Conversation
Benchmark Test ResultsBenchmark results from the latest changes vs base branch |
spiffcs
left a comment
There was a problem hiding this comment.
First pass.
I like the new SBOM struct and how it just encapsulates current information. Is there a future where the underlying Artifacts struct is changed as we build into developing relationships between everything?
Also small comments on data loss across the new outputs of the golden snapshots. I wasn't sure if we wanted to start excluding the information, or if it got lost as a consequence of moving to the new encode architecture.
internal/formats/spdx22json/test-fixtures/snapshot/TestSPDXJSONImagePresenter.golden
Show resolved
Hide resolved
internal/formats/syftjson/test-fixtures/snapshot/TestImagePresenter.golden
Show resolved
Hide resolved
bf6a2cd to
4757c25
Compare
Indeed! I have some of that drafted out in #607, specifically: Line 13 in c3a65f5 This is where the relationships object gets promoted to in the end (the |
Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
4757c25 to
95bd1fd
Compare
Signed-off-by: Alex Goodman <alex.goodman@anchore.com>
15a896f to
7c142e5
Compare
dabefda to
19b9013
Compare
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
19b9013 to
197c27b
Compare
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
luhring
left a comment
There was a problem hiding this comment.
Very nice work! 👏
I had just one question from curiosity
|
@spiffcs thanks a ton for solving what ended up being a caching issue I was having --I was starting to go a little crazy there 🙌 |
* [wip] single sbom doc Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * fix tests Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * fix more tests Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * fix linting Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * update cli tests Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * remove scope in import path Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * swap SPDX tag-value formatter to single sbom document Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * bust CLI cache Signed-off-by: Alex Goodman <alex.goodman@anchore.com> * update fixture to byte diff Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> * byte for byte Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> * bust the cache Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> * who needs cache Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> * add jar for testing Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> * no more bit flips Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> * update apk with the delta for image and directory cases Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> * restore cache workflow Signed-off-by: Christopher Angelo Phillips <christopher.phillips@anchore.com> Co-authored-by: Christopher Angelo Phillips <christopher.phillips@anchore.com>
This PR replaces
poweruser.JSONDocumentConfigwithsbom.SBOM. Specifically this datastructure is being promoted as a first-class object to be used internally for tasking, presentation, and processing concerns. This is proposed in #555 (comment) but closely affects #554 and #556.The main proposal of the PR is the following datastructure:
syft/syft/sbom/sbom.go
Lines 10 to 23 in a26e9e4
All other changes attempt to promote usage of this data structure over passing parts of an SBOM around, for example... replacing:
with:
Notes:
sbom.SBOMin other PRs implementing Prefer artifact relationships over package relationships #556source.Scopewas removed from any presenter output as described in Encapsulate all artifacts in syft JSON output #555 (comment)