Configure tox to use legacy resolver as a temporary workaround
for pypa/pip#9215, which results in huge unnecessary downloads.
Co-authored-by: Jussi Kukkonen <jkukkonen@vmware.com>
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
The GitHub action windows runners (added in a subsequent commit)
choke on a test that runs os.makedirs with a too long directory
name, and expects an OSError with error numbers ENAMETOOLONG or
ENOENT. However, this particular runner returns EINVAL in Python 3,
which according to bugs.python.org/msg295851 is not unlikely.
This commit simply adds EINVAL to the expected error numbers.
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
We recently enabled support for Python 3.9 and also added
3.9 builds on appveyor, but they don't work out of the box.
Instead of troubleshooting this I suggest we soon switch to
GitHub actions:
https://github.com/theupdateframework/tuf/issues/1195.
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
requirements-pinned.txt lists all immediate and transitive
dependencies combined for all supported Python versions.
This commit semi-automatically updates the pinned dependencies
using the instructions in requirements.txt:
Updated dependencies are:
- cryptography (supersedes dependabot's #1237)
- certifi (supersedes dependabot's #1233)
- enum34 (oddly not detected by dependabot)
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
Python 3.5 has now reached its end-of-life and has been retired.
https://www.python.org/dev/peps/pep-0478/
The optional (but highly recommended) 'cryptography' dependency
has also just dropped support for 3.5. Continuing support for 3.5
in TUF does not seem worth the effort.
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
Add a test to ensure that metadata expires at the expiration time, not
after it.
This tests the change to the updater introduced in 4bcd703
Signed-off-by: Joshua Lock <jlock@vmware.com>
The specification, as of 1.0.16, describes an update expiration check as:
> The expiration timestamp in the trusted $ROLE metadata file MUST be
higher than the fixed update expiration time.
Having done some research into how other security providers are comparing
expiration equivalents (i.e. OpenSSL x509 certificate checking code, and
GnuPG expiration checks), and how other TUF implementations are performing
the same check (rust-tuf, go-tuf), we came to a consensus that the correct
way to implement expiration comparisons is:
expiration <= now
Where:
expiration: is the metadata's expiration datetime
now: is the current system time, or the fixed notion of time in the
detailed client workflow (introduced in 1.0.16 of the spec)
Fixes#1231
Signed-off-by: Joshua Lock <jlock@vmware.com>
Bandit just dropped support for Python <3.5. This commit adds
a corresponding constraint to requirements-test.txt.
Note, we run bandit in a dedicated 'lint' tox environment, which
uses Python3.8 on Travis.
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
Use Google style guide with refinements, because the Google style
guide is a comprehensive, well-established style guide that is
mostly based on PEP-8 and was accepted by everyone on the TUF team.
There is no need to replicate these recommendations. However, we do
provide a very slim document with additional refinements, in order
to emphasize on items the we consider especially important, want to
be handled differently, or in one specific way, where the Google
guide would allow multiple.
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
Co-authored-by: Joshua Lock <jlock@vmware.com>
Add MADR that justifies why we want to add custom classes for
complex tuf metadata attributes.
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
Installing securesystemslib in editable mode leads to a problem
in sys.path where we have two "tests" packages.
By not installing securesystemslib in an editable mode we are not
adding the securesystemslib tests to sys.path.
Signed-off-by: Martin Vrachev <mvrachev@vmware.com>
Describe pros of developing TUF 1.0.0 in a subdirectory
of the current implementation against the rest of the options.
Signed-off-by: Teodora Sechkova <tsechkova@vmware.com>
Document the outcome of #1126 to develop TUF 1.0.0
in a subdirectory of the current TUF implementation.
Signed-off-by: Teodora Sechkova <tsechkova@vmware.com>
If during updater.download_target() the download succeeds but a later
check fails (e.g. BadHashError), remember to close the tempfile.
Signed-off-by: Jussi Kukkonen <jkukkonen@vmware.com>
Simplify the loop exit logic in _get_target_file() to simply return a
verified file_object, once we have it, rather than breaking from the loop
and then returning the file_object.
This converts a use of a try/except/else to a try/except and is a little
easier to read.
Signed-off-by: Joshua Lock <jlock@vmware.com>
When verifying newly downloaded root metadata with the keys listed in the
root metadata being verified, multiple signatures with the same keyid
should not be counted towards the threshold. A keyid should only count
once towards the threshold.
This fixes the _verify_root_self_signed() method introduced in PR #1101 to
ensure that keyids are only counted once when verifying a threshold of new
root signatures.
Signed-off-by: Joshua Lock <jlock@vmware.com>
When the updater is verifying that the new root metadata is signed by a
threshold of keys defined by the new root metadata itself, multiple
signatures with the same keyid should not be counted more than once
towards the threshold.
Implement a test for this, which currently fails.
Reported-by: Trishank Karthik Kuppusamy <trishank.kuppusamy@datadoghq.com>
Signed-off-by: Joshua Lock <jlock@vmware.com>
Currently, we are importing the "utils" module in tests/utils
with "import utils".
This could become a problem when there is another module with
the same general name "utils" and could lead to import mistakes.
Signed-off-by: Martin Vrachev <mvrachev@vmware.com>
This quote from the Google Python style guide made me realize
why empty list as a default value for an argument could be
dangerous:
"Default arguments are evaluated once at module load time.
This may cause problems if the argument is a mutable object
such as a list or a dictionary. If the function modifies the object
(e.g., by appending an item to a list), the default value is modified."
Read more here:
https://google.github.io/styleguide/pyguide.html#2123-cons
Signed-off-by: Martin Vrachev <mvrachev@vmware.com>
We don't want to read the whole file in memory as it can be huge. Use
digest_fileobject() instead: This way Securesystemslib will read the
file in chunks.
Securesystemslib already takes care of seeking to beginning of file.
Fixes#1215
Signed-off-by: Jussi Kukkonen <jkukkonen@vmware.com>
In 'repository_lib._generate_and_write_metadata' sort the set of
signing key keyids alphabetically before passing them on to signing
functions, to make the order in which signatures are added
deterministic.
This is above all beneficial for testing.
This commit also adds an exemplary test for signatures on root
metadata using the repository_tool interface to setup all the state
that required to test _generate_and_write_metadata.
Signed-off-by: Lukas Puehringer <lukas.puehringer@nyu.edu>
Update the comments to not mention the usage of temp file
for logging regarding the instances of the TestServerProcess class.
Also, remove one unused import.
Signed-off-by: Martin Vrachev <mvrachev@vmware.com>
