#loads the classify_adm3_names dictionary, used to classify the admin 3 boundaries from adm3_name_dict import classify_adm3_names print classify_service_types dataReader = csv.reader(open(csv_filepathname), delimiter=',', quotechar='"') #The objects can't have the same name as the class, and a new object needs to be created each time a row is looped through for row in dataReader: #print 'hi' if row[0] != 'id': # Ignore the header row, import everything else EffortInstanceObj = EffortInstance() EffortInstanceObj.effort_instance_id = row[0] if row[1] == 'Long-term': EffortInstanceObj.date_start = '9999-12-31' #print 'hey' else: #need to parse and re-enter the date format as year-month-day for django to like it' split = re.split(r'\/', row[1].strip()) print split[2] + '-' + split[0] + '-' + split[1] EffortInstanceObj.date_start = split[2] + '-' + split[0] + '-' + split[1]
hacServiceCols = [
4, 5, 7, 8, 9, 10, 11, 13, 14, 17, 20, 21, 22, 23, 24, 25, 26, 27, 30, 34
]
def upfirstletter(value):
first = value[0] if len(value) > 0 else ''
remaining = value[1:] if len(value) > 1 else ''
return first.upper() + remaining
for row in dataReader:
if row[0] != 'id': # Ignore the header row, import everything else
EffortInstanceObj = EffortInstance()
EffortInstanceObj.effort_instance_id = row[0]
if row[33] == 'Hopital' or row[33] == 'CSL' or row[33] == 'CAL':
#EffortInstanceObj.date_start = '9999-12-31'
print 'long-term recorded'
if row[33] == 'Hopital':
EffortInstanceObj.provider_type = 'HL'
#print 'hospital recorded'
elif row[33] == 'CSL' or row[33] == 'CAL':
EffortInstanceObj.provider_type = 'CL'
#print 'clinic recorded'
def FillTables(locationOrder,adm2Bool,adm3Bool):
if row[0] != 'id': # Ignore the header row, import everything else
#checks to see if sector category is health
if re.search('Health',row[12]):
print 'row: id:'
print row[0]
EffortInstanceObj = EffortInstance()
EffortInstanceObj.effort_instance_id = row[0]
if locationOrder == 2:
EffortInstanceObj.effort_instance_id = row[0] + '02'
if locationOrder == 3:
EffortInstanceObj.effort_instance_id = row[0] + '03'
split = re.split(r'\/', row[9].strip())
print split[2] + '-' + split[0] + '-' + split[1]
EffortInstanceObj.date_start = split[2] + '-' + split[0] + '-' + split[1]
split = re.split(r'\/', row[10].strip())
print split[2] + '-' + split[0] + '-' + split[1]
EffortInstanceObj.date_end = split[2] + '-' + split[0] + '-' + split[1]
#cool way to make Service Providers unique, if obj does not exist then it creates it.
#https://docs.djangoproject.com/en/1.6/ref/models/querysets/#get-or-create
ServiceProviderObj, created = ServiceProvider.objects.get_or_create(provider_name = row[1])
EffortInstanceObj.service_provider = ServiceProvider.objects.get(provider_name=row[1])
EffortInstanceObj.updated_on = utc_datetime
EffortInstanceObj.updated_by = 'Haiti Aid Map scrape'
#HaitiAipMap does not have any lat lon coords, so don't add a location point
#Also, don't use random lat lon. It will break spatial queries
#loc = Location()
#loc.latitude = str(random.randint(0,10))
#loc.longitude = str(random.randint(0,10))
#loc.save()
'''
Insert new code here to match admin boundaries: The location_information column is in column 21 (row [20]). This contains a list
of admin boundaries where the facility is located at. The list contains entries delimited with '>'. The first entry on the list is Haiti.
The second entry is an admin 2 boundary. The third entry is ?.
There can be more than one place listed in the location column. If this is the case then we need to create a new effort instance
for each location in Haiti.
'''
if row[20]:
split_each_location = re.split(r'\|Haiti>', row[20].strip())
if len(split_each_location) > 1:
split_loc = re.split(r'>', split_each_location[locationOrder].strip())
if len(split_loc) > 1:
UpperAdmin1 = split_loc[0].upper()
UpperAdmin1 = re.split(r'\|', UpperAdmin1)
print 'length of split_loc:'
print len(split_loc)
print 'UpperAdmin1:'
print UpperAdmin1[0]
EffortInstanceObj.adm_1 = haiti_adm1_minustah.objects.get(adm1=classify_adm1_names[UpperAdmin1[0]])
if len(split_loc) > 1:
UpperAdmin2 = split_loc[1].upper()
UpperAdmin2 = re.split(r'\|', UpperAdmin2)
print 'UpperAdmin2:'
print UpperAdmin2[0]
try:
EffortInstanceObj.adm_2 = haiti_adm2_minustah.objects.get(adm2=classify_adm2_names[UpperAdmin2[0]])
except:
print "guess no match"
if len(split_loc) > 2:
UpperAdmin3 = split_loc[2].upper()
UpperAdmin3 = re.split(r'\|', UpperAdmin3)
print 'UpperAdmin3:'
print UpperAdmin3[0]
try:
EffortInstanceObj.adm_3 = haiti_adm3_minustah.objects.get(adm3=classify_adm3_names[UpperAdmin3[0]])
except:
print "guess no match"
#need a way first in seeing if a location exists close by
#EffortInstanceObj.location = loc.objects.get
EffortInstanceObj.save()
if len(split_each_location) > 1:
print 'location 1:'
print split_each_location[1]
if len(split_each_location) > 2 and adm2Bool == False:
print 'location 2:'
print split_each_location[2]
FillTables(2,True,False)
