def test_read_mcf(self):
"""Test reading MCFs, strings or dict"""
# test as file
with self.assertRaises(IOError):
mcf = read_mcf(get_abspath('../404.yml'))
mcf = read_mcf(get_abspath('../sample.yml'))
self.assertIsInstance(mcf, dict, 'Expected dict')
# test MCF section
self.assertTrue('version' in mcf['mcf'], 'Expected MCF version')
self.assertTrue('metadata' in mcf, 'Expected metadata section')
# test as string
with open(get_abspath('../sample.yml')) as fh:
mcf_string = fh.read()
mcf = read_mcf(mcf_string)
self.assertTrue('metadata' in mcf, 'Expected metadata section')
# test as dict
mcf_dict = yaml.load(mcf_string)
mcf = read_mcf(mcf_dict)
self.assertTrue('metadata' in mcf, 'Expected metadata section')
def test_read_mcf(self):
"""Test reading MCFs, strings or dict"""
# test as file
with self.assertRaises(IOError):
mcf = read_mcf(get_abspath('../404.yml'))
mcf = read_mcf(get_abspath('../sample.yml'))
self.assertIsInstance(mcf, dict, 'Expected dict')
# test MCF section
self.assertTrue('version' in mcf['mcf'], 'Expected MCF version')
self.assertTrue('metadata' in mcf, 'Expected metadata section')
# test as string
with open(get_abspath('../sample.yml')) as fh:
mcf_string = fh.read()
mcf = read_mcf(mcf_string)
self.assertTrue('metadata' in mcf, 'Expected metadata section')
# test as dict
mcf_dict = yaml.load(mcf_string)
mcf = read_mcf(mcf_dict)
self.assertTrue('metadata' in mcf, 'Expected metadata section')
def test_mcf_version(self):
"""Test MCF version validation"""
with self.assertRaises(MCFReadError):
read_mcf(get_abspath('missing-version.yml'))
with self.assertRaises(MCFReadError):
read_mcf(get_abspath('bad-version.yml'))
def test_mcf_version(self):
"""Test MCF version validation"""
with self.assertRaises(MCFReadError):
read_mcf(get_abspath('missing-version.yml'))
with self.assertRaises(MCFReadError):
read_mcf(get_abspath('bad-version.yml'))
def test_json_output_schema(self):
"""test JSON as dict-based output schemas"""
mcf = read_mcf(get_abspath('../sample.yml'))
record = OGCAPIRecordOutputSchema().write(mcf)
self.assertIsInstance(record, str)
mcf = read_mcf(get_abspath('../sample.yml'))
record = OGCAPIRecordOutputSchema().write(mcf, stringify=False)
self.assertIsInstance(record, dict)
def test_wmo_wigos(self):
"""test WMO WIGOS Metadata support"""
mcf = read_mcf(get_abspath('../sample-wmo-wigos.yml'))
self.assertEqual(len(mcf['facility'].keys()), 1)
self.assertEqual(
len(mcf['facility']['first_station']['spatiotemporal']), 1)
def test_19139_2(self):
"""test ISO 19139-2 Metadata support"""
mcf = read_mcf(get_abspath('../sample.yml'))
self.assertIn('acquisition', mcf)
self.assertIn('platforms', mcf['acquisition'])
self.assertIn('instruments', mcf['acquisition']['platforms'][0])
def test_broken_yaml(self):
"""test against broken YAML"""
iso_os = ISO19139OutputSchema()
with self.assertRaises(MCFReadError):
render_j2_template(read_mcf(get_abspath('broken-yaml.yml')),
iso_os.template_dir)
def test_wmo_wigos(self):
"""test WMO WIGOS Metadata support"""
mcf = read_mcf(get_abspath('../sample-wmo-wigos.yml'))
self.assertEqual(len(mcf['facility'].keys()), 1)
self.assertEqual(
len(mcf['facility']['first_station']['spatiotemporal']), 1)
def test_pre1900_dates(self):
"""test datestrings that are pre-1900"""
iso_os = ISO19139OutputSchema()
xml = render_j2_template(read_mcf(get_abspath('dates-pre-1900.yml')),
iso_os.template_dir)
self.assertIsInstance(xml, str, 'Expected unicode string')
def load_collection_level_metadata(self):
logger.debug('Loading collection level metadata')
for clm in os.listdir(COLLECTION_LEVEL_METADATA):
logger.debug(f'collection metadata file: {clm}')
clm_ = os.path.join(COLLECTION_LEVEL_METADATA, clm)
clm_mcf = read_mcf(clm_)
clm_iso = ISO19139OutputSchema().write(clm_mcf)
logger.debug(f'Upserting metadata: {clm_}')
self._parse_and_upsert_metadata(clm_iso)
def test_validate_mcf(self):
"""test MCF validation"""
mcf = read_mcf(get_abspath('../sample.yml'))
instance = json.loads(json.dumps(mcf, default=json_serial))
is_valid = validate_mcf(instance)
assert is_valid
# validated nested MCF
mcf = read_mcf(get_abspath('./sample-child.yml'))
instance = json.loads(json.dumps(mcf, default=json_serial))
is_valid = validate_mcf(instance)
assert is_valid
with self.assertRaises(MCFValidationError):
