Open Bug 1769327 Opened 4 years ago Updated 13 days ago

make clearHistory & clearSiteData match clearOnShutdown and not remove `persist-data-on-shutdown` exceptions

Categories

(Toolkit :: Data Sanitization, enhancement, P5)

Firefox 91
enhancement

Tracking

()

People

(Reporter: thorin, Unassigned)

Details

User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Firefox/91.0

Steps to reproduce:

Same as Bug 1681701

privacy.clearOnShutdown* and privacy.cpd should follow the same behavior. For site exceptions, users can still manage and clear site data from the Cookies and Site Data section of the UI

Component: Untriaged → Data Sanitization
Product: Firefox → Toolkit
Flags: needinfo?(mail)

any feedback on this yet?

Flags: needinfo?(mail)

I think this is a nice feature, given that users know that sites they have given "cookie" permissions are not cleared. So if we implement this we should also add a checkbox to toggle these exceptions on/off or at least a warning message: "X sites / these sites are exempt and will not be cleared".

Severity: -- → N/A
Priority: -- → P5
Status: UNCONFIRMED → NEW
Ever confirmed: true
Summary: Ensure that cpd sanitization for cookies and site data adheres to cookie exceptions → make clearHistory & clearSiteData match clearOnShitdown and not remove `persist-data-on-shutdown` exceptions
Summary: make clearHistory & clearSiteData match clearOnShitdown and not remove `persist-data-on-shutdown` exceptions → make clearHistory & clearSiteData match clearOnShutdown and not remove `persist-data-on-shutdown` exceptions

It's been a while and looking at the bug I'm not sure what exactly we want to do here. Has this been addressed by the recent switch to a dedicated clear on shutdown exception permission?

Flags: needinfo?(thorin)

Does "Clear recent history" with privacy.cpd.cookies = true honor exceptions (either classic cookie permissions or newer persist-data-on-shutdown permissions)? This isn't indicated in the dialog one way or the other. In a perfect world, there might be an indented checkbox for following or overriding the user's exceptions.

(In reply to jscher2000 from comment #5)

Does "Clear recent history" with privacy.cpd.cookies = true honor exceptions

No, next bit. Note: cpd is obsolete, it's been migrated to clearHistory - e.g. privacy.clearHistory.cookiesAndStorage

(In reply to Emma Zühlcke [:emz] from comment #4)

It's been a while and looking at the bug I'm not sure what exactly we want to do here. Has this been addressed by the recent switch to a dedicated clear on shutdown exception permission?

Sure has been a while :) and no AFAICT

For those following along at home, there are three global methods to sanitize (i.e ignoring site specific Forget about this site)

  • clearOnShutdown_v2 -> automatic with sanitizeOnShutdown
    • items map to options in History>Choose...>Settings
  • clearSiteData -> manual -> items map to options in Privacy and security>Browsing data>Clear browsing data
  • clearHistory -> manual -> items map to options in ctrl-shift-del

How these three behave should be universal IMO

This issue was that the two manual methods, when clearing cookiesAndStorage would not respect the site exception

  • this needs re-testing this still seems to be the case
  • so divergent behavior: auto-sanitize I keep expected site cookies/data .. manually I unexpectedly lose them

Another issue is that the two manual methods, clearing siteSettings also removes the site exception

It would be nice to align the behavior here

Flags: needinfo?(thorin)

without wanting to compound anything, so feel free to ignore ... I wonder why we even need clearHistory anymore - just make the ctrl-shift-del dialog's options map to clearSiteData - i.e you can go to Privacy and security>Browsing data>Clear browsing data if you're in settings .. or shortcut to it with ctrl-shift-del from any origin. They both serve the exact same functionality - they're manual, one-offs, and you change what is cleared each time (e.g when testing/dev work). Unless I'm missing something this would simplify code/tests - yes?

You need to log in before you can comment on or make changes to this bug.