def _get_statements(self):
import requests
from zipfile import ZipFile
from indra.sources.bel.api import process_cbn_jgif_file
import tempfile
cbn_dir = tempfile.mkdtemp('cbn_manager')
logger.info('Retrieving CBN network zip archive')
tmp_zip = os.path.join(cbn_dir, 'cbn_human.zip')
resp = requests.get(self.archive_url)
with open(tmp_zip, 'wb') as f:
f.write(resp.content)
stmts = []
tmp_dir = os.path.join(cbn_dir, 'cbn')
os.mkdir(tmp_dir)
with ZipFile(tmp_zip) as zipf:
logger.info('Extracting archive to %s' % tmp_dir)
zipf.extractall(path=tmp_dir)
logger.info('Processing jgif files')
for jgif in zipf.namelist():
if jgif.endswith('.jgf') or jgif.endswith('.jgif'):
logger.info('Processing %s' % jgif)
pbp = process_cbn_jgif_file(os.path.join(tmp_dir, jgif))
stmts += pbp.statements
uniques, dups = extract_duplicates(stmts,
key_func=KeyFunc.mk_and_one_ev_src)
logger.info("Deduplicating...")
print('\n'.join(str(dup) for dup in dups))
print(len(dups))
return uniques
def test_process_jgif():
test_file_url = 'https://s3.amazonaws.com/bigmech/travis/Hox-2.0-Hs.jgf'
test_file = 'Hox-2.0-Hs.jgf'
request.urlretrieve(url=test_file_url, filename=test_file)
pbp = process_cbn_jgif_file(test_file)
# Clean up
os.remove(test_file)
assert len(pbp.statements) == 26, len(pbp.statements)
assert isinstance(pbp.statements[0], Statement)
assert all(s.evidence[0].source_api == 'bel' for s in pbp.statements)
def test_process_jgif():
test_file_url = 'https://s3.amazonaws.com/bigmech/travis/Hox-2.0-Hs.jgf'
test_file = 'Hox-2.0-Hs.jgf'
request.urlretrieve(url=test_file_url, filename=test_file)
pbp = process_cbn_jgif_file(test_file)
# Clean up
os.remove(test_file)
assert len(pbp.statements) == 26, len(pbp.statements)
assert isinstance(pbp.statements[0], Statement)
assert all(s.evidence[0].source_api == 'bel' for s in pbp.statements)
