def run(args):
if args.agent:
run_agent(args.agent,
lang=args.lang,
hg=hgraph(args.hg),
infile=args.infile,
url=args.url,
sequence=args.sequence)
elif args.system:
run_system(args.system,
lang=args.lang,
hg=hgraph(args.hg),
infile=args.infile,
url=args.url,
sequence=args.sequence)
else:
raise RuntimeError('Either agent or system must be specified.')
def run(args):
if args.agent:
run_agent(args.agent,
lang=args.lang,
parser_class=args.parser,
hg=hgraph(args.hg),
infile=args.infile,
indir=args.indir,
url=args.url,
sequence=args.sequence,
corefs=args.corefs)
elif args.system:
run_system(args.system,
lang=args.lang,
parser_class=args.parser,
hg=hgraph(args.hg),
infile=args.infile,
indir=args.indir,
url=args.url,
sequence=args.sequence,
corefs=args.corefs)
else:
raise RuntimeError('Either agent or system must be specified.')
def setUp(self):
self.hg = hgraph('test.hg')
self.hg.destroy()
concepts = []
concepts.append(hedge('paris/C'))
concepts.append(hedge('(of/B city/C paris/C)'))
concepts.append(hedge('(of/B capital/C france/C)'))
concepts.append(hedge('berlin/C'))
concepts.append(hedge('(of/B city/C berlin/C)'))
concepts.append(hedge('(of/B capital/C germany/C)'))
self.concepts = concepts
self.hg.add('(love/P i/C (of/B city/C paris/C))')
self.hg.add('(hate/P i/C (of/B city/C paris/C))')
def setUp(self):
self.hg = hgraph('test.hg')
def cli():
_show_logo()
parser = argparse.ArgumentParser()
parser.add_argument('command', type=str, help='command to execute')
parser.add_argument('--agent', type=str, help='agent name', default=None)
parser.add_argument('--col', type=str, help='table column', default=None)
parser.add_argument('--corefs',
help='perform coreference resolution',
action='store_true')
parser.add_argument('--fields', type=str, help='field names', default=None)
parser.add_argument('--hg',
type=str,
help='hypergraph db',
default='gb.db')
parser.add_argument('--indir',
type=str,
help='input directory',
default=None)
parser.add_argument('--infile', type=str, help='input file', default=None)
parser.add_argument('--lang', type=str, help='language', default=None)
parser.add_argument('--outdir',
type=str,
help='output directory',
default=None)
parser.add_argument('--outfile',
type=str,
help='output file',
default=None)
parser.add_argument('--parser', type=str, help='parser', default=None)
parser.add_argument('--pattern',
type=str,
help='edge pattern',
default='*')
parser.add_argument('--sequence',
type=str,
help='sequence name',
default=None)
parser.add_argument('--show_namespaces',
help='show namespaces',
action='store_true')
parser.add_argument('--system',
type=str,
help='agent system file',
default=None)
parser.add_argument('--text',
type=str,
help='text identifier',
default='title')
parser.add_argument('--training_data',
type=str,
help='training data for ML tasks',
default=None)
parser.add_argument('--url', type=str, help='url', default=None)
args = parser.parse_args()
# determine language
if args.parser:
plang = parser_lang(args.parser)
if args.lang:
if args.lang != plang:
msg = 'specified language ({}) and parser language ({}) do '\
'not match'.format(args.lang, plang)
error_msg()
sys.exit(-1)
else:
args.lang = plang
# if not lang or parser is specified, default to 'en'
elif not args.lang:
args.lang = 'en'
print(colored('{}\n'.format('command: {}'.format(args.command)), 'white'))
if args.agent:
print('agent: {}'.format(args.agent))
if args.col:
print('column: {}'.format(args.col))
if args.corefs:
print('coreferences: {}'.format(args.corefs))
if args.hg:
print('hypergraph: {}'.format(args.hg))
if args.infile:
print('input file: {}'.format(args.infile))
if args.lang:
print('language: {}'.format(args.lang))
if args.parser:
print('parser: {}'.format(args.parser))
if args.outfile:
print('output file: {}'.format(args.outfile))
if args.sequence:
print('sequence: {}'.format(args.sequence))
if args.system:
print('system: {}'.format(args.system))
if args.training_data:
print('training data: {}'.format(args.training_data))
if args.url:
print('url: {}'.format(args.url))
print()
if args.command == 'create':
hgraph(args.hg)
print('Hypergraph database created.')
elif args.command == 'export':
print('exporting hypergraph...')
hg = hgraph(args.hg)
n = 0
with open(args.outfile, 'w') as f:
for edge, attributes in hg.all_attributes():
row = [str(edge), attributes]
f.write('{}\n'.format(json.dumps(row, ensure_ascii=False)))
n += 1
print('{} edges exported.'.format(n))
elif args.command == 'import':
print('importing hypergraph...')
hg = hgraph(args.hg)
n = 0
with open(args.infile, 'r') as f:
for line in f:
edge_str, attributes = json.loads(line)
hg.add_with_attributes(hedge(edge_str), attributes)
n += 1
print('{} edges imported.'.format(n))
elif args.command == 'txt':
TxtReader(args.infile,
hg=hgraph(args.hg),
sequence=args.sequence,
lang=args.lang,
corefs=args.corefs).read()
elif args.command == 'dir':
DirReader(args.indir,
hg=hgraph(args.hg),
sequence=args.sequence,
lang=args.lang,
corefs=args.corefs).read()
elif args.command == 'csv':
CsvReader(args.infile,
args.col,
hg=hgraph(args.hg),
sequence=args.sequence,
lang=args.lang,
corefs=args.corefs).read()
elif args.command == 'wikipedia':
WikipediaReader(args.url,
hg=hgraph(args.hg),
sequence=args.sequence,
lang=args.lang,
corefs=args.corefs).read()
elif args.command == 'reddit':
RedditReader(args.infile,
hg=hgraph(args.hg),
sequence=args.sequence,
lang=args.lang,
corefs=args.corefs).read()
elif args.command == 'actors':
Actors(hg=hgraph(args.hg), sequence=args.sequence).run()
elif args.command == 'claims':
Claims(hg=hgraph(args.hg), sequence=args.sequence).run()
elif args.command == 'conflicts':
Conflicts(hg=hgraph(args.hg), sequence=args.sequence).run()
elif args.command == 'names':
CorefsNames(hg=hgraph(args.hg), sequence=args.sequence).run()
elif args.command == 'number':
Number(hg=hgraph(args.hg), sequence=args.sequence).run()
elif args.command == 'onto':
CorefsOnto(hg=hgraph(args.hg), sequence=args.sequence).run()
elif args.command == 'taxonomy':
Taxonomy(hg=hgraph(args.hg), sequence=args.sequence).run()
else:
raise RuntimeError('Unknown command: {}'.format(args.command))
print()