def generate_dataset_report(folder):
'''
Generate a descriptive report on the data included in the use case.
:param folder: Path to the folder containing the NARCCAP example data.
:type folder: str
:returns: Path to the report file.
:rtype: str
'''
ocgis.env.DIR_DATA = folder
rds = parse_narccap_filenames(folder)
rdc = ocgis.RequestDatasetCollection(rds)
headers = ['DID', 'Filenames', 'Variable', 'Time Start', 'Time End']
(fd, name) = tempfile.mkstemp(suffix='.csv')
f = open(name, 'w')
writer = csv.writer(f)
writer.writerow(headers)
for rd in rdc:
logging.info(rd)
ds = NcDataset(rd)
to_write = [
rd.did, [os.path.split(uri)[1] for uri in rd.uri], rd.variable,
ds.temporal.value[0], ds.temporal.value[-1]
]
writer.writerow(to_write)
f.close()
return (name)
## subset data by the boundaries of the USA ####################################
ops = ocgis.OcgOperations(dataset=rd1, geom='state_boundaries', agg_selection=True)
ret = ops.execute()
## access the subsetted values. returned data is stored as a dictionary with the
## key the geometry identifier and the value an OcgCollection.
values = ret[1].variables['tasmax'].value
## to get to the geometries
geoms = ret[1].variables['tasmax'].spatial.value
## time points
times = ret[1].variables['tasmax'].temporal.value
## return two variables ########################################################
rdc = ocgis.RequestDatasetCollection([rd1, rd2])
ops = ocgis.OcgOperations(rdc, snippet=True)
ret = ops.execute()
## aggregate the data by state boundary ########################################
ops = ocgis.OcgOperations(dataset=rd1, geom='state_boundaries', aggregate=True,
snippet=True, spatial_operation='clip')
ret = ops.execute()
## returned data has 51 keys. access the selection geometry for a particular
## collection.
ugeom = ret[1].ugeom
## load a selection geometry from disk #########################################
sc = ocgis.ShpCabinet()
## directory holding data files. this directory and it subdirectories will be
## searched for data.
DIR_DATA = '/home/local/WX/ben.koziol/Dropbox/nesii/project/ocg/bin/climate_data/CanCM4'
## where output data files are written.
DIR_OUTPUT = '/home/local/WX/ben.koziol/Dropbox/nesii/project/ocg/presentation/20130225_caspar_demo/output/01'
## state identifiers in the state shapefile that together comprise the
## southwestern united states.
SW_IDS = [25, 23, 24, 37, 42]
## set to true until final execution
SNIPPET = False
## update environmental variables.
ocgis.env.DIR_DATA = DIR_DATA
ocgis.env.DIR_OUTPUT = DIR_OUTPUT
## pull data files together into a single collections of request datasets.
rdc = ocgis.RequestDatasetCollection()
for ii, filename in enumerate(os.listdir(ocgis.env.DIR_DATA)):
variable = filename.split('_')[0]
alias = variable + '_' + str(ii)
rd = ocgis.RequestDataset(filename, variable, alias=alias)
rdc.update(rd)
## SUBSETTING ##################################################################
ops = ocgis.OcgOperations(rdc,
snippet=SNIPPET,
geom='state_boundaries',
select_ugid=SW_IDS,
output_format='shp',
agg_selection=False,
spatial_operation='clip',