Updating SDL to 290561, c-ares to 290399, Boost to 290566, colladadom
to 290576, curl to 290567, fontconfig to 290569, freetype to 290557,
google-mock to 290574, libpng to 290558, libxml2 to 290562, llqtwebkit
to 290578, openssl to 290560, PCRE to 290406, zlib to 290556. Document
compiling and linking in new 00-COMPILE-LINK-RUN.txt in indra/cmake. It
is the README for the building process based on things learned during
this library project. Switch building process back to style used for
new llqtwebkit2 repo. This builds shared libraries for Qt4 products
on all platforms using current build machines and TC task templates.
All platforms are building. All have comparable FPS rates. Builds are
getting quieter. Forward, comrades!
colladadom to 290229, curl to 290186, fontconfig to 290258, freetype to 290255,
google-mock/gmock/gtest to 290199, libpng to 290081, libxml2 to 290157,
llqtwebkit to 290262/250147 (to be fixed shortly), openssl to 290083,
pcre to 290150, zlib to 290080. The freetype2 headers are now normalized
to what they would normally be on a unix-type system. Libxml2 and
pcre are available on all platforms. Libraries generally move in the
direction of static archives rather than shared libraries (e.g. openssl).
The llqtwebkit side is a step back from the more ambitious goal. What
will be used (290262) will be a simple rebuild of the existing 4.7.1
library with updated package dependencies. The bigger work will be in
the future.
Update library dependencies to reference builds using new templates and
modern build hosts for: SDL, c-ares, boost, curl, libpng, llqtwebkit2,
openssl and zlib. Update manifest, library copying and 'test_apps' setup
to reflect that zlib and openssl are now always static (on Mac and
Linux). Viewer and test_apps now functional on 10.7 Mac.
Part of the switchover to dynamic Qt4. This mostly completes the
Mac side. Webkit is functional on Mac, plugins work. The two
test programs, llfbconnectest and llplugintest, need a little
more work to run without manual fixup ('.' on PATH and maybe a
symlink to libz.1.dylib and they run and function on Mac). Also
need to address libz.dylib vs libz.a. Currently resolving against
dylib which is not what I want.
Added the necessary plumbing to get link_hovered events from the webkit plugin through to the viewer UI.
This requires a llqtwebkit library built from revision 1799a899e06d or later in http://hg.secondlife.com/llqtwebkit to function. The viewer source changes are backwards-compatible with earlier versions of llqtwebkit, it just won't see any link_hovered events with previous revisions.
Reviewed by Callum.
Added support to the webkit media plugin and llpluginclassmedia for passing through the auth request/response. We still need an updated build of llqtwebkit for all platforms, as well as some UI work in the viewer to actually display the auth dialog.
This includes a Mac build of llqtwebkit from the following sources:
revision aacdf69cbf5aa12d77c179296e31ef643ed1ef4a of http://qt.gitorious.org/+lindenqt/qt/lindenqt (currently head of the 'lindenqt' branch)
revision 81ab5ae326f0 of http://hg.secondlife.com/llqtwebkit (currently head of the default branch)
Reviewed by Callum.
LLPluginProcessParent can now optionally use a separate thread for reading messages from plugin sockets. If this is enabled, it will spawn a single thread and use apr_pollset_poll to wake up the thread when incoming data arrives instead of polling all the descriptors round-robin every frame. This should be somewhat more efficient, and should also allow blocking requests from plugins to be serviced much more quickly (once we start using them). This is currently disabled by default, until it's had a bit more focused testing on multiple platforms.
Hooked up the switch to use the message read thread to the PluginUseReadThread debug setting and an item in the Advanced menu in the viewer, and to a checkbox in the UI in llmediaplugintest.
Updated some debug logging in the plugin system to have appropriate tags and not log dire-looking warnings during normal operation.
LLPluginProcessParent now once again explicitly kills plugin processes (instead of just closing their sockets and waiting for them to exit). The problem we were attempting to solve by not doing the kill (letting the webkit plugin write its cookie file on exit) has been solved another way.
LLPluginProcessParent::sendMessage() now attempts to write the outgoing message to the socket immediately instead of waiting for the next frame. This should reduce the latency of sending plugin messages.
Added a separate fast timer for LLViewerMedia::updateMedia().
Made all existing plugins send their texture_params message from this init message instead of the LLPLUGIN_MESSAGE_CLASS_BASE "init" message. (This ensures that they won't start to receive 'size_change' messages until after the init has happened.)
Added "set_user_data_path" and "set_language_code" messages to LLPluginClassMedia.
Made webkit plugin deal with the new messages, when they're sent before it receives the media "init".
Removed the user_data_path and language_code arguments from the init function calls throughout the hierarchy.
Made LLViewerMediaImpl queue up the language code and user data path messages before initializing the media.
Reviewed by Callum at http://codereview.lindenlab.com/687006 .
A viewport array passed to glGetIntegerv() and gluUnProject() was declared as
'int'. This compiles on Windows, but fails on Mac. Declaring as 'GLint' works.
Added an LLSD argument to LLPluginClassMedia::keyEvent() and LLPluginClassMedia::textInput() which contains the native key data.
Made LLViewerMediaImpl retrieve the native key data and pass it to keyEvent and textInput.
Added a native_key_data parameter to the text_event and key_event messages.
Made the webkit plugin extract the native_key_data parameter and pass it to the internal keyEvent() and unicodeInput() functions.
Fixed LLMediaPluginTest to match function signature change to LLPluginClassMedia::keyEvent().
callum_linden