Age | Commit message (Collapse) | Author |
|
Reverted Eli's fix 11355 : a55a8ed85023 (VWR-17833 en_xui_change to cover PT (logest translation))
Fixed buttons width in Friends tab in Polish localization.
Also fixed group tab buttons.
Reviewed by Vadim Savchuk - https://codereview.productengine.com/secondlife/r/247/
--HG--
branch : product-engine
|
|
New build of libcurl fixed it.
|
|
|
|
media plugin crashes after loading acrobat.com).
|
|
Squire asked me to route display name lookups via a simulator capability
using HTTP GET. This buys us authentication until the data services team
can implement a public-facing authenticating web service for People API.
|
|
This means display names will only work on grids with name code deployed
|
|
|
|
Updated mac llqtwebkit build from the following sources:
revision aacdf69cbf5aa12d77c179296e31ef643ed1ef4a in http://qt.gitorious.org/+lindenqt/qt/lindenqt (currently head of the 'lindenqt' branch)
revision f35a5eab8c2f in http://bitbucket.org/lindenlab/llqtwebkit/ (currently head of the 'cookie-api' branch)
|
|
|
|
|
|
|
|
|
|
|
|
Bao agreed with my initial version after all.
|
|
|
|
Add yet more paranoia-and-return-or-assert, this time for +-INF in input vector.
To be reviewed by Bao.
(transplanted from 2012f21d23c66179345fb26148cfbe19835b325e)
|
|
|
|
|
|
|
|
|
|
Add yet more paranoia-and-return-or-assert, this time for +-INF in input vector.
To be reviewed by Bao.
|
|
Add yet more paranoia-and-return-or-assert, this time for +-INF in input vector.
To be reviewed by Bao.
(transplanted from 2012f21d23c66179345fb26148cfbe19835b325e)
|
|
|
|
If you or others are a cloud, then your nametag will include a "(Loading...)" string.
Note: this required a small bit of code refactor so that you can query if an avatar (yourself or others) is a cloud; this was basically a rename from updateIsFullyLoaded into getIsCloud.
|
|
|
|
from bao's review.
(transplanted from 061b8b55838634425455ebc4ac046462dfc0e5f1)
|
|
Imported from 2.1 into 2.0.1. This fix is a safe change and gives Residents useful information about why an avatar is a cloud (e.g. because the avatar minimized his window and hence is "away").
|
|
l10n cycle; DE linguistic
|
|
from the end'.
Also, more comments.
Also, try to avoid a seek when offset==0 in read.
To be reviewed by Bao.
(transplanted from 60eb013221a83ff76054ab7dcb1d07c108f51a9d)
|
|
close(). to be reviewed by bao.
(transplanted from 17c6af7a2396f9890aab8226356a556b4d58a7d8)
|
|
handling of "/me " and "/me'" messages by LLFloaterScriptDebug.
Reviewed by Mike Antipov at https://codereview.productengine.com/secondlife/r/237/
--HG--
branch : product-engine
|
|
control floater is opened)
Added force hiding of Stand button to avoid seeing Stand & Move buttons at once for a short moment.
Reviewed by Vadim Savchuk at https://codereview.productengine.com/secondlife/r/233/
--HG--
branch : product-engine
|
|
|
|
|
|
|
|
|
|
|
|
|
|
found out how to fix loading 32-bit cursors from embedded resource and moved them back there instead of freestanding files
|
|
|
|
|
|
reviewed by Richard cc#189
|
|
|
|
|
|
reviewed by monroe
|
|
|
|
DEV-48903 - malformed names in p2p dialogs for Diamondware
In the merge there were some changes that didn't get propagated that resulted in some failures with
respect to diamondware name processing. THe determination as to whether a p2p session was with
an avatar or with an alcatel (pbx) wasn't propagated into the diamondware code.
I fixed that by merging the changes to the vivox module with respect to that into the diamondware module.
Also, the acceptance dialog was vanishing too quickly as the determination as to whether there was still
a p2p invite pending was not updated when the calling code was updated.
The calling code initially was changed to look into vivox state which isn't really valid. that was changed in
the 'trunk' to another mechanism, but that still assumed visibility into the voice code.
I fixed that by creating an API call to ask whether a call was pending.
CR: Karina
|
|
|
|
|
|
The code was using the "authority" part of the URL returned by LLURL when constructing the cookie's domain, which incorrectly included the port number. This change makes sure any port number and username/password that might be in the authority are stripped from the string before handing it to the cookie code.
Reviewed by Gino at http://codereview.lindenlab.com/1297006
|