Meeting minutes
TD PR 1525
<Ege> wot-thing-description PR 1525 - Testing Cleanup and Assertion Extractor
Ege: I added default assertion extraction in my PR in the wot-thing-description repo
McCool: My tool doesn't do that, I need to check
Ege: I also updated the depends.csv
McCool: We need to merge this and make sure we generate the manual.csv
Ege: The weird thing is that we it should have produced false gaps
McCool: Then it would be good to do a sanity check
Kaz: We need to clarify the procedure of the testing, and document it somewhere. probably on the wot-testing repository pointing each spec repo.
McCool: I see that we should merge. Sebastian, do you agree?
Sebastian: Yes we should merge
McCool: merged
Ege: I think we need to document all the things that need to be pushed for testing
McCool: Sure. Also I wanted to check the manual.csv, Ege and Fady, did you update it?
Fady: I pushed the new manual.csv
McCool: PRs are better so we can see the diffs
wot-testing PRs
PR 324 - Discovery: Add results for siemens-logilab
McCool: We will not merge this yet
<kaz> PR 324 - Discovery: Add results for siemens-logilab
PR 326
<kaz> PR 326 - fixes ediTDor TMs
Sebastian: This PR fixes the TMs and makes sure the assertion tester can test them correctly
McCool: merged
Current gaps
<kaz> current implementation report for wot-thing-description
McCool: bearer-format is not implemented by anyone, but can be marked as risk
… security-flow as well
… ComboSecurityScheme is also not implmented anywhere
… ContentEncoding is currently also at risk
… we can easily have examples exclusiveMinimum etc.
… unobserveProperty is also missing
… synchronous ActionAffordance is not implemented yet
… Then we have manual assertions, mostly about security
<Ege> gaps are documented here: https://
Issues
McCool: We are missing implementation descriptions
Issue 325 - Missing Implementation Descriptions
Issue 321 - Missing manual.csv contributions
McCool: New manual.csv are generally needed for all implementations
Kaz: We should clarify the expectations for the implementers
… using a GitHub issue to get data is a possible option, but we should clarify the whole procedure to get which data how and when, and then ask all the implementers to provide which data. Maybe creating a GitHub issue before that mentioning we are missing this data and that would not be very polite.
McCool: Can you update the instructions, Ege?
Ege: Doing it right now
McCool: We will discuss these in the main call afterwards
<McCool_> https://
McCool: meeting adjourned