def geom_area_diffs(**kwargs):
""" Calculates the spatial difference between the first and the last
version of the same area. Function requires at least two areas. """
list_current = current_data(**kwargs)
for current in list_current:
if current.geom_type == 'multipolygons':
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
if len(features) < 2:
result = str(0.0)
else:
area1 = session.query(func.ST_Area(func.ST_GeogFromText(
func.ST_AsText(GeomMultiPolygon.the_geom)))).filter(
GeomMultiPolygon.feature_id == features[0].id).first()[0]
area2 = session.query(func.ST_Area(func.ST_GeogFromText(
func.ST_AsText(GeomMultiPolygon.the_geom)))).filter(
GeomMultiPolygon.feature_id == features[1].id).first()[0]
result = str(area2 - area1)
save_result(kwargs['job_op_id'], current.osm_id, result)
def geom_sum_dist(**kwargs):
""" Calculates the cumulative distance accross points.
Function requires at least three points. """
list_current = current_data(**kwargs)
for current in list_current:
if current.geom_type == 'points':
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
result = 0.0
if len(features) >= 3:
for i, feature in enumerate(features):
if i != 0:
geog1 = session.query(func.ST_GeogFromText(
func.ST_AsText(GeomPoint.the_geom))).filter(
GeomPoint.feature_id == features[i - 1].id).first()[0]
geog2 = session.query(func.ST_GeogFromText(
func.ST_AsText(GeomPoint.the_geom))).filter(
GeomPoint.feature_id == feature.id).first()[0]
result += session.query(func.ST_Distance(
geog1, geog2, True)).first()[0]
save_result(kwargs['job_op_id'], current.osm_id, result)
def geom_from_original(**kwargs):
""" Calculates the distance from the first point.
Function requires at least two points. """
list_current = current_data(**kwargs)
for current in list_current:
if current.geom_type == 'points':
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
if len(features) < 2:
result = '0'
else:
geog1 = session.query(func.ST_GeogFromText(func.ST_AsText(
GeomPoint.the_geom))).filter(
GeomPoint.feature_id == features[0].id).first()[0]
geog2 = session.query(func.ST_GeogFromText(func.ST_AsText(
GeomPoint.the_geom))).filter(
GeomPoint.feature_id == features[-1].id).first()[0]
result = session.query(
func.ST_Distance(geog1, geog2, True)).first()[0]
save_result(kwargs['job_op_id'], current.osm_id, str(result))
def count_inserts(**kwargs):
""" Counts the number of tag insertions. """
# get current data
list_current = current_data(**kwargs)
for i, current in enumerate(list_current):
# select all version of the feature
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
# initialize counter
total = 0
if len(features) >= 2:
for i, feature in enumerate(features):
if i != 0:
previous = features[i - 1]
if feature.all_tags != previous.all_tags:
# go through all tags
for key in feature.all_tags:
if not previous.all_tags.has_key(key):
total += 1
# save results to the DB
save_result(kwargs['job_op_id'], current.osm_id, total)
def count_geom_changes(**kwargs):
""" Counts the number of geometric changes. """
# get current data
list_current = current_data(**kwargs)
for i, current in enumerate(list_current):
if current.osm_version >= kwargs['osm_version']:
# select features based on the osm_id
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
# initialize counter
total = 0
if len(features) >= 2:
for i, feature in enumerate(features):
if i != 0:
previous_geom = find_geom(features[i - 1])
current_geom = find_geom(feature)
if not previous_geom.almost_equals(current_geom):
total += 1
# save results to the DB
save_result(kwargs['job_op_id'], current.osm_id, total)
def count_updates(**kwargs):
""" Counts the number of modified tags or the number of modifications to a
certain tag of interest. """
# get current data
list_current = current_data(**kwargs)
# find TOI
try:
tag = kwargs['tag_of_interest']
except KeyError:
tag = None
for i, current in enumerate(list_current):