is_valid = validate_mcf({'foo': 'bar'})
def test_nested_mcf(self):
"""test nested mcf support"""
mcf = read_mcf(get_abspath('child.yml'))
self.assertEqual(mcf['metadata']['identifier'], 1234,
'Expected specific identifier')
self.assertEqual(mcf['identification']['title_en'], 'title in English',
'Expected specific title')
self.assertIsInstance(mcf, dict, 'Expected dict')
def test_pretty_print(self):
"""Test pretty-printing"""
iso_os = ISO19139OutputSchema()
xml = render_j2_template(read_mcf(get_abspath('../sample.yml')),
iso_os.template_dir)
xml2 = pretty_print(xml)
self.assertIsInstance(xml2, str, 'Expected unicode string')
self.assertEqual(xml2[-1], '>', 'Expected closing bracket')
self.assertTrue(xml2.startswith('<?xml'), 'Expected XML declaration')
def test_mcf_model(self):
"""test mcf model and types"""
mcf = read_mcf(get_abspath('../sample.yml'))
self.assertIsInstance(mcf['identification']['dates'], dict,
'Expected list')
self.assertIsInstance(mcf['identification']['keywords'], dict,
'Expected dict')
self.assertIsInstance(mcf['identification']['topiccategory'], list,
'Expected list')
self.assertIsInstance(mcf['contact'], dict, 'Expected dict')
self.assertIsInstance(mcf['distribution'], dict, 'Expected dict')
def test_mcf_model(self):
"""test mcf model and types"""
mcf = read_mcf(get_abspath('../sample.yml'))
self.assertIsInstance(mcf['identification']['dates'], dict,
'Expected list')
self.assertIsInstance(mcf['identification']['keywords'], dict,
'Expected dict')
self.assertIsInstance(mcf['identification']['topiccategory'], list,
'Expected list')
self.assertIsInstance(mcf['contact'], dict, 'Expected dict')
self.assertIsInstance(mcf['distribution'], dict, 'Expected dict')
def test_nested_mcf(self):
"""test nested mcf support"""
mcf = read_mcf(get_abspath('child.yml'))
self.assertEqual(mcf['metadata']['identifier'], 1234,
'Expected specific identifier')
self.assertEqual(mcf['identification']['title_en'],
'title in English',
'Expected specific title')
self.assertIsInstance(mcf, dict, 'Expected dict')
def test_deep_nested_mcf(self):
"""test deep nested mcf support"""
mcf = read_mcf(get_abspath('deep-nest-child.yml'))
self.assertEqual(mcf['metadata']['identifier'], 'MYID',
'Expected specific identifier')
self.assertEqual(mcf['identification']['title_en'], 'child title',
'Expected specific title')
self.assertEqual(mcf['distribution']['waf']['url'],
'http://dd.meteo.gc.ca', 'Expected specific URL')
self.assertEqual(mcf['contact']['main']['positionname'],
'Senior Systems Scientist', 'Expected specific name')
def test_deep_nested_mcf(self):
"""test deep nested mcf support"""
mcf = read_mcf(get_abspath('deep-nest-child.yml'))
self.assertEqual(mcf['metadata']['identifier'], 'MYID',
'Expected specific identifier')
self.assertEqual(mcf['identification']['title_en'], 'child title',
'Expected specific title')
self.assertEqual(mcf['distribution']['waf']['url'],
'http://dd.meteo.gc.ca', 'Expected specific URL')
self.assertEqual(mcf['contact']['main']['positionname'],
'Senior Systems Scientist', 'Expected specific name')
def test_nested_mcf(self):
"""test nested mcf support"""
mcf = read_mcf(get_abspath('child.yml'))
self.assertEqual(mcf['metadata']['identifier'], 5678,
'Expected specific identifier')
self.assertEqual(mcf['distribution']['waf']['type'], 'WWW:LINK',
'Expected specific distribution type')
self.assertEqual(mcf['distribution']['waf']['url'],
'http://example.org/waf',
'Expected specific distribution url')
self.assertEqual(mcf['metadata']['datestamp'],
datetime.date(2011, 11, 11),
'Expected specific metadata datestamp')
self.assertIsInstance(mcf, dict, 'Expected dict')
def test_nested_mcf(self):
"""test nested mcf support"""
mcf = read_mcf(get_abspath('child.yml'))
self.assertEqual(mcf['metadata']['identifier'], 5678,
'Expected specific identifier')
self.assertEqual(mcf['distribution']['waf']['type'], 'WWW:LINK',
'Expected specific distribution type')
self.assertEqual(mcf['distribution']['waf']['url'],
'http://example.org/waf',
'Expected specific distribution url')
self.assertEqual(mcf['metadata']['datestamp'],
datetime.date(2011, 11, 11),
'Expected specific metadata datestamp')
self.assertIsInstance(mcf, dict, 'Expected dict')
def test_render_j2_template(self):
