def gen_data(fetch=None, **kwargs):
"""Generates data from records or file"""
result = fetch(**kwargs)
if result.get("f"):
f = result.pop("f")
ext = result.pop("ext", "csv")
reader = io.get_reader(ext)
records = reader(f, sanitize=True, **result)
elif result.get("records"):
records = result["records"]
else:
msg = "`fetch` must return a dict with either `records` or `f`."
raise TypeError(msg)
if kwargs.get("normalize"):
normalized = kwargs["normalize"](records, **kwargs)
else:
normalized = records
if kwargs.get("filterer"):
filtered = it.ifilter(partial(kwargs["filterer"], **kwargs), normalized)
else:
filtered = normalized
if kwargs.get("parse"):
parsed = it.imap(partial(kwargs["parse"], **kwargs), filtered)
else:
parsed = filtered
return parsed
def gen_data(fetch=None, **kwargs):
"""Generates data from records or file"""
result = fetch(**kwargs)
if result.get('f'):
f = result.pop('f')
ext = result.pop('ext', 'csv')
reader = io.get_reader(ext)
records = reader(f, sanitize=True, **result)
elif result.get('records'):
records = result['records']
else:
msg = '`fetch` must return a dict with either `records` or `f`.'
raise TypeError(msg)
if kwargs.get('normalize'):
normalized = kwargs['normalize'](records, **kwargs)
else:
normalized = records
if kwargs.get('filterer'):
filtered = it.ifilter(partial(kwargs['filterer'], **kwargs),
normalized)
else:
filtered = normalized
if kwargs.get('parse'):
parsed = it.imap(partial(kwargs['parse'], **kwargs), filtered)
else:
parsed = filtered
return parsed
def update_datastore(self, resource_id, filepath, **kwargs):
verbose = not kwargs.get('quiet')
chunk_rows = kwargs.get('chunksize_rows')
primary_key = kwargs.get('primary_key')
content_type = kwargs.get('content_type')
type_cast = kwargs.get('type_cast')
method = 'upsert' if primary_key else 'insert'
keys = ['aliases', 'primary_key', 'indexes']
try:
extension = p.splitext(filepath)[1].split('.')[1]
except (IndexError, AttributeError):
# no file extension given, e.g., a tempfile
extension = cv.ctype2ext(content_type)
try:
reader = io.get_reader(extension)
except TypeError:
print('Error: plugin for extension `%s` not found!' % extension)
return False
else:
records = reader(filepath, **kwargs)
first = records.next()
keys = first.keys()
records = it.chain([first], records)
if type_cast:
records, results = pr.detect_types(records)
types = results['types']
casted_records = pr.type_cast(records, types)
else:
types = [{'id': key, 'type': 'text'} for key in keys]
casted_records = records
if verbose:
print('Parsed types:')
pprint(types)
create_kwargs = {k: v for k, v in kwargs.items() if k in keys}
if not primary_key:
self.delete_table(resource_id)
insert_kwargs = {'chunksize': chunk_rows, 'method': method}
self.create_table(resource_id, types, **create_kwargs)
args = [resource_id, casted_records]
return self.insert_records(*args, **insert_kwargs)
def update_datastore(self, resource_id, filepath, **kwargs):
verbose = not kwargs.get('quiet')
chunk_rows = kwargs.get('chunksize_rows')
primary_key = kwargs.get('primary_key')
content_type = kwargs.get('content_type')
type_cast = kwargs.get('type_cast')
method = 'upsert' if primary_key else 'insert'
keys = ['aliases', 'primary_key', 'indexes']
try:
extension = p.splitext(filepath)[1].split('.')[1]
except (IndexError, AttributeError):
# no file extension given, e.g., a tempfile
extension = cv.ctype2ext(content_type)
try:
reader = io.get_reader(extension)
except TypeError:
print('Error: plugin for extension `%s` not found!' % extension)
return False
else:
records = reader(filepath, **kwargs)
first = records.next()
keys = first.keys()
records = it.chain([first], records)
if type_cast:
records, results = pr.detect_types(records)
types = results['types']
casted_records = pr.type_cast(records, types)
else:
types = [{'id': key, 'type': 'text'} for key in keys]
casted_records = records
if verbose:
print('Parsed types:')
pprint(types)
create_kwargs = {k: v for k, v in kwargs.items() if k in keys}
if not primary_key:
self.delete_table(resource_id)
insert_kwargs = {'chunksize': chunk_rows, 'method': method}
self.create_table(resource_id, types, **create_kwargs)
args = [resource_id, casted_records]
return self.insert_records(*args, **insert_kwargs)
def test_get_reader(self):
nt.assert_true(callable(io.get_reader("csv")))
with nt.assert_raises(KeyError):
io.get_reader("")
