def create_animation(self, dataset_name, year_list, indicator_name, viz_metadata):
map_file_list = []
for year in year_list:
map_file_list.append(
os.path.join(
self.storage_location,
"anim_"
+ dataset_name
+ "_map_"
+ str(year)
+ "_"
+ indicator_name
+ "."
+ MapnikMap.get_file_extension(self),
)
)
table_name = dataset_name + "_animated_map_" + str(min(year_list)) + "_" + indicator_name
animation_file_name = str(os.path.join(self.storage_location, table_name + "." + self.get_file_extension()))
os.system("convert -delay 100 %s -loop 0 %s" % (" ".join(map_file_list), animation_file_name))
# delete intermediate png files
for i in range(map_file_list.__len__()):
os.remove(map_file_list[i])
metadata = ([indicator_name], table_name, [min(year_list)])
viz_metadata.append(metadata)
def create_animation(self, dataset_name, year_list, indicator_name,
viz_metadata):
map_file_list = []
for year in year_list:
map_file_list.append(
os.path.join(
self.storage_location,
'anim_' + dataset_name + '_map_' + str(year) + '_' +
indicator_name + '.' + MapnikMap.get_file_extension(self)))
table_name = dataset_name + '_animated_map_' + str(
min(year_list)) + '_' + indicator_name
animation_file_name = str(
os.path.join(self.storage_location,
table_name + '.' + self.get_file_extension()))
os.system('convert -delay 100 %s -loop 0 %s' %
(' '.join(map_file_list), animation_file_name))
# delete intermediate png files
for i in range(map_file_list.__len__()):
os.remove(map_file_list[i])
metadata = ([indicator_name], table_name, [min(year_list)])
viz_metadata.append(metadata)
def visualize(self, indicators_to_visualize, computed_indicators):
"""Create a map for the given indicator, save it to the cache
directory's 'indicators' sub-directory."""
# TODO: eliminate this example indicator stuff
example_indicator = computed_indicators[indicators_to_visualize[0]]
source_data = example_indicator.source_data
dataset_to_attribute_map = {}
package_order = source_data.get_package_order()
self._create_input_stores(years=source_data.years)
for name, computed_indicator in computed_indicators.items():
if name not in indicators_to_visualize:
continue
if computed_indicator.source_data != source_data:
raise Exception("result templates in indicator batch must all be the same.")
dataset_name = computed_indicator.indicator.dataset_name
if dataset_name not in dataset_to_attribute_map:
dataset_to_attribute_map[dataset_name] = []
dataset_to_attribute_map[dataset_name].append(name)
viz_metadata = []
for dataset_name, indicator_names in dataset_to_attribute_map.items():
attributes = [
(name, computed_indicators[name].get_computed_dataset_column_name()) for name in indicator_names
]
for year in source_data.years:
SessionConfiguration(new_instance=True, package_order=package_order, in_storage=AttributeCache())
SimulationState().set_cache_directory(source_data.cache_directory)
SimulationState().set_current_time(year)
dataset = SessionConfiguration().get_dataset_from_pool(dataset_name)
dataset.load_dataset()
if dataset.get_coordinate_system() is not None:
dataset.compute_variables(names=dataset.get_coordinate_system())
for indicator_name, computed_name in attributes:
indicator = computed_indicators[indicator_name]
table_data = self.input_stores[year].load_table(
table_name=dataset_name, column_names=[computed_name]
)
if computed_name in table_data:
table_name = self.get_name(
dataset_name=dataset_name, years=[year], attribute_names=[indicator_name]
)
if self.scale:
min_value, max_value = self.scale
else:
min_value, max_value = (None, None)
file_path = os.path.join(
self.storage_location, "anim_" + table_name + "." + MapnikMap.get_file_extension(self)
)
dataset.add_attribute(name=str(computed_name), data=table_data[computed_name])
dataset.plot_map(
name=str(computed_name),
min_value=min_value,
max_value=max_value,
file=str(file_path),
my_title=str(indicator_name),
color_list=self.color_list,
range_list=self.range_list,
label_list=self.label_list,
is_animation=True,
year=year,
resolution=self.resolution,
page_dims=self.page_dims,
map_lower_left=self.map_lower_left,
map_upper_right=self.map_upper_right,
legend_lower_left=self.legend_lower_left,
legend_upper_right=self.legend_upper_right
# filter = where(table_data[computed_name] != -1)
# filter = 'urbansim.gridcell.is_fully_in_water'
)
# metadata = ([indicator_name], table_name, [year])
# viz_metadata.append(metadata)
else:
logger.log_warning("There is no computed indicator %s" % computed_name)
for indicator_name, computed_name in attributes:
self.create_animation(
dataset_name=dataset_name,
year_list=source_data.years,
indicator_name=str(indicator_name),
viz_metadata=viz_metadata,
)
visualization_representations = []
for indicator_names, table_name, years in viz_metadata:
visualization_representations.append(
self._get_visualization_metadata(
computed_indicators=computed_indicators,
indicators_to_visualize=indicator_names,
table_name=table_name,
years=years,
)
)
return visualization_representations
def visualize(self, indicators_to_visualize, computed_indicators):
"""Create a map for the given indicator, save it to the cache
directory's 'indicators' sub-directory."""
