Inconsistent handling of prescontext's link handler and container

RESOLVED FIXED

Status

()

Core
Document Navigation
RESOLVED FIXED
13 years ago
10 years ago

People

(Reporter: bz, Assigned: Brian Ryner (not reading))

Tracking

Trunk
x86
Linux
Points:
---
Dependency tree / graph

Firefox Tracking Flags

(Not tracked)

Details

Attachments

(1 attachment)

Before bfcache we have the invariant that the prescontext's link handler and
container are both the same -- the docshell.   With bfcache, we null out the
link handler but not the container in cached presentations.  We should sort our
exactly what these do and figure out whether we should be nulling both, neither,
or only one and why, then document it.

I also think we should just nuke both members and use the document's container,
but that may confuse printing....
(Assignee)

Comment 1

13 years ago
Created attachment 186773 [details] [diff] [review]
null out the prescontext container
Assignee: nobody → bryner
Status: NEW → ASSIGNED
Attachment #186773 - Flags: superreview?(dbaron)
Attachment #186773 - Flags: review?(dbaron)
Comment on attachment 186773 [details] [diff] [review]
null out the prescontext container

r+sr=dbaron, assuming it doesn't need to be re-set elsewhere :-)
Attachment #186773 - Flags: superreview?(dbaron)
Attachment #186773 - Flags: superreview+
Attachment #186773 - Flags: review?(dbaron)
Attachment #186773 - Flags: review+
(Assignee)

Comment 3

13 years ago
Comment on attachment 186773 [details] [diff] [review]
null out the prescontext container

Requesting approval; only impacts fastback.
Attachment #186773 - Flags: approval1.8b3?
Comment on attachment 186773 [details] [diff] [review]
null out the prescontext container

a=me for 1.8b3.

/be
Attachment #186773 - Flags: approval1.8b3? → approval1.8b3+
(Assignee)

Comment 5

13 years ago
checked in
Status: ASSIGNED → RESOLVED
Last Resolved: 13 years ago
Resolution: --- → FIXED

Updated

13 years ago
Blocks: 297488
I'd still like to see some documentation on what these members are and why just
nulling out both makes sense (in particular why no one ever assumes they're
non-null, which they generally are in normal browsing).

Updated

10 years ago
Component: History: Session → Document Navigation
QA Contact: history.session → docshell
You need to log in before you can comment on or make changes to this bug.