def import_data(f, log):
""" Import the data in 'f', recording errors etc into 'log'.
'f' is a temporary file holding the data to import, and 'log' is a list
of strings.
We import the data from 'f', updating the Location records, etc, as we
go along. If we encounter any errors, or want to notify the user of an
action which has occurred, we append the error or notification to the
'log' list.
"""
locs_with_names = set() # Set of location codes which we had names for.
cur_loc = None # Location object we are currently importing.
for line in f:
line = line.rstrip()
if len(line) == 0: continue # Ignore blanks.
parts = line.split("\t")
if len(parts) < 2:
log.append("File contains invalid data. Are you sure it's " \
+ "a tab-delimited text file?")
break
# Process this line.
loc_code = parts[0]
field = parts[1].lower()
if len(parts) >= 3:
value = parts[2]
else:
value = ""
if cur_loc == None or loc_code != cur_loc.code:
# We're starting a new location.
if cur_loc != None:
# Save the previous location to disk.
cur_loc.save()
if is_new:
log.append("Added location " + cur_loc.code)
else:
log.append("Updated location " + cur_loc.code)
# Load the new location into memory.
try:
cur_loc = Location.objects.get(code=loc_code)
is_new = False
except Location.DoesNotExist:
cur_loc = Location() # Create a new location.
cur_loc.code = loc_code
is_new = True
# Import the field specified by this line in the import field.
if field == "level":
try:
level = Level.objects.get(level=int(value))
except:
level = None
if level != None:
cur_loc.level = level
elif field == "full_name":
cur_loc.name = value
elif field == "display_name":
cur_loc.display_name = value
elif field == "min_zoom_lat":
try:
cur_loc.min_zoom_lat = decimal.Decimal(value)
except:
log.append("Invalid min_zoom_lat value '" + value +
"' for location " + loc_code)
elif field == "min_zoom_long":
try:
cur_loc.min_zoom_long = decimal.Decimal(value)
except:
log.append("Invalid min_zoom_long value '" + value +
"' for location " + loc_code)
elif field == "max_zoom_lat":
try:
cur_loc.max_zoom_lat = decimal.Decimal(value)
except:
log.append("Invalid max_zoom_lat value '" + value +
"' for location " + loc_code)
elif field == "max_zoom_long":
try:
cur_loc.max_zoom_long = decimal.Decimal(value)
except:
log.append("Invalid max_zoom_long value '" + value +
"' for location " + loc_code)
elif field == "population":
try:
cur_loc.population = int(value)
except:
log.append("Invalid population value '" + value +
"' for location " + loc_code)
elif field == "area":
try:
cur_loc.area = int(value)
except:
log.append("Invalid area value '" + value +
"' for location " + loc_code)
elif field == "income":
try:
cur_loc.averageIncome = decimal.Decimal(value)
except:
log.append("Invalid income value '" + value +
"' for location " + loc_code)
elif field == "parents":
parents = []
ok = True # initially.
for parent_code in parts[2:]:
try:
parent = Location.objects.get(code=parent_code)
except Location.DoesNotExist:
log.append("Invalid parent location code: " + parent_code)
ok = False
break
parents.append(parent)
if ok:
cur_loc.parents.clear()
cur_loc.parents.add(*parents)
elif field == "children":
children = []
ok = True # initially.
for child_code in parts[2:]:
try:
child = Location.objects.get(code=child_code)
except Location.DoesNotExist:
log.append("Invalid child location code: " + child_code)
ok = False
break
children.append(child)
if ok:
cur_loc.children.clear()
cur_loc.children.add(*children)
elif field == "neighbors":
neighbours = []
ok = True # initially.
for neighbour_code in parts[2:]:
try:
neighbour = Location.objects.get(code=neighbour_code)
except Location.DoesNotExist:
log.append("Invalid neighbour location code: " +
neighbour_code)
ok = False
break
neighbours.append(neighbour)
if ok:
cur_loc.neighbors.clear()
cur_loc.neighbors.add(*neighbours)
elif field == "outline":
wkt = "".join(parts[2:])
try:
outline = Outline.objects.get(location=cur_loc)
except Outline.DoesNotExist:
if is_new:
# We have to save the current location before our outline
# can refer to it.
cur_loc.save()
outline = Outline()
outline.location = cur_loc
outline.outline = wkt
outline.save()
elif field == "name" or field == "addname":
filter = {}
ok = True # initially.
for filter_src in parts[3:]:
if "=" not in filter_src:
log.append("Invalid name filter: '" + filter_src + "'")
ok = False
break
key,value = filter_src.split("=", 1)
key = key.strip().lower()
value = value.strip()
if key not in ["source", "country", "state",
"metro", "region", "county", "city"]:
log.append("Invalid name filter: '" + filter_src + "'")
ok = False
break
try:
filter_loc = Location.objects.get(code=value)
except Location.DoesNotExist:
log.append("The filter '" + filter_src + " refers to " + \
"a non-existent location.")
