Towards improving read and rendering of results

[ooprathamm]

closes #991

Checklist

• No CHANGELOG update needed

• No new tests needed

• No documentation update needed

[github-actions]

Please add bug fixes, new features, breaking changes and anything else you think is worthwhile mentioning to the master (unreleased) section of CHANGELOG.md. If no CHANGELOG update is needed add the following to the PR description: [x] No CHANGELOG update needed

[williballenthin]

im sorry we didn't provide more detail on how we were thinking about going about this. i think we should rely on --json > output.json to save the result document. for new functionality, we'll have to find a way to identify that the input file is a result document, and then do the rendering. determining the file type might be done like: 1) check for json content, a few strings we expect to be present, and then trying to deserialize it, or 2) requiring the user to specify format=result or similar as a cli argument.

please dont hesitate to reach out with questions or suggestions - we're excited to work with you on this enhancement!

[mr-tz]

determining the file type might be done like: 1) check for json content, a few strings we expect to be present, and then trying to deserialize it, or 2) requiring the user to specify format=result or similar as a cli argument.

I'm also in favor of this so we don't need a new command line argument and instead reuse sample. This can also simplify some of the things you're concerned about.

[williballenthin]

i've updated #991 with more explicit details on how we were thinking this might work. please use this as guidance or coordinate with us if you want to propose alternative logic. we're open to other ideas but also sensitive to the current architecture of capa and what's on the roadmap.

[ooprathamm]

@williballenthin @mr-tz Made the final changes :)
Able to successfully parse the meta and capabilities from the output json

Output

CHANGELOG updated or no update needed, thanks! 😄

[williballenthin]

see feedback inline and let me know what questions you have

[williballenthin]

also, please prepare a few unit tests that demonstrate how the new functionality proposed here works as intended.

[ooprathamm]

@williballenthin @mr-tz I hope I have resolved all the above discussions.

Passing Testsuite -

[williballenthin]

we are really close! i have a few requests inline, but i dont anticipate any of these should be difficult to address. if they are, dont hesitate to reach out and i'd be hapy to lend a hand. overall, i feel that the quality of this PR is really good, nice job!

[williballenthin]

there's also some conflicts to deal with from master, and we'll want the linters to pass again (code style is broken right now but should be fixed soon. if this becomes a blocker we can find a way around it).

[ooprathamm]

@williballenthin I have addressed the suggestions in inline

[williballenthin]

awesome, logic looks good!

please:

• resolve the conflicts with master, and
• ensure isort and black linters are happy with your changes (mypy is broken right now)

[williballenthin]

“xfail” means “expected to fail” so in fact we want these entries. I don’t see any “fail” in that output so it looks great!

[williballenthin]

this failure is unexpected, but im not sure if its from master. let me triage it further. otherwise, PR looks great and i'll merge shortly!

[williballenthin]

great, thanks for the arch fix, @ooprathamm!

once the remaining tests pass (aside from codestyle) i'll merge, thank again!

[ooprathamm]

@williballenthin I have fixed the reason of failing elf arch extraction.
I hope these were the final changes to this PR. It was awesome working on this one, got me really brainstorming.
Thanks for providing your guidance throughout.

Passing local test suite

[williballenthin]

🎉