if len(split_each_location) > 3 and adm3Bool == False and adm2Bool == True:
print 'location 3:'
print split_each_location[3]
FillTables(3,True,True)
hacHeaders = ["id","title","address ","arv","cardiovascular","dental","department ","diabetes","emergency","free_obstetric_care_program","general_consultation","hypertension","institution_code","intensive_care","laboratory","latitude ","longitude","malaria","number_of_adult_beds","number_of_pediatric_beds","obgyn","operating_room","orl ","palliative_care","pediatrics","pharmacy","physical_therapy","psychology_service","section","status","surgery","telephone","town ","type","vaccination"]
# TODO: Need to add column 3 to this below list, but don't know what "arv" means and its not yet in dictionary
hacServiceCols = [4,5,7,8,9,10,11,13,14,17,20,21,22,23,24,25,26,27,30,34];
def upfirstletter(value):
first = value[0] if len(value) > 0 else ''
remaining = value[1:] if len(value) > 1 else ''
return first.upper() + remaining
for row in dataReader:
if row[0] != 'id': # Ignore the header row, import everything else
EffortInstanceObj = EffortInstance()
EffortInstanceObj.effort_instance_id = row[0]
if row[33] == 'Hopital' or row[33] == 'CSL' or row[33] == 'CAL':
#EffortInstanceObj.date_start = '9999-12-31'
print 'long-term recorded'
if row[33] == 'Hopital':
EffortInstanceObj.provider_type = 'HL'
#print 'hospital recorded'
elif row[33] == 'CSL' or row[33] == 'CAL':
EffortInstanceObj.provider_type = 'CL'
#print 'clinic recorded'
def FillTables(locationOrder, adm2Bool, adm3Bool):
if row[0] != 'id': # Ignore the header row, import everything else
#checks to see if sector category is health
if re.search('Health', row[12]):
print 'row: id:'
print row[0]
EffortInstanceObj = EffortInstance()
EffortInstanceObj.effort_instance_id = row[0]
if locationOrder == 2:
EffortInstanceObj.effort_instance_id = row[0] + '02'
if locationOrder == 3:
EffortInstanceObj.effort_instance_id = row[0] + '03'
split = re.split(r'\/', row[9].strip())
print split[2] + '-' + split[0] + '-' + split[1]
EffortInstanceObj.date_start = split[2] + '-' + split[
0] + '-' + split[1]
split = re.split(r'\/', row[10].strip())
print split[2] + '-' + split[0] + '-' + split[1]
EffortInstanceObj.date_end = split[2] + '-' + split[
0] + '-' + split[1]
#cool way to make Service Providers unique, if obj does not exist then it creates it.
#https://docs.djangoproject.com/en/1.6/ref/models/querysets/#get-or-create
ServiceProviderObj, created = ServiceProvider.objects.get_or_create(
provider_name=row[1])
EffortInstanceObj.service_provider = ServiceProvider.objects.get(
provider_name=row[1])
EffortInstanceObj.updated_on = utc_datetime
EffortInstanceObj.updated_by = 'Haiti Aid Map scrape'
#HaitiAipMap does not have any lat lon coords, so don't add a location point
#Also, don't use random lat lon. It will break spatial queries
#loc = Location()
#loc.latitude = str(random.randint(0,10))
#loc.longitude = str(random.randint(0,10))
#loc.save()
'''
Insert new code here to match admin boundaries: The location_information column is in column 21 (row [20]). This contains a list
of admin boundaries where the facility is located at. The list contains entries delimited with '>'. The first entry on the list is Haiti.
The second entry is an admin 2 boundary. The third entry is ?.
There can be more than one place listed in the location column. If this is the case then we need to create a new effort instance
for each location in Haiti.
'''
if row[20]:
split_each_location = re.split(r'\|Haiti>', row[20].strip())
if len(split_each_location) > 1:
split_loc = re.split(
r'>', split_each_location[locationOrder].strip())
if len(split_loc) > 1:
UpperAdmin1 = split_loc[0].upper()
UpperAdmin1 = re.split(r'\|', UpperAdmin1)
print 'length of split_loc:'
print len(split_loc)
print 'UpperAdmin1:'
print UpperAdmin1[0]
EffortInstanceObj.adm_1 = haiti_adm1_minustah.objects.get(
adm1=classify_adm1_names[UpperAdmin1[0]])
if len(split_loc) > 1:
UpperAdmin2 = split_loc[1].upper()
UpperAdmin2 = re.split(r'\|', UpperAdmin2)
print 'UpperAdmin2:'
print UpperAdmin2[0]
try:
EffortInstanceObj.adm_2 = haiti_adm2_minustah.objects.get(
adm2=classify_adm2_names[UpperAdmin2[0]])
except:
print "guess no match"
if len(split_loc) > 2:
UpperAdmin3 = split_loc[2].upper()
UpperAdmin3 = re.split(r'\|', UpperAdmin3)
print 'UpperAdmin3:'
print UpperAdmin3[0]
try:
EffortInstanceObj.adm_3 = haiti_adm3_minustah.objects.get(
adm3=classify_adm3_names[
UpperAdmin3[0]])
except:
print "guess no match"
#need a way first in seeing if a location exists close by
#EffortInstanceObj.location = loc.objects.get
EffortInstanceObj.save()
if len(split_each_location) > 1:
print 'location 1:'
print split_each_location[1]
if len(split_each_location) > 2 and adm2Bool == False:
print 'location 2:'
print split_each_location[2]
FillTables(2, True, False)
if len(split_each_location
) > 3 and adm3Bool == False and adm2Bool == True:
print 'location 3:'
print split_each_location[3]
FillTables(3, True, True)