Closed Bug 66301 Opened 25 years ago Closed 8 years ago

C style function callback change notification needs to be removed.

Categories

(Core :: Preferences: Backend, defect, P3)

All
Windows 2000
defect

Tracking

()

RESOLVED WONTFIX
mozilla1.2alpha

People

(Reporter: jud, Unassigned)

References

Details

(4 keywords, Whiteboard: [T2])

nsIPref has two listener/callback registration mechanisms causing confusion and inconsistency. The nsIObserver mechanism is preferred, and the old PrefCallback mechanism needs to be pulled out (and existing users of it need to be changed over to the new nsIObserver mechanism).
Keywords: embed
Bulk assigning bugs covered by the rewrite of nsPrefs to myself.
Assignee: dveditz → bnesse
Bulk updating some fields and accepting
Status: NEW → ASSIGNED
Keywords: nsbeta1
Hardware: PC → All
Target Milestone: --- → mozilla0.9
Blocks: 70229
Target Milestone: mozilla0.9 → mozilla0.9.1
The C style callbacks are not supported for users nsIPrefBranch. This bug now serves as a reminder that all current consumers of nsIPref need to be converted to using nsIPrefService and nsIPrefBranch.
Keywords: nsbeta1arch
Target Milestone: mozilla0.9.1 → mozilla0.9.2
Priority: -- → P3
Bumping to 0.9.3.
Target Milestone: mozilla0.9.2 → mozilla0.9.3
Updating milestones
Target Milestone: mozilla0.9.3 → mozilla1.0
Currently nsIPref is used in 208 files in about 10 modules... Does it makes sense to file separate bugs for each module and then make this one dependant on them? Otherwise patch is gonna be a way too big
Yes, the removal of nsIPref is a big job. Filing separate bugs on the modules and making this a meta bug would be a reasonable way to attack this. Especially if the module owners would take ownership of their own cleanup. :)
I don't think we can expect module owners to be responsible for converting their own code, since this is such a well-used interface. We can request that they start using the new API however :)
But having separate (smaller!) bugs on modules you can get some help outside. Also it'll be possible to get rid of it step by step. Otherwise this bug will leave forever. :-) Another point - having such bug assigned to module will make it more visible to owners ;-)
Removing the 1.0 milestone to avoid having Asa do it for me. :)
Target Milestone: mozilla1.0 → ---
Target Milestone: --- → mozilla1.2
Keywords: topembed
Keywords: mozilla1.0+
EDT topembed-, don't think it is a blocker. embed since it would still be nice to have
Keywords: topembedtopembed-
->peterl
Assignee: bnesse → peterl
Status: ASSIGNED → NEW
Whiteboard: [T2]
--->reassign to default owner of prefs module
Assignee: peterl → bnesse
bnesse is no longer around. any takers?
Assignee: bnesse → nobody
Keywords: helpwanted
90 callers, not exactly light lifting. There are a few options. * we could try and get each module owner to sign up to do the work * we could inform the module owner that if they don't the work will be done for them and their disinterest will be treated as moa for it. cookie -- morse pics -- can someone just let me cvs remove it? wallet -- morse typeahead -- aaronl -- this is new code, some sr's should be strung up for allowing it in, but aaronl knows. security -- kaie proxies -- darin content [intl] -- smontagu xulcache -- (jrgm?) -- i'll do it dom [caret] -- aaronl dom -- jst gfx -- (each port) -- i'll do it layout -- dbaron widget [mac] -- pinkerton widget [xp] -- bryner bootstrap [mac] -- pinkerton search -- rjc addrbook -- dmose in short, i can do this work at my leisure, and will if people give me approval in advance or some specific instructions for how to get prompt approval. if you were cc'd and feel that i wrongfully selected you as being responsive for some module, please indicate someone else. the list is based on an lxr search for 'RegisterCallback'. What i don't want to have to do is spend time hunting down or appologizing to 15module owners.
I think this has a much better chance of getting done if you change all of the callers yourself. As far as I know I'm not the module owner of anything though.
Frankly, since this is only on nsIPref, which is now deprecated in favor of nsIPrefBranch / nsIPrefService - I would say this is just a WONTFIX. I don't see the advantage of doing the work.
QA Contact: bugzilla → preferences-backend
Removing QAwanted from this bug since it didn't come with any specific requests and it was added 12 years ago. Please re-add if you need anything specific from QA.
Keywords: qawanted
nsIPref is long gone. libpref still has two listener/callback registration mechanisms -- an observer one and a callback one -- but this is reasonable and not causing problems. WONTFIX.
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.