def import_data(self, metadata, output):
source, entity_type = self._get_source(), self._get_entity_type()
parser = etree.XMLParser(load_dtd=True)
parser.resolvers.add(BBCTPEGResolver())
xml = etree.parse(urllib.urlopen(self._tpeg_url), parser=parser)
entities, seen = {}, set()
for entity in Entity.objects.filter(source=source):
if 'bbc-tpeg' in entity.identifiers:
entities[entity.identifiers['bbc-tpeg']] = entity
for message in xml.getroot().findall('tpeg_message'):
id = message.find('road_traffic_message').attrib['message_id']
road_traffic_message = message.find('road_traffic_message')
try:
entity = entities[id]
except KeyError:
entity = Entity()
entities[id] = entity
entity.source = source
entity.primary_type = entity_type
locs = map(self._wgs84_to_point, road_traffic_message.findall('location_container/location_coordinates/WGS84'))
if len(locs) > 1:
entity.geometry = LineString(*locs)
elif len(locs) == 1:
entity.geometry = locs[0]
else:
continue
entity.location = Point(
sum(p.x for p in locs) / len(locs),
sum(p.y for p in locs) / len(locs),
srid=4326,
)
entity.metadata['bbc_tpeg'] = {
'xml': etree.tostring(message),
'severity': road_traffic_message.attrib['severity_factor'],
'generated': road_traffic_message.attrib['message_generation_time'],
'version': int(road_traffic_message.attrib['version_number']),
}
entity.save(identifiers={'bbc-tpeg': id})
set_name_in_language(entity, 'en',
title=message.find('summary').text)
entity.all_types = [entity_type]
entity.update_all_types_completion()
seen.add(entity.pk)
for entity in Entity.objects.filter(source=source):
if not entity.pk in seen:
entity.delete()
def _load_from_csv(self, reader, entity_type, source):
j = 0
for i, line in enumerate(reader):
postcode_abbrev, (easting, northing) = line[0], line[10:12]
postcode_abbrev = postcode_abbrev.replace(' ', '')
# Now try to figure out where to put the space in
if re.match(r'[A-Z][0-9]{2}[A-Z]{2}', postcode_abbrev):
# A9 9AA
postcode = '%s %s' % (postcode_abbrev[:2], postcode_abbrev[2:])
elif re.match(r'[A-Z][0-9]{3}[A-Z]{2}', postcode_abbrev):
# A99 9AA
postcode = '%s %s' % (postcode_abbrev[:3], postcode_abbrev[3:])
elif re.match(r'[A-Z]{2}[0-9]{2}[A-Z]{2}', postcode_abbrev):
# AA9 9AA
postcode = '%s %s' % (postcode_abbrev[:3], postcode_abbrev[3:])
elif re.match(r'[A-Z]{2}[0-9]{3}[A-Z]{2}', postcode_abbrev):
# AA99 9AA
postcode = '%s %s' % (postcode_abbrev[:4], postcode_abbrev[4:])
elif re.match(r'[A-Z][0-9][A-Z][0-9][A-Z]{2}', postcode_abbrev):
# A9A 9AA
postcode = '%s %s' % (postcode_abbrev[:3], postcode_abbrev[3:])
elif re.match(r'[A-Z]{2}[0-9][A-Z][0-9][A-Z]{2}', postcode_abbrev):
# AA9A 9AA
postcode = '%s %s' % (postcode_abbrev[:4], postcode_abbrev[4:])
else:
postcode = postcode_abbrev
try:
easting, northing = int(easting), int(northing)
except ValueError:
continue
j += 1
try:
entity = Entity.objects.get(source=source, _identifiers__scheme='postcode', _identifiers__value=postcode_abbrev)
except Entity.DoesNotExist:
entity = Entity(source=source)
entity.location = Point(easting, northing, srid=3857)
entity.geometry = entity.location
entity.primary_type = entity_type
identifiers = {
'postcode': postcode_abbrev,
'postcode-canonical': postcode,
}
entity.save(identifiers=identifiers)
set_name_in_language(entity, 'en',
title=postcode)
entity.all_types.add(entity_type)
entity.update_all_types_completion()
def endElement(self, name):
if name in ('node', 'way') and self.valid:
try:
types = self.find_types(self.tags)
except ValueError:
self.ignore_count += 1
return
# Ignore ways that lay partly outside our bounding box
if name == 'way' and not all(id in self.node_locations for id in self.nodes):
