Opened on 11/11/2016 at 09:59:50 AM
Closed on 10/09/2019 at 08:21:26 PM
#4632 closed change (rejected)
Store sitekeys by frame ID instead of frame URL
Reported by: | kzar | Assignee: | |
---|---|---|---|
Priority: | P3 | Milestone: | |
Module: | Platform | Keywords: | |
Cc: | sebastian, greiner, esanchez | Blocked By: | #4551 |
Blocking: | Platform: | Unknown / Cross platform | |
Ready: | no | Confidential: | no |
Tester: | Unknown | Verified working: | no |
Review URL(s): |
Description
Background
So far we store sitekeys in a PageMap of frame URL to sitekey Objects. This generally worked OK but did require a workaround for sitekey whitelisting to work directly after a redirection (#4483).
We did this since Safari didn't give us separate page IDs for pre rendered pages and therefore we couldn't simply use the page and frame IDs. Now that we're no longer concerned with Safari support this is no longer the case.
What to change
- Track sitekeys by frame ID instead of frame URL, still in a PageMap for now. (PageMaps will be removed in issue #4580.)
- Remove the work around required for #4483 where we update the page structure eagerly from the onHeadersReceived listener and keep track of which pages were updated eagerly.
- While at it remove the logic to take a sitekey from the data-adblockkey attribute of the html element which was only required for Safari anyway.
Attachments (0)
Change History (5)
comment:1 Changed on 11/11/2016 at 10:01:32 AM by kzar
comment:2 Changed on 11/11/2016 at 08:16:42 PM by esanchez
- Cc esanchez added
comment:3 Changed on 11/17/2016 at 04:40:12 PM by kzar
- Owner kzar deleted
comment:4 Changed on 10/09/2019 at 06:28:41 PM by greiner
- Component changed from Unknown to Platform
comment:5 Changed on 10/09/2019 at 08:21:26 PM by sebastian
- Resolution set to rejected
- Status changed from new to closed
The changes seem reasonable, but rejecting the issue as we no longer use Trac. If anyone wants to work on it, please file a new issue on GitLab.
Note: See
TracTickets for help on using
tickets.
The discussion in the review for #4599 reminded me about this. Does the description look correct to you Sebastian?