W3C

– DRAFT –
WoT Testfest - Day 5

29 July 2022

Attendees

Present
Cristiano_Aguzzi, David_Ezell, Ege_Korkan, Fady_Salama, Farshid_Tavakolizadeh, Kaz_Ashimura, Michael_McCool, Tomoaki_Mizushima
Regrets
-
Chair
McCool
Scribe
cris, kaz

Meeting minutes

PRs

<kaz> PR 391 - Update intel-nodejs-speak TD

<kaz> PR 390 - Update results/intel-nodejs

<kaz> PR 389 - Update intel-nodejs-speak TD

<kaz> PR 388 - 2022 07 result updates

<kaz> PR 387 - 2022 07 Updates to Input Data - Day 4

McCool: I merged five PRs yesterday
… let's start from there

McCool: updated the rendering script
… some work on intel PRs
… I created a custom token scheme
… also auto
… later on some other updates on schemes
… I worked on combo
… thanks to it I was able to clean up the list of forms
… and it cleaned also some assertions

PR 393

<kaz> PR 393 - add VAIMEE's Thing Description Directory

Cristiano: implemented a TDD for an european project, it is scalable and it implements JSONPath and events

McCool: there is a minor problem, you should concatenate active and manual csv
… I'll do it right now
… merging

PR 394

<kaz> PR 394 - improve node-wot TM results

Cristiano: I checked a bunch of assertions for node-wot
… there are two controversial assertions

Ege: actually we are checking for it

McCool: there is a TODO left
… might cause problems with tooling
… but I like the idea of having TODO
… might integrate in the tool
… other than this PR looks fine
… merging

PR 392

<kaz> PR 392 - more 2022 07 result updates

McCool: there are some controversial topics
… I added a new Huawei TD
… I use combo
… this time, the controversial point is that is an update of an old implementation from Huawei
… is it ok to put the update under huawaii folder?
… if we cannot I should implement a proxy to assert the same stuff

Ege: it is fine by me

Cristiano: +1

<kaz> merged

PR 374

<kaz> PR 374 - TUM manual.csv

McCool: is this still an example PR?

fady: I'm working on it

Updates to report

<McCool> More 2022 07 testfest updates for Implementation Report #1634

McCool: I did some updates to TD implementation report
… there are 49 at risk item left
… it is still not perfect but we are moving forward
… running the batch script again
… there are some errors
… are those coming from the playground?

Ege: it does not sound right

McCool: it should be a playground problem, I didn't change anything
… anyhow back to implementation report: I fixed a bunch of assertions
… OAuth is left not easy to implement

Ege: ditto has an implementation
… also node-wot

McCool: TM does not count as implementation

Cristiano: node-wot should implement it

Ege: code flow is there for sure

McCool: there are some missing implementations for op keywords

Ege: I did some experimental work on some of them

McCool: pretty good shape so far
… I'll clean up some security
… probably the only one left will be proxy
… about manual assertions
… UUIDs should be easy

Ege: they should be automatically tested, right?

McCool: I guess we can clear that off with automatic tests
… there is an assertion that looks like a policy
… I think we should remove it, I'll track it in a issue

Ege: regarding updating it is a bit a little too much for the spec

McCool: I'm marking this as something will do later on
… regarding the offending assertion they are policy assertions

Kaz: we should clarify which part of the expected system behavior to be defined by our specs and which parts are implementation-dependent. We could clarify that using an Editor's Note for the moment, and should clearly define that by another document about implementation guidelines
… where to put those things
… it is good that it is in the next charter

McCool: we have the security guidelines document
… we can use it for this matter
… body extension is weird

Ege: there are some assertions that can be moved to auto csv

McCool: easy, as long as you are not using html strings in TD you should check pass for this assertion

McCool: is any implementation that counts characters when loading TDs?

Ege: playground yes

Cristiano: node-wot sort of, Nodejs should have limtis for buffers and strings

<McCool> https://github.com/w3c/wot-thing-description/issues/1635

Kaz: as the next step, would it be possibile to categorize the unimplemented features in who's going to tackle it?

McCool: I would like to down grade to un informal statment

Kaz: my suggestion is starting with categorizing the remaining issues and think about how to cover them next :)

McCool: let's do that next week

[adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 147 (Thu Jun 24 22:21:39 2021 UTC).