Closed
Bug 6297
Opened 25 years ago
Closed 25 years ago
JS Navigator object returns bogus platform
Categories
(Core :: DOM: Core & HTML, defect, P3)
Tracking
()
People
(Reporter: bobj, Assigned: jud)
References
()
Details
Running xxx build on US NT4, and going to the URL
http://rocknroll/users/nhotta/publish/html/nav_obj.html
it returns:
navigator.appVersion=5
navigator.platform=Win95
navigator.language=en
navigator.userAgent=Mozilla/5.0 [en] (Win95; I)
screen.width=-1
screen.height=-1
screen.pixelDepth=-1 bits
screen.colorDepth=-1 bits, 0.5 colors
but the platform is NT4, not Win95.
I tried this on 4.51 and it returns "Win32". Better, but not
very precise...
Updated•25 years ago
|
Assignee: norris → vidur
Component: JavaScript → DOM Level 0
Updated•25 years ago
|
Assignee: vidur → warren
Comment 1•25 years ago
|
||
I'm just getting this from the nsINetService interface. Reassigning to the
netlib guys.
Comment 2•25 years ago
|
||
Verify in necko.
Assignee | ||
Comment 3•25 years ago
|
||
as it stands right now, necko will return null for all user agent components;
they aren't being initialized right now. How do we want to init them now that
they're in necko (pure xp code)? In the old world the FE would set the various
pieces as they are FE dependent. I suppose I should just have the code that is
setting them now grab an nsINetService and set them that way, but there are two
problems with that:
1. Before any urls are loaded we have to be sure that they get set via
nsINetService.
2. How do we protect ourselves from some rogue code grabbing the nsINetService
and changing the values? (there is a certain portion of the UA string that is
static and will act as our brand).
Assignee | ||
Updated•25 years ago
|
Status: NEW → RESOLVED
Closed: 25 years ago
Resolution: --- → DUPLICATE
Updated•25 years ago
|
Status: RESOLVED → VERIFIED
Comment 5•25 years ago
|
||
Maring verified.
You need to log in
before you can comment on or make changes to this bug.
Description
•