| Age | Commit message (Collapse) | Author |
|---|
|
Now that the viewer's own background updater logic is responsible for
downloading a new installer, the only functionality we still use in
linux-updater that couldn't be expressed more simply in bash is the UI. But
since most Linux distros capable of running SL at all have zenity, and all
will have xmessage, we can handle even the UI part. Add xmenity wrapper script
so update_install doesn't have to care which is present, and make the bash
script that used to launch linux-updater do the real work.
|
|
|
|
- Fixed an issue where a slash was prepended to the beginning of the filename (which happened to be the full path)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
- LLAppDelegate header relocated to LLWindow. Definition is still present in secondlife-bin (for compatibility reasons when loading a nib).
- Return key handling fixed.
- Command key now acts the same as the control key by issuing control character codes when the command key is pressed.
- We now retrieve the window pointer directly from the app delegate in LLWindow.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Fix dangling references.
|
|
characters.
|
|
initialized properly within LLAppViewer::init(). Otherwise, the viewer will sporadically crash on start up.
|
|
|
|
|
|
|
|
- Removed the callback system in favor of simply defining functions in a header to later be implemented in whichever file is most convenient for what we want to do (i.e., calling LLWindow callbacks within LLWindowMacOSX, setting cursors in llwindowmacosx-objc.mm, etc.)
- Viewer shutdown now works appropriately
- Added a bit of debugging code to test if a key has been handled by the UI or not (useful for tracking down the mystery of the enter key not being handled)
- Setup a cocoa quit handler within the application delegate that intercepts any termination requests
|
|
|
|
- Updated to .xib format. To produce a new nib (which is required after changing the xib file), use "ibtool SecondLife.xib --compile SecondLife.nib" within the newview directory.
- xib file now defines the viewer's window. VIews are still being pragmatically added to the main window. This may change in the future.
- LLAppViewer's main loop has been slightly refactored to be executed on a timer for OS X. This probably needs a bit more work.
- Event handling still needs more work to work within a timer based event loop. It works fairly sporadically at the moment, as if events are being dropped between timer executions, at least for the mouse.
- Carbon has been purged from the viewer's startup, and from LLWindow entirely. There's likely still a few odds and ends fragmented throughout the viewer and its dependencies. Need to track these down.
- LLAppViewerMacOSX now uses NSApplicationMain, and also implements the NSApplicationDelegate protocol in LLAppDelegate.
- Fullscreen support has been implemented for OS X 10.7+
This is still a work in progress, however at this stage everything needed for a functional viewer is (mostly) complete. From here, it's mostly just bug hunting and fixing.
|
|
|
|
|
|
|
|
|
|
|
|
http phase 1 Some missing counter initialization kept the debugger
from entering the startup state giving the appearance of a do-
nothing floater. Also found some unbound recursion that might need
looking at in the future. (There's a comment.)
|
|
request. During readcallback, would generate an overrun-type message
about reading position beyond end-of-data. Mistake was is messaging
when state is exactly at end of data (which is expected) versus an
overrun. Both result in declaring end-of-data to libcurl. Also
changed some of the status logging for the metrics payload to be
less chatty on success, more informative on error.
|
|
|
|
Now that we've backed out the premature push, reapply the actual fix to
propagate it appropriately.
|
|
This changeset was prematurely pushed to the wrong repository. Back it out
to propagate it the right way.
|
|
llhttpclient_test.cpp is only supposed to be run by test_llsdmessage_peer.py,
as specified in llmessage/CMakeLists.txt. test_llsdmessage_peer.py sets the
environment variable PORT before running INTEGRATION_TEST_llhttpclient. While
it's not yet clear under what circumstances INTEGRATION_TEST_llhttpclient
could find PORT not set (getenv() returning NULL), it's obvious that in that
case, the previous code might well crash. Make it at least fail with an
explanatory message rather than a mysterious large negative integer. In
theory, occurrences of that message could help us solve the underlying issue.
|
|
|
|
|
|
|
|
|
|
|
|
|
