def results_procLocation():
user_location = request.form.get("location")
if user_location == '':
user_location = 'San Francisco, CA, United States'
lat,lon,full_add,data = maps.geocode(user_location)
# set the bounding box for the requested area
res = data['results'][0]
lng_sw = res['geometry']['viewport']['southwest']['lng']
lng_ne = res['geometry']['viewport']['northeast']['lng']
lat_sw = res['geometry']['viewport']['southwest']['lat']
lat_ne = res['geometry']['viewport']['northeast']['lat']
if lat_sw >= 36.94 and lat_ne <= 38.0 and lng_ne >= -122.53 and lng_sw <= -121.8:
# get the times
endTime = pd.datetime.replace(pd.datetime.now(), microsecond=0)
startTime = pd.datetime.isoformat(endTime - pd.tseries.offsets.Hour(timeWindow_hours))
endTime = pd.datetime.isoformat(endTime)
return redirect(url_for('.results', lng_sw=lng_sw, lng_ne=lng_ne, lat_sw=lat_sw, lat_ne=lat_ne, startTime=startTime, endTime=endTime, city=full_add))
else:
this_lon, this_lat = sd.set_get_boundBox(area_str='bayarea')
# for our loc, just set the average
user_lon = np.mean(this_lon)
user_lat = np.mean(this_lat)
latlng_sw = [this_lat[1], this_lon[1]]
latlng_ne = [this_lat[0], this_lon[0]]
selected = "1"
return render_template('out_of_bounds.html', examples=examples,\
user_lat=user_lat, user_lon=user_lon,\
latlng_sw=latlng_sw, latlng_ne=latlng_ne,\
selected=selected)
def food_groups_page():
user_location = request.args.get("origin")
lat,lon,full_add,data = maps.geocode(user_location)
clusters = foodgroups.foodGroups(lat,lon)
restaurants = []
for i in range(len(clusters['X'])):
restaurants.append(dict(lat=clusters['X'][i][0], long=clusters['X'][i][1], clusterid=clusters['labels'][i]))
return render_template('results.html',results=restaurants,user_lat = lat, user_long = lon, faddress = full_add, ncluster = clusters['n_clusters'])
def food_groups_page():
user_location = request.args.get("origin")
lat, lon, full_add, data = maps.geocode(user_location)
clusters = foodgroups.foodGroups(lat, lon)
restaurants = []
for i in range(len(clusters['X'])):
restaurants.append(
dict(lat=clusters['X'][i][0],
long=clusters['X'][i][1],
clusterid=clusters['labels'][i]))
return render_template('results.html',
results=restaurants,
user_lat=lat,
user_long=lon,
faddress=full_add,
ncluster=clusters['n_clusters'])
def food_groups_page():
try:
user_location = request.args.get("origin")
if len(user_location) == 0:
return render_template('oops.html')
lat,lon,full_add,data = maps.geocode(user_location)
sortkey = int(request.args.get("keychain"))
clusters,restdata, cluster_info = foodgroups.foodGroups(lat,lon,key = sortkey,cache=True)
restaurants = []
for ix,a in restdata.iterrows():
thisdat = a
restaurants.append(dict(lat=thisdat['latitude']
,long=thisdat['longitude']
,clusterid=thisdat['ranking']
,url=thisdat['url']
,name=thisdat['name']
,pic=thisdat['photo']
))
return render_template('results3.html',results=restaurants,c_info = cluster_info, user_lat = lat, user_long = lon, faddress = full_add, ncluster = clusters['n_clusters'])
except:
# Well something went wrong in here
return render_template('oops.html')
def cleanData(data):
'''
Clean YELP result dictionaries into handy dataframes
Want business location (lat, long), business name,
'''
businesslist = data['businesses']
df = pd.DataFrame(columns = ('name','full_address','rating',
'review_count','distance','categories','snippet','latitude','longitude','latlongfound'))
failures = []
for i in range(len(businesslist)):
try:
full_address = ''
for a in businesslist[i]['location']['display_address']:
full_address = full_address +' %s'%a
except:
continue
# full_address = businesslist[i]['location']['display_address'][0] + ' '+businesslist[i]['location']['display_address'][1]
try:
loc = businesslist[i]['location']['coordinate']
lat = loc['latitude']
long = loc['longitude']
latlongfound = 1
