Opened 4 years ago
Closed 3 years ago
#15551 closed enhancement (fixed)
WebPositive's user agent is always x86
| Reported by: | jmairboeck | Owned by: | pulkomandy |
|---|---|---|---|
| Priority: | normal | Milestone: | R1/beta4 |
| Component: | Kits/Web Kit | Version: | R1/Development |
| Keywords: | Cc: | ||
| Blocked By: | Blocking: | ||
| Platform: | All |
Description
The platform part of the user agent should be actually platform-dependent. For example, on x86_64, I had expected something like Mozilla/5.0 (Macintosh; Intel Haiku R1 x86_64) AppleWebKit/608.1.6 (KHTML, like Gecko) WebPositive/1.2 Version/11.1 Safari/608.1.6.
This is similar to #14709, but just for the platform part.
Change History (4)
comment:1 by , 4 years ago
comment:2 by , 4 years ago
All Linux browsers which I have tested contain the architecture, e.g. Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.99 Safari/537.36 Vivaldi/2.9.1705.41. For Konqueror, this is even configurable and optional. It can be as long as Mozilla/5.0 (X11; Linux 4.12.14-lp151.28.36-default x86_64; German) KHTML/5.55.0 (like Gecko) Konqueror/5 KIO/5.55 but also just Mozilla/5.0 (X11) KHTML/5.55.0 (like Gecko) Konqueror/5 KIO/5.55.
If this can not be fixed, it is not a big problem and I would rather not change it if it causes any problems with websites. However, it looks wrong on anything but 32bit x86 right now.
comment:3 by , 3 years ago
| Component: | Applications/WebPositive → Kits/Web Kit |
|---|
comment:4 by , 3 years ago
| Milestone: | Unscheduled → R1/beta4 |
|---|---|
| Resolution: | → fixed |
| Status: | new → closed |
I would just remove the architecture from there, I see no reason for leaking this informnation to websites, how is it relevant to them?
I'm not sure if that will still allow us to pass as desktop Safari and get the non-mobile version of websites, however.