#TODO: eliminate this example indicator stuff
example_indicator = computed_indicators[indicators_to_visualize[0]]
source_data = example_indicator.source_data
dataset_to_attribute_map = {}
package_order = source_data.get_package_order()
self._create_input_stores(years=source_data.years)
for name, computed_indicator in computed_indicators.items():
if name not in indicators_to_visualize: continue
if computed_indicator.source_data != source_data:
raise Exception(
'result templates in indicator batch must all be the same.'
)
dataset_name = computed_indicator.indicator.dataset_name
if dataset_name not in dataset_to_attribute_map:
dataset_to_attribute_map[dataset_name] = []
dataset_to_attribute_map[dataset_name].append(name)
viz_metadata = []
for dataset_name, indicator_names in dataset_to_attribute_map.items():
attributes = [
(name,
computed_indicators[name].get_computed_dataset_column_name())
for name in indicator_names
]
for year in source_data.years:
SessionConfiguration(new_instance=True,
package_order=package_order,
in_storage=AttributeCache())
SimulationState().set_cache_directory(
source_data.cache_directory)
SimulationState().set_current_time(year)
dataset = SessionConfiguration().get_dataset_from_pool(
dataset_name)
dataset.load_dataset()
if dataset.get_coordinate_system() is not None:
dataset.compute_variables(
names=dataset.get_coordinate_system())
for indicator_name, computed_name in attributes:
indicator = computed_indicators[indicator_name]
table_data = self.input_stores[year].load_table(
table_name=dataset_name, column_names=[computed_name])
if computed_name in table_data:
table_name = self.get_name(
dataset_name=dataset_name,
years=[year],
attribute_names=[indicator_name])
if self.scale:
min_value, max_value = self.scale
else:
min_value, max_value = (None, None)
file_path = os.path.join(
self.storage_location, 'anim_' + table_name + '.' +
MapnikMap.get_file_extension(self))
dataset.add_attribute(name=str(computed_name),
data=table_data[computed_name])
dataset.plot_map(
name=str(computed_name),
min_value=min_value,
max_value=max_value,
file=str(file_path),
my_title=str(indicator_name),
color_list=self.color_list,
range_list=self.range_list,
label_list=self.label_list,
is_animation=True,
year=year,
resolution=self.resolution,
page_dims=self.page_dims,
map_lower_left=self.map_lower_left,
map_upper_right=self.map_upper_right,
legend_lower_left=self.legend_lower_left,
legend_upper_right=self.legend_upper_right
#filter = where(table_data[computed_name] != -1)
#filter = 'urbansim.gridcell.is_fully_in_water'
)
#metadata = ([indicator_name], table_name, [year])
#viz_metadata.append(metadata)
else:
logger.log_warning(
'There is no computed indicator %s' %
computed_name)
for indicator_name, computed_name in attributes:
self.create_animation(dataset_name=dataset_name,
year_list=source_data.years,
indicator_name=str(indicator_name),
viz_metadata=viz_metadata)
visualization_representations = []
for indicator_names, table_name, years in viz_metadata:
visualization_representations.append(
self._get_visualization_metadata(
computed_indicators=computed_indicators,
indicators_to_visualize=indicator_names,
table_name=table_name,
years=years))
return visualization_representations
