| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| | | |
|
| |\ \
| | | |
| | | |
| | | |
| | | | |
Conflicts:
linden/indra/newview/viewer_manifest.py
|
| | |\ \ |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Warning was because writeToLog expected a char pointer, but was being
given string constants. Exact warning message was:
deprecated conversion from string constant to ‘char*’
|
| |\ \ \ \
| | | |_|/
| | |/| | |
|
| | | | | |
|
| | | | | |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | | |
because openal wasn't initing too, removed the check)
|
| |\ \ \ \
| | |/ / /
| | | | /
| | |_|/
| |/| | |
|
| | | |
| | | |
| | | |
| | | | |
however the old reference in install.xml is kept commented for linux32 in case it causes trouble
|
| | | |
| | | |
| | | |
| | | | |
conservative than debian oldstable (lenny)
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
llplugin* classes belong to
Yes, I confess: I grepped for unique strings in the binaries, to be absolutely sure.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
anymore after navigating to a broken one.
This also shows an error page if the login screen page is broken.
Todo: replace the trademarked logo.
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | |/ |
|
| |\ \ |
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Not sure why this happens, but is related to mForceToSaveRawImage=TRUE
(e.g. in the texture preview floater.) This workaround seems to work for
Singularity, as suggested by Siana Gears.
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Make sure partial textures are cached as partial, so that they will
finish fetching in consequent sessions. This is a rare case, but does
happen. Backported from viewer2.
|
| | |\ \ |
|
| | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Port from V2 (Snowglobe2) that adds to each texture a list of faces and
sculpt volumes that use that texture, and rebuild all sculpts when the
texture is decoded. This is needed so that we don't depend on LOD changes
which may never/rarely happen for some megaprim sculpts.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
This is a partial fix for cases where the sculpt map is larger than
128x128, like in some old sculpted trees. This was done by content
creators to beat bugs in old viewers. The sculpties still need a lod
change to trigger a rebuild once they are fully decoded, and that doesn't
always happen, because we abuse RenderVolumeLODFactor.
Ported from Snowglobe2 (GPL)
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The code to render this tool is in the viewer, but the menu option was
dropped when the advanced menu was moved to xml. This adds it back.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
The new option is Advanced -> Rendering -> Info Displays -> Sculpt.
The code for displaying this debug info is old, but the menu option was
lost when the Advanced menu was converted to xml.
|
| | | | |
| | | | |
| | | | |
| | | | | |
Missing “POOL_VOIDWATER” item in pool names array caused crash.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Prevent null dereference when the selection has no primary object,
e.g. rubber band selection.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Fix offset reading from local files (non-cached.) This is a bug, so needs
fixing, even though the offset is always 0 here, currently.
Taken from Cool Viewer (patch orginally by Max Fox.)
|
| | | | |
| | | | |
| | | | |
| | | | | |
Taken from V2/Henry.
|
| | | | |
| | | | |
| | | | |
| | | | | |
from V2.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Enable per-account settings only after we have an account. Also, make
settings that are saved locally not depend on UserInfoReply being received.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Ignore the background color specified in PNG's bKGD chunk, since it's only
there as a fallback.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Make sure partial textures are cached as partial, so that they will
finish fetching in consequent sessions. This is a rare case, but does
happen. Backported from viewer2.
|
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
Not sure why this happens, but is related to mForceToSaveRawImage=TRUE
(e.g. in the texture preview floater.) This workaround seems to work for
Singularity, as suggested by Siana Gears.
|
| | | | | |
|
| | | | | |
|
| | |_|/
| |/| | |
|
| | | |
| | | |
| | | |
| | | | |
worked the very first time. Damn.
|
| | | | |
|
| | | |
| | | |
| | | |
| | | | |
Based on a change to the old meta viewer by Tom Meta, but implimented differently.
|
| | | | |
|
| | | |
| | | |
| | | |
| | | |
| | | | |
Yes, this makes the meta-impy viewer non compliant with LL's TPVP, we don't care this is a viewer for the meta grid.
We care about privacy. We have our own effective methods of implementing bans.
|