def update(repos, verbose=True):
ecoregions = [(er['properties']['eco_code'], shape(er['geometry']))
for er in jsonlib.load(
data_file('ecoregions.json', repos=repos))['features']
if er['geometry']
and er['properties']['eco_code'] not in INVALID_ECO_CODES]
with CsvData('distribution', repos=repos) as data:
res = {i.id: i for i in data.items}
occurrence_data = list(
data_file('external', 'gbif', repos=repos).glob('*.json'))
if verbose: # pragma: no cover
occurrence_data = tqdm(occurrence_data)
for fname in occurrence_data:
sid = fname.stem
d = res.get(sid, Distribution(sid, '', ''))
if not d.countries__ids or not d.ecoregions__ids:
occurrences = jsonlib.load(fname).get('results', [])
if not d.ecoregions__ids:
d.ecoregions__ids = list(match(occurrences, ecoregions))
if not d.countries__ids:
d.countries__ids = list(
r.get('countryCode') for r in occurrences)
res[sid] = d
data.items = [res[key] for key in sorted(res.keys())]
def upload_images(args):
"""
tsammalex upload_images path/to/cdstar/catalog
"""
images_path = data_file('images.csv', repos=args.tsammalex_data)
staged_images_path = data_file('staged_images.csv',
repos=args.tsammalex_data)
checksums = set(
d.id for d in models.CsvData('images', repos=args.tsammalex_data))
providers = [prov(args.tsammalex_data) for prov in PROVIDERS]
with MediaCatalog('cdstar.json',
repos=args.tsammalex_data,
json_opts=dict(indent=4)) as mcat:
with Catalog(args.args[0],
cdstar_url=os.environ['CDSTAR_URL'],
cdstar_user=os.environ['CDSTAR_USER'],
cdstar_pwd=os.environ['CDSTAR_PWD']) as cat:
for item in models.CsvData('staged_images',
repos=args.tsammalex_data):
for provider in providers:
if item in provider:
img = provider.retrieve(item, cat, checksums, mcat)
if img:
try:
add_rows(images_path, img.csv_row())
except:
print(img)
raise
filter_rows(staged_images_path,
lambda d: d['id'] != item.id)
break
def upload_images(args):
"""
tsammalex upload_images path/to/cdstar/catalog
"""
images_path = data_file('images.csv', repos=args.tsammalex_data)
staged_images_path = data_file('staged_images.csv', repos=args.tsammalex_data)
checksums = set(d.id for d in models.CsvData('images', repos=args.tsammalex_data))
providers = [prov(args.tsammalex_data) for prov in PROVIDERS]
with MediaCatalog(
'cdstar.json', repos=args.tsammalex_data, json_opts=dict(indent=4)) as mcat:
with Catalog(
args.args[0],
cdstar_url=os.environ['CDSTAR_URL'],
cdstar_user=os.environ['CDSTAR_USER'],
cdstar_pwd=os.environ['CDSTAR_PWD']) as cat:
for item in models.CsvData('staged_images', repos=args.tsammalex_data):
for provider in providers:
if item in provider:
img = provider.retrieve(item, cat, checksums, mcat)
if img:
try:
add_rows(images_path, img.csv_row())
except:
print(img)
raise
filter_rows(staged_images_path, lambda d: d['id'] != item.id)
break
def test():
if not REPOS.exists():
return
data = {
n: OrderedDict([(item.id, item)
for item in models.CsvData(n, on_error=error)])
for n in CSV
}
data['ecoregions'] = {}
for ecoregion in jsonlib.load(data_file('ecoregions.json'))['features']:
data['ecoregions'][ecoregion['properties']['eco_code']] = ecoregion
data['refs'] = {}
with data_file('sources.bib').open(encoding='utf8') as fp:
for line in fp:
match = BIB_ID_PATTERN.match(line.strip())
if match:
data['refs'][match.group('id')] = 1
data['countries'] = {country.alpha2: country for country in countries}
for name in ['names', 'taxa']:
for line, item in enumerate(data[name].values()):
for ref in item.refs__ids:
if '[' in ref:
source_id, pages = ref.split('[', 1)
if not pages.endswith(']'): # pragma: no cover
error('invalid reference %s' % (ref, ), name, line + 2)
else:
source_id = ref
if source_id not in data['refs']: # pragma: no cover
error('invalid id referenced: %s' % (source_id, ), name,
line + 2)
for name, model in [(n, getattr(models, n.capitalize())) for n in CSV]:
for line, item in enumerate(data[name].values()):
for col in [f.name for f in attr.fields(model)]:
if '__' in col:
ref, cardinality = col.split('__', 1)
#if ref not in data:
# continue
ids = getattr(item, col)
if cardinality == 'id':
assert not isinstance(ids, list)
ids = [ids]
for v in ids:
if ref not in data:
raise ValueError(ref) # pragma: no cover
if ref == 'refs' and '[' in v:
v = v.split('[')[0]
if v not in data[ref]: # pragma: no cover
error('invalid %s id referenced: %s' % (ref, v),
name, line + 2)
if not SUCCESS: # pragma: no cover
raise ValueError('integrity checks failed!')