# select features based on the osm_id
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
# initialize counter
total = 0
if len(features) >= 2:
for i, feature in enumerate(features):
if i != 0:
previous = features[i - 1]
# if no tag was specified
if not tag:
for key in feature.all_tags:
# go through all tags
if previous.all_tags.has_key(key) and \
feature.all_tags.has_key(key) and \
previous.all_tags[key] != feature.all_tags[key]:
total += 1
# if a tag was specified
else:
# check the specified tag
if previous.all_tags.has_key(tag) and \
feature.all_tags.has_key(tag) and \
previous.all_tags[tag] != feature.all_tags[tag]:
total += 1
# save results to the DB
save_result(kwargs['job_op_id'], current.osm_id, total)
def currency_coefficient(**kwargs):
""" Assigns a currency coefficient to each feature. The larger the
coefficient, the greater the need for updating the object. """
# get current data
features = current_data(**kwargs)
# find the temporal boundary of the specified group of OSM features
ts_boundary = time_boundary(**kwargs)
for feature in features:
# the result is the timestamp difference in days
result = (ts_boundary - convert_timestamp(feature.osm_timestamp)).days
# save results to the DB
save_result(kwargs['job_op_id'], feature.osm_id, result)
def count_deletions(**kwargs):
""" Counts the number of deleted tags. """
# get current data
list_current = current_data(**kwargs)
# check if any specific tag was specified
try:
tag = kwargs['tag_of_interest']
except KeyError:
tag = None
for i, current in enumerate(list_current):
# select features based on the osm_id
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
# initialize counter
total = 0
if len(features) >= 2:
for i, feature in enumerate(features):
if i != 0:
previous = features[i - 1]
# if no tag was specified
if tag == None:
# go through all tags
for key in previous.all_tags:
if previous.all_tags.has_key(key) == True and \
feature.all_tags.has_key(key) == False:
total += 1
# if a tag was specified
else:
# check the specified tag
if previous.all_tags.has_key(tag) == True and \
feature.all_tags.has_key(tag) == False:
total += 1
# save results to the DB
save_result(kwargs['job_op_id'], current.osm_id, total)
def currency_revision_needed(**kwargs):
""" Marks features in need of revision. """
# get current data
features = current_data(**kwargs)
# find the temporal boundary of the specified group of OSM features
ts_boundary = time_boundary(**kwargs)
for feature in features:
# revision is not needed yet
if convert_timestamp(feature.osm_timestamp) >= ts_boundary:
result = 0
# revision is needed
else:
result = 1
# save results to the DB
save_result(kwargs['job_op_id'], feature.osm_id, result)
def geom_length_diffs(**kwargs):
""" Calculates the spatial difference between the first and the last
version of the same line. Function requires at least two lines. """
list_current = current_data(**kwargs)
for current in list_current:
if (current.geom_type == 'lines'
or current.geom_type == 'multilinestrings'):
features = session.query(Feature).filter(
Feature.osm_id == current.osm_id).order_by(
Feature.osm_version).all()
if len(features) < 2:
result = str(0.0)
else:
if current.geom_type == 'lines':
length1 = session.query(func.ST_Length(func.ST_GeogFromText(
func.ST_AsText(GeomLine.the_geom)))).filter(
GeomLine.feature_id == features[0].id).first()[0]
length2 = session.query(func.ST_Length(func.ST_GeogFromText(
func.ST_AsText(GeomLine.the_geom)))).filter(
GeomLine.feature_id == features[1].id).first()[0]
else:
length1 = session.query(func.ST_Length(func.ST_GeogFromText(
func.ST_AsText(GeomMultiLine.the_geom)))).filter(
GeomMultiLine.feature_id == features[0].id).first()[0]
length2 = session.query(func.ST_Length(func.ST_GeogFromText(
func.ST_AsText(GeomMultiLine.the_geom)))).filter(
GeomMultiLine.feature_id == features[1].id).first()[0]
result = str(length2 - length1)
save_result(kwargs['job_op_id'], current.osm_id, result)