test(reporters): Add three scan issues to report test assets #8552
Conversation
fviernau
force-pushed
the
webapp-filter-excluded-preparations
branch
from
e01a939 to
50aeb43
Compare
| @@ -65,59 +65,80 @@ issue_resolutions: | |||
| comment: "A comment explaining why the issue can be ignored." | |||
| issues: | |||
| - _id: 0 | |||
| timestamp: "2024-04-22T10:36:10.661544294Z" | |||
There was a problem hiding this comment.
Add one issue to an included- and another to an excluded project
"Add one issue to an included and another one to an excluded project"
This prepares adding the ability
"This prepares for adding the ability" or "This prepares to add the ability"
fviernau
force-pushed
the
webapp-filter-excluded-preparations
branch
from
50aeb43 to
e89a519
Compare
fviernau
force-pushed
the
webapp-filter-excluded-preparations
branch
from
e89a519 to
6b893b2
Compare
The replacement has been introduced for replacing the report creation timestamp only. However, `replace()` replaces all matches and the pattern would all so match the timestamp value of issues. So, replace only the first match, so that issues (with timestamps) can be added in a following change. [1]: ba0e4cb Signed-off-by: Frank Viernau <[email protected]>
fviernau
force-pushed
the
webapp-filter-excluded-preparations
branch
from
6b893b2 to
03dd6ae
Compare
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
plugins/reporters/evaluated-model/src/funTest/kotlin/EvaluatedModelReporterFunTest.kt
Fixed
Show fixed
Hide fixed
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #8552 +/- ##
=========================================
Coverage 67.68% 67.68%
Complexity 1006 1006
=========================================
Files 246 246
Lines 7924 7924
Branches 883 883
=========================================
Hits 5363 5363
Misses 2181 2181
Partials 380 380
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
Add one issue with an included and one issue with an excluded affected
path to an included project. Furthermore, add on issue with an included
affected path to an excluded project. This prepares to add the ability
to vizualize issues with excluded effected path as excluded, in the
WebApp as well as in the static HTML report.
Note: There are three copies of `reporter-test-input.yml` which are all
updated for consistency. Furthermore, the affected paths of the
issues have been choosen to be within the VCS path, so that the
issues do net get filtered out by
`OrtResult.getScanResultsForId()`.
Signed-off-by: Frank Viernau <[email protected]>
fviernau
force-pushed
the
webapp-filter-excluded-preparations
branch
from
03dd6ae to
5ff7c71
Compare
| @@ -190,6 +221,97 @@ scan_results: | |||
| package_verification_code: "da39a3ee5e6b4b0d3255bfef95601890afd80709" | |||
| packages: | |||
| - _id: 0 | |||
| id: "Gradle:org.ossreviewtoolkit:nested-fake-project:1.0.0" | |||
There was a problem hiding this comment.
Any idea why the numeric IDs of the packages seem to be swapped now? That makes the diff a bit difficult to read...
There was a problem hiding this comment.
No idea. This is always imho difficult to figure out as the IDs are assigned by Jackson.
There was a problem hiding this comment.
@sschuberth - I've debugged this for quite a bit, and it seems to just depend on the order things get serialized. E.g. the first identifier which gets serialized gets the 0 assigned. The next identifier the 1 and so forth. So, with this change Gradle:org.ossreviewtoolkit:nested-fake-project:1.0.0 gets serialized first.
I believe such diffs we have to live with, as long as we use the above mentioned approach to assign identifiers.
There was a problem hiding this comment.
E.g. the first identifier which gets serialized gets the
0assigned.
Right, but why does the order of identifiers change with your PR, shouldn't it all be deterministic? In any case, thanks for looking into it.
There was a problem hiding this comment.
Right, but why does the order of identifiers change with your PR, shouldn't it all be deterministic?
Hm, each time one make a small change to the test asset one needs to debug why jackson assigns the IDs in the way it does is imo raising the bar for contributions quite a bit. Can we shortcut this and live without the explanation ?
There was a problem hiding this comment.
Investigating further. Maybe @mnonnenmacher does the issue here ring any bell? (can you help?)
There was a problem hiding this comment.
I believe it is as follows:
- The
EvaluatedModel.issuesproperty comes beforeEvaluatedModel.packages. - The issues do reference packages
- So, the assignment of the package identifiers start when
issuesare serialized, and the ordering of
the issues determines the IDs of the referenced packages. - As this commit adds a new first issue, the package referenced gets the first id:
0.
So, it seemed natural to just change property order so that packages are serialized before the issues, but
that would re-create a similar issues, because there is a cyclic reference between these instances.
@sschuberth what do you think?
There was a problem hiding this comment.
That sounds reasonable as an explanation, good enough for me, thanks!
fviernau
changed the title
First preparations for implementing #8373.