def take_action(self, parsed_args):
self.log.debug('[+] showing overview of CTU datasets')
# TODO(dittrich): Getting really not DRY: Move this into class.
pages = []
# Expand scenario names if abbreviated
scenarios = [CTU_Dataset.get_fullname(name=s)
for s in parsed_args.scenario]
if 'ctu_metadata' not in dir(self):
self.ctu_metadata = CTU_Dataset(
cache_file=parsed_args.cache_file,
ignore_cache=parsed_args.ignore_cache,
debug=self.app_args.debug)
self.ctu_metadata.load_ctu_metadata()
if len(scenarios) == 0:
print("{}".format(CTU_Dataset.get_disclaimer()))
pages.append(CTU_Dataset.get_ctu_datasets_overview_url())
else:
for scenario in scenarios:
page = self.ctu_metadata.get_scenario_data(scenario,
'Capture_URL')
if page is not None:
pages.append(page)
for page in pages:
open_browser(page=page,
browser=parsed_args.browser,
force=parsed_args.force)
def test_get_disclaimer(self):
disclaimer = CTU_Dataset.get_disclaimer()
self.assertTrue("http://dx.doi.org/10.1016/j.cose.2014.05.011" in disclaimer)
def get_parser(self, prog_name):
parser = super().get_parser(prog_name)
parser.formatter_class = argparse.RawDescriptionHelpFormatter
parser.add_argument(
'--force',
action='store_true',
dest='force',
default=False,
help="Force over-writing files if they exist (default: ``False``)")
parser.add_argument('--no-subdir',
action='store_true',
dest='no_subdir',
default=False,
help=('Do not maintain scenario name subdirectory '
'(default: ``False``)'))
_default_protocols = ",".join(DEFAULT_PROTOCOLS)
parser.add_argument('-P',
'--protocols',
metavar='<protocol-list>',
dest='protocols',
type=lambda s: [i for i in s.split(',')],
default=_default_protocols,
help=("Protocols to include, or 'any' "
f'(default: ``{_default_protocols}``)'))
parser.add_argument(
'-L',
'--maxlines',
metavar='<lines>',
dest='maxlines',
default=None,
help="Maximum number of lines to get (default: ``None``)")
cache_file = CTU_Dataset.get_cache_file()
parser.add_argument('--cache-file',
action='store',
dest='cache_file',
default=cache_file,
help=('Cache file path for CTU metadata '
'(Env: ``LIM_CTU_CACHE``; '
f'default: ``{cache_file}``)'))
parser.add_argument(
'--ignore-cache',
action='store_true',
dest='ignore_cache',
default=False,
help="Ignore any cached results (default: ``False``)")
parser.add_argument('scenario',
nargs=1,
type=normalize_ctu_name,
default=None)
data_types = str(", ".join(
[f'{i.lower()}' for i in CTU_Dataset.get_data_columns()]))
parser.add_argument(
'data',
nargs='+',
type=str.lower,
choices=[
c.lower() for c in CTU_Dataset.get_data_columns() + ['all']
],
default=None)
parser.epilog = textwrap.dedent(f"""\
Get one or more data components from a scenario. These
components are the raw PCAP file, Netflow file, and
other analytic products from intrusion detection system
processing, etc.
See ``lim ctu list --help`` for more on the ``scenario`` argument.
For the ``data`` argument, you can use ``all`` to recursively
download all scenario data, or one or more of the data
files by type: ``{data_types}``
By default, or when using the ``all`` attribute identifier,
the file(s) are placed in a subdirectory with the full name
of the scenario to better organize data across multiple
scenarios. You can override this when getting specific files
(i.e., not using ``all``) with the ``--no-subdir`` option.
\n""") + CTU_Dataset.get_disclaimer() # noqa
return parser
def get_parser(self, prog_name):
parser = super().get_parser(prog_name)
parser.formatter_class = argparse.RawDescriptionHelpFormatter
cache_file = CTU_Dataset.get_cache_file()
parser.add_argument('--cache-file',
action='store',
dest='cache_file',
default=cache_file,
help=('Cache file path for CTU metadata '
'(Env: ``LIM_CTU_CACHE``; '
f'default: ``{cache_file}``)'))
parser.add_argument(
'--ignore-cache',
action='store_true',
dest='ignore_cache',
default=False,
help="Ignore any cached results (default: ``False``)")
parser.add_argument('--date-starting',
dest='date_starting',
metavar='<YYYY-MM-DD>',
default='1970-01-01',
help=('List scenarios starting from this date '
"(default: '1970-01-01')"))
TODAY_DATE = arrow.now().format('YYYY-MM-DD')
parser.add_argument('--date-ending',
dest='date_ending',
metavar='<YYYY-MM-DD>',
default=TODAY_DATE,
help=('List scenarios up to this date '
f"(default: '{TODAY_DATE}')"))
parser.add_argument('--fullnames',
action='store_true',
dest='fullnames',
default=False,
help=("Show full names"))
parser.add_argument('-a',
'--everything',
action='store_true',
dest='everything',
default=False,
help=("Show all metadata columns "
"(default : False)"))
parser.add_argument('--hash',
dest='hash',
metavar='<{md5_hash|sha256_hash}>',
default=None,
help=('Only list scenarios that involve a '
'specific hash (default: ``None``)'))
parser.add_argument(
'--malware-includes',
dest='malware_includes',
metavar='<string>',
default=None,
help=('Only list scenarios including this string'
"in the 'Malware' column (default: ``None``)"))
parser.add_argument(
'--name-includes',
dest='name_includes',
metavar='<string>',
default=None,
help=('Only list scenario including this string'
"in the 'Capture_Name' column (default: ``None``)"))
parser.add_argument('--description-includes',
dest='description_includes',
metavar='<string>',
default=None,
help=('Only list scenarios including this string'
'in the description (default: ``None``)'))
parser.add_argument('scenario',
nargs='*',
type=normalize_ctu_name,
default=None)
all_columns = ", ".join(
[f'{i.lower()}' for i in CTU_Dataset.get_all_columns()])
default_columns = ", ".join(
[f"{i.lower()}" for i in CTU_Dataset.get_index_columns()])
parser.epilog = textwrap.dedent(f"""\
List scenarios (a.k.a., "captures") and related metadata.
