             for dset in datasets)
+    plt.suptitle(f'{units} measurements, {title_dsets}')
     buf = BytesIO()
     fig.savefig(buf, format='png')
     png_data = base64.b64encode(buf.getbuffer()).decode('ascii')