Update consentManagementUsp.js to not store and reuse null consent

[patmmccann]

Fixes bug described at #5440 to comply with intent of #4425

Type of change

• Bugfix
• Feature
• New bidder adapter
• Code style update (formatting, local variables)
• Refactoring (no functional changes, no api changes)
• Build related changes
• CI related changes
• Does this change affect user-facing APIs or examples documented on http://prebid.org?
• Other

Description of change

No longer checks to see if consentData are stored; should recheck _uspapi

[patmmccann]

of note, build failed because of bad merge d8e5796

[jsnellbaker]

If this is something we wish to implement, then the changes here should work.

As a note - if we are just taking this aspect of the feature out, then we could just take out the entire unit test instead of modifying it. It's not really serving a unique purpose now and would be doing the same job as the other basic workflow unit test.

Assigning to @harpere to 2nd review on this change.

[jsnellbaker]

On an additional note - I retested the browserstack tests locally and they passed fine. I'm not sure why circleci tests are failing. It seems like it's trying to use a previous commit where a flatMap was added for the amxBidAdapter, but the config/stats for the job appear to be pointing to the most recent commit (where the flatMap was removed). I think we can just ignore it unless someone has another idea.

[patmmccann]

Thanks so much for all the education on yesterday's call. You're right, the circleci fail is from a reverted amx adapter commit.

…

On Tue, Jul 7, 2020 at 2:12 PM jsnellbaker ***@***.***> wrote: On an additional note - I retested the browserstack tests locally and they passed fine. I'm not sure why circleci tests are failing. It seems like it's trying to use a previous commit where a flatMap was added for the amxBidAdapter, but the config/stats for the job appear to be pointing to the most recent commit (where the flatMap was removed). I think we can just ignore it unless someone has another idea. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#5452 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAM25Z4TFDKGKACE4JM3YRDR2NQQ7ANCNFSM4OPJ4ZCQ> .

[bretg]

@harpere is out right now -- assigning over to @mkendall07

[mkendall07]

This change looks good if it's the behavior we want. The only slight caveat here is this part (from #4425)

the USP (CCPA) api function __uspapi() always responds synchronously, whether or not privacy data is available, while the GDPR CMP may respond asynchronously, especially if waiting for a user response. Unfortunately this doesn't necessarily mean it's always a synchronous transaction because if there are nested cross-domain iframes we may need to do an async postMessage().

Note the bolded part, it's possible this puts an async delay into every auction request. As long as we are ok living with that it's a LGTM.

[mkendall07]

@bretg @patmmccann for weigh in ^^

[patmmccann]

I don't think we have a choice, usp is opt out and if someone opts out at any point, refreshes and subsequent auctions need to know. If I were a regulator and wanted to test for compliance, I would go to a page, opt out, and look at the subsequent calls.

…

On Thu, Jul 16, 2020 at 9:51 AM Matt Kendall ***@***.***> wrote: @bretg <https://github.com/bretg> @patmmccann <https://github.com/patmmccann> for weigh in ^^ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#5452 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAM25Z7THP7SHUMMSNTBVTDR34AXFANCNFSM4OPJ4ZCQ> .

[mkendall07]

makes sense. It looks like circleci is continuously failing. Is this branch up to date with master?

[patmmccann]

there was a bad merge to master of the amx adapter that was later fixed

[patmmccann]

@mkendall07 it is up to date with master now

[bretg]

Re-ran the build and it passed. Since this is a bug, I'll assume that we're not going to need documentation. Will merge and add it to the rel notes.