By default, all scenarios are listed. You can limit the output
by filtering on several attributes (e.g., by ``Capture_Name``
field, by date range, contents of the malware name or web page
description, etc.) You can also limit the number of items
shown if necessary when the number of results is large.
The ``scenario`` argument equates to the field ``Capture_Name`` in
the index. This can be the scenario's full name (e.g.,
``CTU-IoT-Malware-Capture-34-1``) or an abbreviated form of the
name (e.g., ``IoT-34-1`` or just ``34-1``).
::
$ lim ctu list IoT-34-1 Botnet-42
+----------------+-------------------------------+---------+
| Infection_Date | Capture_Name | Malware |
+----------------+-------------------------------+---------+
| 2011-08-10 | CTU-Malware-Capture-Botnet-42 | Neeris |
| 2018-12-21 | CTU-IoT-Malware-Capture-34-1 | Mirai |
+----------------+-------------------------------+---------+
A larger number of attributes are available. You can get all of them
using the ``-a`` (``--everything``) flag. The subset of columns shown
by default is: ``{default_columns}``
Valid column labels for options ``-c``, ``--column``, ``--sort-column``,
or to be shown with ``-a``, include:
``{all_columns}``
Using ``lim ctu list -a`` produces very wide output. Even if many fields
are ``None`` and ``--fit-width`` is included, it is still unwieldy for just
one scenario as you can see here. Consider using ``lim ctu show`` instead.
::
$ lim ctu list --name-includes IoT --malware-includes muhstik --fit-width -a
+----------------+------------------------+---------+------------------------+------------------------+------------------------+------------------------+---------+------------------------+------------------------+-----------------------------+
| Infection_Date | Capture_Name | Malware | MD5 | SHA256 | Capture_URL | ZIP | LABELED | BINETFLOW | PCAP | WEBLOGNG |
+----------------+------------------------+---------+------------------------+------------------------+------------------------+------------------------+---------+------------------------+------------------------+-----------------------------+
| 2018-05-19 | CTU-IoT-Malware- | Muhstik | b8849fe97e39ae3afd6def | 5ce13670bc875e913e6f08 | https://mcfp.felk.cvut | fce7b8bbd1c1fba1d75b9d | None | 2018-05-21_capture.bin | 2018-05-21_capture.pca | 2018-05-21_capture.weblogng |
| | Capture-3-1 | | 618568bb09 | 7a4ac0a9e343347d5babb3 | .cz/publicDatasets/IoT | c1a60b25f49f68c9ec16b3 | | etflow | p | |
| | | | | b5c63e1d1b199371f69a | Datasets/CTU-IoT- | 656b52ed28290fc93c72.z | | | | |
| | | | | | Malware-Capture-3-1 | ip | | | | |
+----------------+------------------------+---------+------------------------+------------------------+------------------------+------------------------+---------+------------------------+------------------------+-----------------------------+
There are also a number of filters that can be applied, including MD5
and SHA256 hash, substrings in the ``Capture_Name`` or ``Malware``
fields, start and end dates, or description of the scenario in its
web page.
The ``--hash`` option makes an exact match on any of the stored hash
values. This is the hash of the executable binary referenced in the
``ZIP`` column. This example uses the most frequently occuring MD5
hash as seen in ``lim ctu stats --help``::
$ lim ctu list --hash e515267ba19417974a63b51e4f7dd9e9
+----------------+----------------------------------+---------+
| Infection_Date | Capture_Name | Malware |
+----------------+----------------------------------+---------+
| 2015-03-04 | CTU-Malware-Capture-Botnet-110-1 | HTBot |
| 2015-03-04 | CTU-Malware-Capture-Botnet-110-2 | HTBot |
| 2015-03-09 | CTU-Malware-Capture-Botnet-110-3 | HTBot |
| 2015-03-09 | CTU-Malware-Capture-Botnet-111-2 | HTBot |
| 2015-04-09 | CTU-Malware-Capture-Botnet-110-4 | HTBot |
| 2015-04-09 | CTU-Malware-Capture-Botnet-111-3 | HTBot |
| 2015-04-22 | CTU-Malware-Capture-Botnet-110-5 | HTBot |
| 2015-04-22 | CTU-Malware-Capture-Botnet-111-4 | HTBot |
| 2015-04-23 | CTU-Malware-Capture-Botnet-110-6 | HTBot |
| 2015-06-09 | CTU-Malware-Capture-Botnet-111-5 | HTBot |
+----------------+----------------------------------+---------+
The ``--malware-includes`` option is rather simplistic, matching any
occurance of the substring (case insensitive) in the ``Malware`` field.
The same applies for the ``--name-includes`` option with respect to the
``Capture_Name`` field. For more accurate matching, you may want to use
something like the ``-f csv`` option and match on regular expressions
using one of the ``grep`` variants. Or add regular expression handling
and submit a pull request! ;)
\n""") + CTU_Dataset.get_disclaimer() # noqa
return parser