return
# We already have these from OxPoints, so leave them alone.
if self.tags.get('amenity') == 'library' and self.tags.get('operator') == 'University of Oxford':
return
# Ignore disused and under-construction entities
if self.tags.get('life_cycle', 'in_use') != 'in_use' or self.tags.get('disused') in ('1', 'yes', 'true'):
return
reset_queries()
try:
entity = Entity.objects.get(source=self.source,
_identifiers__scheme='osm',
_identifiers__value=self.id)
created = True
except Entity.DoesNotExist:
entity = Entity(source=self.source)
created = False
if not 'osm' in entity.metadata or \
entity.metadata['osm'].get('attrs', {}).get('timestamp', '') < self.attrs['timestamp']:
if created:
self.create_count += 1
else:
self.modify_count += 1
if name == 'node':
entity.location = Point(self.node_location, srid=4326)
entity.geometry = entity.location
elif name == 'way':
cls = LinearRing if self.nodes[0] == self.nodes[-1] else LineString
entity.geometry = cls([self.node_locations[n] for n in self.nodes], srid=4326)
min_, max_ = (float('inf'), float('inf')), (float('-inf'), float('-inf'))
for lon, lat in [self.node_locations[n] for n in self.nodes]:
min_ = min(min_[0], lon), min(min_[1], lat)
max_ = max(max_[0], lon), max(max_[1], lat)
entity.location = Point((min_[0] + max_[0]) / 2 , (min_[1] + max_[1]) / 2 , srid=4326)
else:
raise AssertionError("There should be no other types of entity we're to deal with.")
names = dict()
for lang_code, lang_name in settings.LANGUAGES:
with override(lang_code):
if '-' in lang_code:
tags_to_try = ('name:%s' % lang_code, 'name:%s' % lang_code.split('-')[0], 'name', 'operator')
else:
tags_to_try = ('name:%s' % lang_code, 'name', 'operator')
name = None
for tag_to_try in tags_to_try:
if self.tags.get(tag_to_try):
name = self.tags.get(tag_to_try)
break
if name is None:
try:
name = reverse_geocode(*entity.location)[0]['name']
if not name:
raise IndexError
name = u"↝ %s" % name
except IndexError:
name = u"↝ %f, %f" % (self.node_location[1], self.node_location[0])
names[lang_code] = name
entity.metadata['osm'] = {
'attrs': dict(self.attrs),
'tags': dict(zip((k.replace(':', '_') for k in self.tags.keys()), self.tags.values()))
}
entity.primary_type = self.entity_types[types[0]]
entity.save(identifiers={'osm': self.id})
for lang_code, name in names.items():
set_name_in_language(entity, lang_code, title=name)
entity.all_types = [self.entity_types[et] for et in types]
entity.update_all_types_completion()
else:
self.unchanged_count += 1
def add_stop(self, meta, entity_type, source):
# Check this entity is in an area
if self.areas != None:
in_area = False
for area in self.areas:
if meta["atco-code"].startswith(area):
in_area = True
if not in_area:
return
# See if we're updating an existing object, or creating a new one
try:
entity = Entity.objects.get(
source=source, _identifiers__scheme="atco", _identifiers__value=meta["atco-code"]
)
except Entity.DoesNotExist:
entity = Entity(source=source)
except Entity.MultipleObjectsReturned:
# Handle clashes
Entity.objects.filter(
source=source, _identifiers__scheme="atco", _identifiers__value=meta["atco-code"]
).delete()
entity = Entity(source=source)
common_name, indicator, locality, street = [
meta.get(k) for k in ("common-name", "indicator", "locality-ref", "street")
]
if (
(common_name or "").endswith(" DEL")
or (indicator or "").lower() == "not in use"
or "to define route" in (common_name or "")
or "to def rte" in (common_name or "")
or "to def route" in (common_name or "")
or "def.rte" in (common_name or "")
):
# In the NaPTAN list, but indicates it's an unused stop
return
if self.meta["stop-type"] in ("MET", "GAT", "FER", "RLY"):