def test():
if not REPOS.exists():
return
data = {
n: OrderedDict([(item.id, item) for item in models.CsvData(n, on_error=error)])
for n in CSV}
data['ecoregions'] = {}
for ecoregion in jsonlib.load(data_file('ecoregions.json'))['features']:
data['ecoregions'][ecoregion['properties']['eco_code']] = ecoregion
data['refs'] = {}
with data_file('sources.bib').open(encoding='utf8') as fp:
for line in fp:
match = BIB_ID_PATTERN.match(line.strip())
if match:
data['refs'][match.group('id')] = 1
data['countries'] = {country.alpha2: country for country in countries}
for name in ['names', 'taxa']:
for line, item in enumerate(data[name].values()):
for ref in item.refs__ids:
if '[' in ref:
source_id, pages = ref.split('[', 1)
if not pages.endswith(']'): # pragma: no cover
error('invalid reference %s' % (ref,), name, line + 2)
else:
source_id = ref
if source_id not in data['refs']: # pragma: no cover
error('invalid id referenced: %s' % (source_id,), name, line + 2)
for name, model in [(n, getattr(models, n.capitalize())) for n in CSV]:
for line, item in enumerate(data[name].values()):
for col in [f.name for f in attr.fields(model)]:
if '__' in col:
ref, cardinality = col.split('__', 1)
#if ref not in data:
# continue
ids = getattr(item, col)
if cardinality == 'id':
assert not isinstance(ids, list)
ids = [ids]
for v in ids:
if ref not in data:
raise ValueError(ref) # pragma: no cover
if ref == 'refs' and '[' in v:
v = v.split('[')[0]
if v not in data[ref]: # pragma: no cover
error(
'invalid %s id referenced: %s' % (ref, v), name, line + 2)
if not SUCCESS: # pragma: no cover
raise ValueError('integrity checks failed!')
def test_JsonData(self):
from pytsammalex.util import JsonData, data_file
tmpdir = create_repos(self.tmp_path())
with JsonData('test.json', repos=tmpdir) as jdat:
jdat['a'] = 1
self.assertTrue(data_file('test.json', repos=tmpdir).exists())
with JsonData('test.json', repos=tmpdir) as jdat:
self.assertEqual(len(jdat), 1)
self.assertEqual(jdat['a'], 1)
def test_json_data(tmpdir):
tmp_ = create_repos(tmpdir)
with JsonData('test.json', repos=Path(tmp_)) as jdat:
jdat['a'] = 1
assert (data_file('test.json', repos=Path(tmp_)).exists() is True)
with JsonData('test.json', repos=Path(tmp_)) as jdat:
assert (len(jdat) == 1)
assert (jdat['a'] == 1)
def cached_metadata(self, sid, id=None, name=None, refresh=False):
if data_file('external', self.name, repos=self.repos).is_dir():
fname = data_file('external', self.name, sid + '.json', repos=self.repos)
if not fname.exists() or refresh:
try:
data = self.metadata(id or self.identify(name))
except: # pragma: no cover
data = None
if not data:
return # pragma: no cover
jsonlib.dump(data, fname)
return data
return jsonlib.load(fname)
if sid not in self.items or refresh:
try:
self.items[sid] = self.metadata(id or self.identify(name))
except:
return
return self.items[sid]
def update(repos, log):
ecoregions = [
(er['properties']['eco_code'], shape(er['geometry']))
for er in jsonlib.load(data_file('ecoregions.json', repos=repos))['features']
if er['geometry'] and er['properties']['eco_code'] not in INVALID_ECO_CODES]
with CsvData('distribution', repos=repos) as data:
res = {i.id: i for i in data.items}
occurrence_data = list(data_file('external', 'gbif', repos=repos).glob('*.json'))
for fname in tqdm(occurrence_data):
sid = fname.stem
d = res.get(sid, Distribution(sid, '', ''))
if not d.countries__ids or not d.ecoregions__ids:
occurrences = jsonlib.load(fname).get('results', [])
if not d.ecoregions__ids:
d.ecoregions__ids = list(match(occurrences, ecoregions, log))
if not d.countries__ids:
d.countries__ids = list(r.get('countryCode') for r in occurrences)
res[sid] = d
data.items = [res[key] for key in sorted(res.keys())]
def cached_metadata(self, sid, id=None, name=None, refresh=False):
if data_file('external', self.name, repos=self.repos).is_dir():
fname = data_file('external',
self.name,
sid + '.json',
repos=self.repos)
if not fname.exists() or refresh:
try:
data = self.metadata(id or self.identify(name))
except: # pragma: no cover
data = None
if not data:
return # pragma: no cover
jsonlib.dump(data, fname)
return data
return jsonlib.load(fname)
if sid not in self.items or refresh:
try:
self.items[sid] = self.metadata(id or self.identify(name))
except:
return
return self.items[sid]
def identify(self, item):
p = data_file('staged_images', item.id, repos=self.repos)
if p.exists():
return p