"""test template rendering"""
test_mcf_paths = [
'../sample.yml', 'unilingual.yml',
'nil-identification-language.yml'
]
for mcf_path in test_mcf_paths:
iso_os = ISO19139OutputSchema()
# working template directory
xml = render_j2_template(read_mcf(get_abspath(mcf_path)),
iso_os.template_dir)
self.assertIsInstance(xml, str, 'Expected unicode string')
# no template directory or local schema provided
with self.assertRaises(RuntimeError):
render_j2_template(read_mcf(get_abspath(mcf_path)))
# bad template directory provided
with self.assertRaises(RuntimeError):
xml = render_j2_template(read_mcf(get_abspath(mcf_path)),
'bad_dir')
# bad j2 template_dir provided
with self.assertRaises(RuntimeError):
xml = render_j2_template(read_mcf(get_abspath(mcf_path)),
template_dir='/bad_schema/path')
# good j2 template_dir provided
xml = render_j2_template(
read_mcf(get_abspath(mcf_path)),
template_dir=get_abspath('sample_schema_j2')) # noqa
# good sample output schema
s_os = SampleOutputSchema()
_ = s_os.write(read_mcf(get_abspath(mcf_path)))
def _describe_file(filepath):
"""
Helper function to describe a geospatial data
First checks if a sidecar mcf file is available, if so uses that
if not, script will parse the file to retrieve some info from the file
:param filepath: path to file
:returns: `dict` of GeoJSON item
"""
content = {'bbox': None, 'geometry': None, 'properties': {}}
mcf_file = '{}.yml'.format(os.path.splitext(filepath)[0])
if os.path.isfile(mcf_file):
try:
from pygeometa.core import read_mcf, MCFReadError
from pygeometa.schemas.stac import STACItemOutputSchema
md = read_mcf(mcf_file)
stacjson = STACItemOutputSchema.write(STACItemOutputSchema, md)
stacdata = loads(stacjson)
for k, v in stacdata.items():
content[k] = v
except ImportError:
LOGGER.debug('pygeometa not found')
except MCFReadError as err:
LOGGER.warning('MCF error: {}'.format(err))
else:
LOGGER.debug('No mcf found at: {}'.format(mcf_file))
if content['geometry'] is None and content['bbox'] is None:
try:
import rasterio
from rasterio.crs import CRS
from rasterio.warp import transform_bounds
except ImportError as err:
LOGGER.warning('rasterio not found')
LOGGER.warning(err)
return content
try:
import fiona
except ImportError as err:
LOGGER.warning('fiona not found')
LOGGER.warning(err)
return content
try: # raster
LOGGER.debug('Testing raster data detection')
d = rasterio.open(filepath)
content['bbox'] = [
d.bounds.left, d.bounds.bottom, d.bounds.right, d.bounds.top
]
content['geometry'] = {
'type':
'Polygon',
'coordinates': [[[d.bounds.left, d.bounds.bottom],
[d.bounds.left, d.bounds.top],
[d.bounds.right, d.bounds.top],
[d.bounds.right, d.bounds.bottom],
[d.bounds.left, d.bounds.bottom]]]
}
for k, v in d.tags(d.count).items():
content['properties'][k] = v
if k in ['GRIB_REF_TIME']:
value = int(v.split()[0])
datetime_ = datetime.fromtimestamp(value)
content['properties']['datetime'] = datetime_.isoformat(
) + 'Z' # noqa
except rasterio.errors.RasterioIOError:
try:
LOGGER.debug('Testing vector data detection')
d = fiona.open(filepath)
scrs = CRS(d.crs)
if scrs.to_epsg() is not None and scrs.to_epsg() != 4326:
tcrs = CRS.from_epsg(4326)
bnds = transform_bounds(scrs, tcrs, d.bounds[0],
d.bounds[1], d.bounds[2],
d.bounds[3])
content['properties']['projection'] = scrs.to_epsg()
else:
bnds = d.bounds
if d.schema['geometry'] not in [None, 'None']:
content['bbox'] = [bnds[0], bnds[1], bnds[2], bnds[3]]
content['geometry'] = {
'type':
'Polygon',
'coordinates':
[[[bnds[0], bnds[1]], [bnds[0], bnds[3]],
[bnds[2], bnds[3]], [bnds[2], bnds[1]],
[bnds[0], bnds[1]]]]
}
for k, v in d.schema['properties'].items():
content['properties'][k] = v
if d.driver == 'ESRI Shapefile':
id_ = os.path.splitext(os.path.basename(filepath))[0]
content['assets'] = {}
for suffix in ['shx', 'dbf', 'prj']:
fullpath = '{}.{}'.format(
os.path.splitext(filepath)[0], suffix)
if os.path.exists(fullpath):
filectime = file_modified_iso8601(fullpath)
filesize = os.path.getsize(fullpath)
content['assets'][suffix] = {
'href': './{}.{}'.format(id_, suffix),
'created': filectime,
'file:size': filesize
}
except fiona.errors.DriverError:
LOGGER.debug('Could not detect raster or vector data')
return content
def indexDir(dir, dir_out, dir_out_mode, mode, dbtype, profile, db):
if not dir:
dir = "."
