def add_resource(mt_file, ref, cache):
"""Add a resources entry, downloading the intuiting the file, replacing entries with
the same reference"""
if isinstance(mt_file, MetapackDoc):
doc = mt_file
else:
doc = MetapackDoc(mt_file)
if 'Resources' not in doc:
doc.new_section('Resources')
doc['Resources'].args = [
e for e in set(doc['Resources'].args +
['Name', 'StartLine', 'HeaderLines', 'Encoding']) if e
]
seen_names = set()
u = parse_app_url(ref)
# The web and file URLs don't list the same.
if u.proto == 'file':
entries = u.list()
else:
entries = [ssu for su in u.list() for ssu in su.list()]
for e in entries:
add_single_resource(doc, e, cache=cache, seen_names=seen_names)
write_doc(doc, mt_file)
def process_schemas(mt_file, resource=None, cache=None, clean=False, report_found=True, force=False, min_rows=5000,
allow_codes=True):
from metapack import MetapackDoc, MetapackResourceUrl, MetapackDocumentUrl
if isinstance(mt_file, MetapackDoc):
doc = mt_file
write_doc_to_file = False
else:
doc = MetapackDoc(mt_file)
write_doc_to_file = True
try:
if clean:
doc['Schema'].clean()
else:
doc['Schema']
except KeyError:
doc.new_section('Schema', ['DataType', 'AltName', 'Description'])
schemas_processed = 0
for r in doc['Resources'].find('Root.Resource'):
if resource and r.name != resource:
continue
schema_term = r.schema_term
col_count = len(list(r.columns()))
datatype_count = sum(1 for c in r.columns() if c['datatype'])
if schema_term and col_count == datatype_count and force is False:
if report_found:
prt("Found table for '{}'; skipping".format(r.schema_name))
continue
if col_count != datatype_count:
prt("Found table for '{}'; but {} columns don't have datatypes"
.format(r.schema_name, col_count - datatype_count))
schemas_processed += 1
rr = r.resolved_url
rmtree(get_materialized_data_cache(doc), ignore_errors=True)
if isinstance(rr, MetapackDocumentUrl):
warn('{} is a MetapackDocumentUrl; skipping', r.name)
elif isinstance(rr, MetapackResourceUrl):
_process_metapack_resource(doc, r, force)
else:
_process_normal_resource(doc, r, force, skip_start=min_rows, allow_codes=allow_codes)
if write_doc_to_file and schemas_processed:
write_doc(doc, mt_file)
def run_url_add(args):
"""Add a resources entry, downloading the intuiting the file, replacing entries with
the same reference"""
m = MetapackCliMemo(args, downloader)
update_name(m.mt_file, fail_on_missing=False, report_unchanged=False)
if isinstance(m.mt_file, MetapackDoc):
doc = m.mt_file
else:
doc = MetapackDoc(m.mt_file)
if 'Resources' not in doc:
doc.new_section('Resources')
doc['Resources'].args = [
e for e in set(doc['Resources'].args +
['Name', 'StartLine', 'HeaderLines', 'Encoding']) if e
]
seen_names = set()
u = parse_app_url(args.url)
# The web and file URLs don't list the same.
if u.proto == 'file':
entries = u.list()
else:
entries = [ssu for su in u.list() for ssu in su.list()]
errors = []
for e in entries:
if not add_single_resource(
doc, e, cache=m.cache, seen_names=seen_names):
errors.append(e)
if errors:
prt()
warn("Found, but failed to add these urls:")
for e in errors:
print(' ', e)
write_doc(doc)
