def init_argparse(cls, ap) -> None:
ap.add_argument(
"content_type",
choices=[t.get_name() for t in meta.get_all_content_types()],
help="what kind of content to hash",
)
ap.add_argument(
"--signal-type",
"-S",
choices=[t.get_name() for t in meta.get_all_signal_types()],
help="only generate these signal types",
)
ap.add_argument(
"--as-text",
"-T",
action="store_true",
help=
"force input to be interpreted as text instead of as filenames",
)
ap.add_argument(
"content",
nargs="+",
help=("what to match against. Accepts filenames, "
"quoted strings, or '-' to read newline-separated stdin"),
)
def _apply_transformers(self, features: Features) -> Features:
transformed = features
for t in self.transformers:
transformed = t.transform(transformed)
transformed.validate()
logging.info("Applied %s", t.get_name())
return transformed
def init_argparse(cls, ap) -> None:
ap.add_argument(
"content_type",
choices=[t.get_name() for t in meta.get_all_content_types()],
help="what kind of content to match",
)
ap.add_argument(
"--hashes",
"-H",
action="store_true",
help=(
"instead of content (i.e. videos), "
"input contains intermediate representations (i.e. video MD5s)"
),
)
ap.add_argument(
"--as-text",
"-T",
action="store_true",
help="force input to be interpreted as text instead of as filenames",
)
ap.add_argument(
"content",
nargs="+",
help=(
"what to match against. Accepts filenames, "
"quoted strings, or '-' to read newline-separated stdin"
),
)
ap.add_argument(
"--show-false-positives",
action="store_true",
help="show matches even if you've marked them false_positive",
)
ap.add_argument(
"--hide-disputed",
action="store_true",
help="hide matches if someone has disputed them",
)
def _fit_transform(self, features: Features):
transformed = features
for t in self.transformers:
transformed = t.fit_transform(transformed)
logging.info("Trained %s", t.get_name())
return transformed