if not dir_out_mode or dir_out_mode not in ["flat", "nested"]:
dir_out_mode = "flat"
if not mode or mode not in ["init", "update", "export"]:
mode = "init"
if not dbtype or dbtype not in ["path", "sqlite", "postgres"]:
dbtype = "path"
if not db:
db = dir
if not profile or profile not in ["iso19139", "dcat"]:
profile = "iso19139"
print(mode + ' metadata in ' + dir + ' as ' + profile + ' in ' + db)
if mode == "export":
if dbtype == 'sqlite':
dir_out = os.path.join(dir_out, db)
createIndexIfDoesntExist(dir_out)
elif dbtype == "path":
if not os.path.exists(dir_out):
print('creating out folder ' + dir_out)
os.makedirs(dir_out)
else:
print("postgis not supported")
# core metadata gets populated by merging the index.yaml content from parent folders
coreMetadata = {}
# identify if there is a path change
prvPath = "dummy"
for path, dirs, files in os.walk(dir):
if mode == 'export':
# if dir has index.yaml merge it to parent
f = os.path.join(path, 'index.yaml')
if os.path.exists(f):
if prvPath != path:
print('Indexing path ' + path)
prvPath = path
with open(os.path.join(f), mode="r",
encoding="utf-8") as yf:
pathMetadata = yaml.load(yf, Loader=SafeLoader)
pathMetadata.pop('index')
pathMetadata.pop('mode')
dict_merge(pathMetadata, coreMetadata)
coreMetadata = pathMetadata
else:
print(f + ' does not exist') # create it?
for file in files:
fname = os.path.join(path, file)
if '.' in file:
base, extension = file.rsplit('.', 1)
if extension.lower() in SPATIAL_FILE_TYPES:
print('Indexing file ' + fname)
yf = os.path.join(path, base + '.yaml')
if (mode == 'update'
or (not os.path.exists(yf) and mode == 'init')):
# mode init for spatial files without metadata or update
cnt = indexSpatialFile(fname, extension)
if (mode == 'update'
): # keep manual changes on the original
try:
with open(os.path.join(yf),
mode="r",
encoding="utf-8") as f:
orig = yaml.load(f, Loader=SafeLoader)
dict_merge(
orig, cnt
) # or should we overwrite some values from cnt explicitely?
cnt = orig
except Exception as e:
print('Failed to merge original:', f, e)
md = asPGM(cnt)
# write yf
try:
with open(os.path.join(yf), 'w') as f:
yaml.dump(md, f, sort_keys=False)
except Exception as e:
print('Failed to dump yaml:', e)
elif mode == 'export':
try:
with open(os.path.join(yf),
mode="r",
encoding="utf-8") as f:
cnf = yaml.load(f, Loader=SafeLoader)
dict_merge(cnf, coreMetadata)
if dbtype == 'sqlite' or dbtype == 'postgres':
insert_or_update(cnt, dir_out)
elif dbtype == "path":
#load yml as mcf
md = read_mcf(cnf)
#yaml to iso/dcat
if schemaPath and os.path.exists(
schemaPath):
print('Using schema', schemaPath)
xml_string = render_j2_template(
md,
template_dir="{}/iso19139".format(
schemaPath))
else:
print('Using default iso19139 schema')
iso_os = ISO19139OutputSchema()
xml_string = iso_os.write(md)
if dir_out_mode == "flat":
pth = os.path.join(
dir_out,
cnf['metadata']['identifier'] +
'.xml')
else:
pth = os.path.join(path, base + '.xml')
print("write to file: " + pth)
with open(pth, 'w+') as ff:
ff.write(xml_string)
print('iso19139 xml generated at ' +
pth)
except Exception as e:
print('Failed to create xml:', e)
else:
None
# print('Skipping {}, not approved file type: {}'.format(fname, extension))
else:
None
