Closed
Bug 1180092
Opened 9 years ago
Closed 9 years ago
implement cache-pinning of packages
Categories
(Core :: Networking: Cache, defect, P1)
Core
Networking: Cache
Tracking
()
RESOLVED
DUPLICATE
of bug 1190290
blocking-b2g | 2.5+ |
People
(Reporter: pauljt, Unassigned)
References
Details
Instead of installing, signed packages will have some kind of pinning mechanism where the user decides they want to "pin" the content. Liek bookmarking, but we will also pin the package to in the cache, so the content is persisted like an installed app.
Reporter | ||
Comment 1•9 years ago
|
||
As part of this bug we also need to take care of updating of pinned packages.
Reporter | ||
Comment 2•9 years ago
|
||
(In reply to Paul Theriault [:pauljt] from comment #1)
> As part of this bug we also need to take care of updating of pinned packages.
And also handle the case where the developer has removed the package altogether from the website. I wonder what we should do here? If the user has pinned the packaged, thats like an install, so maybe we should keep the old content, and assume the user wants to still use the content?
Comment 3•9 years ago
|
||
At the moment, web packaged apps are loaded into the cache under the system's principal, but I think each packaged app should have its own appId/principa.
So when do we generate that appId? I assume it would be when we first load the package, so we can immediately put the cache entries into the right cache-jar.
Any thoughts?
Flags: needinfo?(jonas)
The way this will work is that we'll use a particular set appid/browserflag when fetching the package. When pinning, it should be pinned in the cache with that appid/browserflag. So only network loads that use that same appid/browserflag should receive that package.
So at no point does necko need to transfer the package from one appid/browserflag to another.
(On a separate note, we're working on converting code from explicitly using appid/browserflag to instead using the OriginAttributes struct. This struct is currently mainly a wrapper around appid/browserflag, but will let us add other forms of segmenting storage in a more generic way eventually.)
Flags: needinfo?(jonas)
Reporter | ||
Updated•9 years ago
|
Priority: -- → P1
Reporter | ||
Updated•9 years ago
|
blocking-b2g: --- → 2.5+
Is bug 1190290 the same as this one? Should we dupe either one to the other?
Blocks: 1190290
Reporter | ||
Comment 6•9 years ago
|
||
(In reply to Jonas Sicking (:sicking) from comment #5)
> Is bug 1190290 the same as this one? Should we dupe either one to the other?
Yes as far as I can tell. There is further discussion in the other bug, so i'm resolving this one.
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → DUPLICATE
You need to log in
before you can comment on or make changes to this bug.
Description
•