names = self.names
else:
names = dict()
for lang_code, lang_name in settings.LANGUAGES:
with override(lang_code):
# Try and find one in our preferred order
for lang in (lang_code, "en", None):
if lang in self.names:
common_name = self.names[lang]
break
# Expand abbreviations in indicators
if indicator is not None:
parts = []
for part in indicator.split():
parts.append(
{
# Translators: This is referring to bus stop location descriptions
"op": ugettext("Opposite"),
"opp": ugettext("Opposite"),
"opposite": ugettext("Opposite"),
# Translators: This is referring to bus stop location descriptions
"adj": ugettext("Adjacent"),
# Translators: This is referring to bus stop location descriptions
"outside": ugettext("Outside"),
"o/s": ugettext("Outside"),
# Translators: This is referring to bus stop location descriptions
"nr": ugettext("Near"),
# Translators: This is referring to bus stop location descriptions
"inside": ugettext("Inside"),
# Translators: This is referring to bus stop location descriptions
"stp": ugettext("Stop"),
}.get(part.lower(), part)
)
indicator = " ".join(parts)
if indicator is None and self.meta["stop-type"] in ("AIR", "FTD", "RSE", "TMU", "BCE"):
# Translators: This is referring to public transport entities
title = ugettext("Entrance to %s") % common_name
elif indicator is None and self.meta["stop-type"] in ("FBT",):
# Translators: This is referring to ferry ports
title = ugettext("Berth at %s") % common_name
elif indicator is None and self.meta["stop-type"] in ("RPL", "PLT"):
# Translators: This is referring to rail and metro stations
title = ugettext("Platform at %s") % common_name
elif indicator is not None and indicator.lower() != "none" and indicator not in common_name:
title = indicator + " " + common_name
else:
title = common_name
if street != None and street != "-":
# Deal with all-caps street names
if street.upper() == street:
fixedstreet = ""
wordstart = True
for letter in street:
if wordstart:
wordstart = False
fixedstreet += letter
continue
elif letter == " ":
wordstart = True
fixedstreet += letter
continue
else:
fixedstreet += letter.lower()
street = fixedstreet
if street not in title:
title += ", " + street
locality_lang = self.nptg_localities.get(locality)
if locality_lang != None:
for lang in (lang_code, "en", "cy"):
if lang in locality_lang:
if locality_lang[lang] != street:
title += ", " + locality_lang[lang]
break
names[lang_code] = title
entity.primary_type = entity_type
if not entity.metadata:
entity.metadata = {}
entity.metadata["naptan"] = meta
entity.location = Point(float(meta["longitude"]), float(meta["latitude"]), srid=4326)
entity.geometry = entity.location
if meta["atco-code"] in self.tube_references:
entity.metadata["london-underground-identifiers"] = self.tube_references[meta["atco-code"]]
identifiers = {"atco": meta["atco-code"]}
if "naptan-code" in meta:
meta["naptan-code"] = "".join(map(self.naptan_dial, meta["naptan-code"]))
identifiers["naptan"] = meta["naptan-code"]
if "plate-code" in meta:
identifiers["plate"] = meta["plate-code"]
if "crs" in meta:
identifiers["crs"] = meta["crs"]
if indicator != None and re.match("Stop [A-Z]\d\d?", indicator):
identifiers["stop"] = indicator[5:]
entity.save(identifiers=identifiers)
for lang_code, name in names.items():
# This is the NaPTAN, so default to English
if lang_code is None:
lang_code = "en"
set_name_in_language(entity, lang_code, title=name)
entity.all_types = (entity_type,)
entity.update_all_types_completion()
entity.groups.clear()
for stop_area in self.stop_areas:
sa, created = EntityGroup.objects.get_or_create(source=source, ref_code=stop_area)
entity.groups.add(sa)
entity.save()
return entity