ok = False
break
filter[key] = filter_loc
if ok:
# If this is the first time we've received a name for this
# location, and we aren't adding to the list of names, erase
# the existing names (if any).
if field == "name":
if loc_code not in locs_with_names and not is_new:
for loc_name in LocationName.objects.filter(
location=cur_loc):
name = loc_name.name
loc_name.delete()
if name.locationname_set.count() == 0:
# We've removed the last occurrence of this
# name -> delete the name as well.
name.delete()
# Remember that we've got a name for this location.
locs_with_names.add(loc_code)
# Create a new LocationName record for this name.
if is_new:
# We have to save the current location before our
# LocationName object can refer to it.
cur_loc.save()
try:
name = Name.objects.get(level=cur_loc.level,
name=parts[2])
except Name.DoesNotExist:
name = Name()
name.level = cur_loc.level
name.name = parts[2]
name.save()
loc_name = LocationName()
loc_name.name = name
loc_name.location = cur_loc
for field,filter_loc in filter.items():
if field == "source":
loc_name.sourceFilter = filter_loc
elif field == "country":
loc_name.countryFilter = filter_loc
elif field == "state":
loc_name.stateFilter = filter_loc
elif field == "metro":
loc_name.metroFilter = filter_loc
elif field == "region":
loc_name.regionFilter = filter_loc
elif field == "county":
loc_name.countyFilter = filter_loc
elif field == "city":
loc_name.cityFilter = filter_loc
loc_name.save()
elif field == "delete":
log.append("Deleting locations is not supported yet.")
else:
log.append("Invalid field for location " + loc_code +
": '" + field + "'")
f.close()
if cur_loc != None:
# Save the last location to disk.
cur_loc.save()
if is_new:
log.append("Added location " + cur_loc.code)
else:
log.append("Updated location " + cur_loc.code)
def main():
""" Our main program.
"""
# Start by deleting our existing location data.
print "Deleting existing name data..."
for name in Name.objects.all():
name.level = None
name.save()
Name.objects.all().delete()
for locName in LocationName.objects.all():
locName.name = None
locName.location = None
locName.countryFilter = None
locName.stateFilter = None
locName.metroFilter = None
locName.regionFilter = None
locName.cityFilter = None
locName.save()
LocationName.objects.all().delete()
# Load the Level objects into memory, for speed.
levels = {} # Maps level number to Level object.
for level in range(1, 9):
levels[level] = Level.objects.get(level=level)
# Load all the Location objects into memory, for speed of access.
print "Loading locations into memory..."
locations = {} # Maps location code to Location object.
for location in Location.objects.all():
locations[location.code] = location
print
print "Loaded %d locations" % len(locations)
print
# Process the "names.txt" file, one line at a time.
print "Processing 'names.txt' file..."
names = {} # Maps (level, name) tuple to Name object.
f = open("../Location Data/names.txt", "r")
for i,line in enumerate(f.readlines()[1:]):
if i % 1000 == 0:
print " %d names processed" % i
parts = line.rstrip().split("\t")
levelNum = int(parts[0][0])
nameString = parts[1]
countryCode = parts[2]
stateCode = parts[4]
metroCode = parts[6]
regionCode = parts[8]
cityCode = parts[10]
locationCode = parts[12]
level = levels[levelNum]
country = locations.get(countryCode)
state = locations.get(stateCode)
metro = locations.get(metroCode)
region = locations.get(regionCode)
city = locations.get(cityCode)
location = locations.get(locationCode)
if location == None:
print "Missing location: " + repr(locationCode)
continue
# Find the Name object for this row in the table. If necessary, we
# create a new Name record.
nameKey = (levelNum, nameString)
if nameKey in names:
name = names[nameKey]
else:
name = Name()
name.level = level
name.name = nameString
name.save()
names[nameKey] = name
# Finally, create a new LocationName object for this row in the table.
locName = LocationName()
locName.name = name
locName.location = location
locName.countryFilter = country
locName.stateFilter = state
locName.metroFilter = metro
locName.regionFilter = region
locName.cityFilter = city
locName.save()
# That's all, folks!
print "All done!"
def save_name(data, location, loc_name):
""" Save the given location name into the database.
The parameters are as follows:
'data'
The entered form data, as a dictionary.
'location'
The Location object we are entering a name for.
'loc_name'
The existing LocationName record we are editing, if any. If we
are adding a new name, this should be set to None.
We save the given name into the database, either updating 'loc_name' or
creating a new LocationName record as appropriate.
"""
# Get the Name object to use for this location name. If necessary, we
# create a new Name object for this name.
name_value = helpers.tidy_name(data['name'].upper())
try:
name = Name.objects.get(level=location.level, name=name_value)
except Name.DoesNotExist:
name = Name()
name.level = location.level
name.name = name_value
name.save()
# If necessary, create a new LocationName object for this name.
if loc_name == None:
loc_name = LocationName()
# Save the entered data into this location name.
loc_name.name = name
loc_name.location = location
loc_name.sourceFilter = getSource(data['source_id'])
if location.level.level >= 2:
if data['country_field'] != "":
loc_name.countryFilter = getLocation(data['country_id'])
else:
loc_name.countryFilter = None
else:
loc_name.countryFilter = None
if location.level.level >= 3:
if data['state_field'] != "":
loc_name.stateFilter = getLocation(data['state_id'])
else:
loc_name.stateFilter = None
else:
loc_name.stateFilter = None
if location.level.level >= 4:
if data['metro_field'] != "":
loc_name.metroFilter = getLocation(data['metro_id'])
else:
loc_name.metroFilter = None
else:
loc_name.metroFilter = None
if location.level.level >= 5:
if data['region_field'] != "":
loc_name.regionFilter = getLocation(data['region_id'])
else:
loc_name.regionFilter = None
else:
loc_name.regionFilter = None
if location.level.level >= 6:
if data['county_field'] != "":
loc_name.countyFilter = getLocation(data['county_id'])
else:
loc_name.countyFilter = None
else:
loc_name.countyFilter = None
if location.level.level >= 7:
if data['city_field'] != "":
loc_name.cityFilter = getLocation(data['city_id'])
else:
loc_name.cityFilter = None
else:
loc_name.cityFilter = None
loc_name.save()
# Finally, send signals telling the rest of the system that the Name and
# LocationName tables have been chagned.
name_changed.send(sender=None)
location_name_changed.send(sender=None)