except:
# Probably doesn't have a coordinate
failures.append(full_address)
latlongfound = 0
lat = 0.0
long = 0.0
categories = ''
try:
for a in businesslist[i]['categories']:
categories+= a[1]+',' #lowercase
except:
pass
try:
df.loc[len(df)+1] = [businesslist[i]['name']
, full_address
, businesslist[i]['rating']
, businesslist[i]['review_count']
, businesslist[i]['distance']
, categories
, businesslist[i]['snippet_text']
, lat
, long
, latlongfound
]
except:
pass
try:
url = 'http://www.datasciencetoolkit.org/street2coordinates'
# Convert failures to json and read from web API (datasciencetoolkit.org)
req = urllib2.Request(url,json.dumps(failures))
response = urllib2.urlopen(req)
healing = json.loads(response.read())
except:
healing = {}
for a in failures:
try:
lat,lon,full_add,data = maps.geocode(a)
healing[a] = {}
healing[a]['latitude'] = lat
healing[a]['longitude'] = lon
except:
break
for a in healing.keys():
try:
df.loc[df[df['full_address']==a].index,'latitude'] = healing[a]['latitude']
df.loc[df[df['full_address']==a].index,'longitude'] = healing[a]['longitude']
df.loc[df[df['full_address']==a].index,'latlongfound'] = 1
except:
pass
return df[df['latlongfound']==1]
def cleanData4Square(data):
'''
Clean 4Square result dictionaries into handy dataframes
Want business location (lat, long), business name,
'''
businesslist = data['groups'][0]['items'];
df = pd.DataFrame(columns = ('name','full_address','street','rating',
'review_count','distance','categories','price','latitude','longitude','latlongfound',
'photo','hasphoto','IsOpenNow','url'))
failures = []
for i in range(len(businesslist)):
try:
full_address = ''
full_address = full_address + businesslist[i]['venue']['location']['address']+', '
full_address = full_address + businesslist[i]['venue']['location']['formattedAddress'][1]
except:
continue
try:
lat = businesslist[i]['venue']['location']['lat']
long = businesslist[i]['venue']['location']['lng']
latlongfound = 1
except:
# Probably doesn't have a coordinate
failures.append(full_address)
latlongfound = 0
lat = 0.0
long = 0.0
categories = ''
try:
for a in businesslist[i]['venue']['categories']: categories+= a['shortName']+','
except:
pass
photourl = ''
try:
prefix = businesslist[i]['venue']['specials']['items'][0]['photo']['prefix']
suffix = businesslist[i]['venue']['specials']['items'][0]['photo']['suffix']
photourl = prefix + '100x100' + suffix
hasphoto = 2
except:
try:
prefix = businesslist[i]['venue']['featuredPhotos']['items'][0]['prefix']
suffix = businesslist[i]['venue']['featuredPhotos']['items'][0]['suffix']
photourl = prefix + '100x100' + suffix
hasphoto = 1
except:
hasphoto = 0
pass
try:
df.loc[len(df)+1] = [businesslist[i]['venue']['name']
, full_address
, businesslist[i]['venue']['location']['address']
, businesslist[i]['venue']['rating']
, businesslist[i]['venue']['ratingSignals']
, businesslist[i]['venue']['location']['distance']
, categories
, businesslist[i]['venue']['price']['tier']
, lat
, long
, latlongfound
, photourl
, hasphoto
, businesslist[i]['venue']['hours']['isOpen']
, 'https://foursquare.com/v/'+businesslist[i]['venue']['id']
]
except:
pass
try:
url = 'http://www.datasciencetoolkit.org/street2coordinates'
# Convert failures to json and read from web API (datasciencetoolkit.org)
req = urllib2.Request(url,json.dumps(failures))
response = urllib2.urlopen(req)
healing = json.loads(response.read())
except:
healing = {}
for a in failures:
try:
lat,lon,full_add,data = maps.geocode(a)
healing[a] = {}
healing[a]['latitude'] = lat
healing[a]['longitude'] = lon
except:
break
for a in healing.keys():
try:
df.loc[df[df['full_address']==a].index,'latitude'] = healing[a]['latitude']
df.loc[df[df['full_address']==a].index,'longitude'] = healing[a]['longitude']
df.loc[df[df['full_address']==a].index,'latlongfound'] = 1
except:
pass
return df[df['latlongfound']==1]
