videodata working

rowers/dataprep.py

@@ -10,6 +10,7 @@ from __future__ import unicode_literals, absolute_import
 from rowers.models import Workout, Team
 
 import pytz
+import collections
 
 from rowingdata import rowingdata as rrdata
 
@@ -122,24 +123,35 @@ from scipy.signal import savgol_filter
 
 import datetime
 
-def get_video_data(w):
-    df = getsmallrowdata_db(['time','velo','spm','catch'],ids=[w.id],
-    workstrokesonly=False,doclean=False,compute=False)
+def get_video_data(w,groups=['basic']):
+    columns = ['time','velo','spm']
+    columns += [name for name,d in rowingmetrics if d['group'] in groups]
+    columns = list(set(columns))
+    df = getsmallrowdata_db(columns,ids=[w.id],
+                            workstrokesonly=False,doclean=False,compute=False)
     df['time'] = (df['time']-df['time'].min())/1000.
     df.sort_values(by='time',inplace=True)
 
 
     df.set_index(pd.to_timedelta(df['time'],unit='s'),inplace=True)
     df2 = df.resample('1s').mean().interpolate()
+    if 'pace' in columns:
+        df2['pace'] = df2['pace']/1000.
+        p = df2['pace']
+        p = p.apply(lambda x:timedeltaconv(x))
+        p = nicepaceformat(p)
+        df2['pace'] = p
+
+
 
 
     #mask = df2['time'] < delay
     #df2 = df2.mask(mask).dropna()
     df2['time'] = (df2['time']-df2['time'].min())
 
+    df2 = df2.round(decimals=2)
+
     boatspeed = (100*df2['velo']).astype(int)/100.
-    spm = (10*df2['spm']).astype(int)/10.
-    catch = (10*df2['catch']).astype(int)/10.
 
     coordinates = get_latlon_time(w.id)
 
@@ -155,32 +167,36 @@ def get_video_data(w):
 
     # bundle data
     data = {
-        'boatspeed':[ v for v in boatspeed.values],
-        'latitude':[ l for l in latitude.values],
-        'longitude':[ l for l in longitude.values],
-        'spm':[ s for s in spm.values ],
-        'catch': [c for c in catch.values]
+        'boatspeed':boatspeed.values.tolist(),
+        'latitude':latitude.values.tolist(),
+        'longitude':longitude.values.tolist(),
         }
 
+    # metrics = {
+    #     'boatspeed': {
+    #         'unit': 'm/s',
+    #         'metric': 'boatspeed',
+    #         'name': 'Boat Speed'
+    #     },
+    # }
+
+    metrics = {}
+
+    for c in columns:
+        if c != 'time':
+            data[c] = df2[c].values.tolist()
+            metrics[c] = {
+                'name': dict(rowingmetrics)[c]['verbose_name'],
+                'metric': c,
+                'unit': ''
+            }
+
+    metrics['boatspeed'] = metrics.pop('velo')
+    # metrics['workperstroke'] = metrics.pop('driveenergy')
+    metrics = collections.OrderedDict(sorted(metrics.items()))
+
     maxtime = coordinates['time'].max()
 
-    metrics = {
-        'spm': {
-            'unit': '',
-            'metric': 'spm',
-            'name': 'SPM',
-        },
-        'boatspeed': {
-            'unit': 'm/s',
-            'metric': 'boatspeed',
-            'name': 'Boat Speed'
-        },
-        'catch': {
-            'unit': 'degrees',
-            'metric': 'catch',
-            'name': 'Catch Angle'
-        }
-    }
 
     return data, metrics, maxtime