Weekly#14
Open
AlericInglewood wants to merge 133 commits intomccabe:b5074f1a8e3306f1be9e93172d26c97ff088cea0from
AlericInglewood:weekly
Open
Weekly#14AlericInglewood wants to merge 133 commits intomccabe:b5074f1a8e3306f1be9e93172d26c97ff088cea0from AlericInglewood:weekly
AlericInglewood wants to merge 133 commits intomccabe:b5074f1a8e3306f1be9e93172d26c97ff088cea0from
AlericInglewood:weekly
Conversation
There was a problem with #include "hippolimits.h" because the file is called hippoLimits.h (with an uppercase L). After some discussion on IRC it was reluctantly decided that filenames should be lowercase, as is done for every Linden file. Moreover, hippolimits.h is about the class HippoLimits, with uppercase H, so even that doesn't match. Of course, then we immediately rename all hippo* files to lowercase. This patch ONLY renames the file, with no other changes. This is done in order to make sure that git won't get confused ;). All other necessary changes will be in the next commit.
There was a problem with #include "hippolimits.h" because the file is called hippoLimits.h (with an uppercase L). After some discussion on IRC it was reluctantly decided that filenames should be lowercase, as is done for every Linden file. This is part 2 or 2 (the first part just renamed the files). This part changes all files to make the viewer compile again after the renaming.
The loading.html page is no longer used. Instead embedded html found in panel_login.xml is used. I couldn't get it to show an image, but at least the "loading..." text is now centralized and I changed the background color from black to imprudence-purple.
Basically, this Linden code is just horribly broken. I spend already too much time on this (two days) for dead code, so I settled for a minimal fix for this particular case that works, instead of fixing their code properly. The problem is that without specifying "bottom" the code adds 4 pixels (VPAD) between whatever widget that is added and it's parents top (in LLView::createRect). In order to be sure that bottom is exactly minus the height, I had to specify the height too; hopefully this value is the same for everyone cause theorectically it depends on the font size (4 + fontheight + 4). The default height of the Linden code is only 4 + fontheight (they leave off the VPAD).
libpangocairo-1.0.so* was missing. I did a complete new rebuild of the package with, as usual, debian Lenny (stable) package. This tar ball contains the following debian packages: libatk1.0-0_1.22.0-1_i386.deb libatk1.0-dev_1.22.0-1_i386.deb libcairo2_1.6.4-7_i386.deb libcairo2-dev_1.6.4-7_i386.deb libpango1.0-common_1.20.5-6_all.deb libpango1.0-0_1.20.5-6_i386.deb libpango1.0-dev_1.20.5-6_i386.deb libgtk2.0-0_2.12.12-1~lenny2_i386.deb libgtk2.0-dev_2.12.12-1~lenny2_i386.deb libpixman-1-0_0.10.0-2_i386.deb libpixman-1-dev_0.10.0-2_i386.deb But only the 'copyright' file from doc/*, only the include/* files and only the lib*.so files. Most notably, not pkgconfig, resources, configuration files and modules... (those are absent in the previous prebuilts too).
…#685), patch by Nicky Perian
…es to gstreamer post-webkit_plugins in viewer_manifest.py
… changes to llSensor LSL function
There was a problem with #include "hippolimits.h" because the file is called hippoLimits.h (with an uppercase L). After some discussion on IRC it was reluctantly decided that filenames should be lowercase, as is done for every Linden file. Moreover, hippolimits.h is about the class HippoLimits, with uppercase H, so even that doesn't match. Of course, then we immediately rename all hippo* files to lowercase. This patch ONLY renames the file, with no other changes. This is done in order to make sure that git won't get confused ;). All other necessary changes will be in the next commit.
There was a problem with #include "hippolimits.h" because the file is called hippoLimits.h (with an uppercase L). After some discussion on IRC it was reluctantly decided that filenames should be lowercase, as is done for every Linden file. This is part 2 or 2 (the first part just renamed the files). This part changes all files to make the viewer compile again after the renaming.
The loading.html page is no longer used. Instead embedded html found in panel_login.xml is used. I couldn't get it to show an image, but at least the "loading..." text is now centralized and I changed the background color from black to imprudence-purple.
Basically, this Linden code is just horribly broken. I spend already too much time on this (two days) for dead code, so I settled for a minimal fix for this particular case that works, instead of fixing their code properly. The problem is that without specifying "bottom" the code adds 4 pixels (VPAD) between whatever widget that is added and it's parents top (in LLView::createRect). In order to be sure that bottom is exactly minus the height, I had to specify the height too; hopefully this value is the same for everyone cause theorectically it depends on the font size (4 + fontheight + 4). The default height of the Linden code is only 4 + fontheight (they leave off the VPAD).
…ting (removes the 'Extras' panel from Prefs > Advanced and moves the color options from 'Text Chat')
* Text Chat -> Chat * Voice Chat -> Voice * Communication -> IMs & Logging * Colors -> Chat Colors
…chat colors more logically
… to the Imprudence support page
… General tab. This will be handled by the login manager
…eds to be redone and turned into bottom_deltas)
…ckname#' to match the rest of settings.xml. Nicknames set before this commit will need to be set again due to the new preferences name
Uncomment the code needed for this panel to work. This was commented out in the inital Snowglobe media plugings merge, and somehow we got that too.
…, and imposter count. Also increased the draw distance range to 1024 from 512 and changed the increments from 8 to 4 as better defaults
…xed the layout and the default being too large (2048 isn't supported on SL. This problem needs a different solution)
…when using it anymore
* 1280x720 (HDV720) * 1440x1080 (HDV1080) * 1920x1080 (HD1080)
Apparently the code of Imprudence was from before this bug was introduced. Nevertheless, decided to add this one-liner so that the code becomes the same (in this regard) as snowglobe 1.x and (soon) viewer 2.x (see https://jira.secondlife.com/browse/STORM-423).
Bind control-shift-G to Groups. This is also done in snowglobe.
Patch by Lance Corrimal taken from SNOW-717.
…ng show group in profile option in the group window from Ascent Viewer" since it actually causes the bug it's supposed to fix This reverts commit 46bcc05.
…nsock2.h is always included before windows.h throughout the source to prevent collisions
… in CheckLevelMap Wrapped gSettings and the global objects returned by Settings::get() and Globals::get() in AIThreadSafe, forcing thread-safe access.
Force UseExternalBrowser to on. For two cases (automatic script help and explicitely opening a browser from the menu), call LLWeb::loadURL rather than LLWeb::loadURLInternal, which now will default to calling LLWeb::loadURLExternal. Finally, removed weird /icanhas* builtin gestures that open all kinds of imprudence specific 'easter egg' webpages. This remove made more sense than changing opening the imprudence webpages in an external browser or even reproducing the pages on a moonworld website %-).
Also removed most other grids. Will have to clean up this file later. Change the setting 'GridUpdateList' to a temporary test URL.
The following menu items are retained: 1. File Menu - Only Snaphot and Quit. 2. Edit Menu - Only Preferences. 3. View Menu - Only Mouselook, Reset View and Zoom Level. 4. World Menu - Only Stop Animating My Avatar. 5. Tools Menu - Completely hidden. 6. Help Menu - Completely hidden. 7. Advanced Menu - Completely hidden. This is achieved by adding attributes moonworld="true" to those items, in indra/newview/skins/default/xui/en-us/menu_login.xml and in indra/newview/skins/default/xui/en-us/menu_viewer.xml that should remain, and of course by adding support for this new attribute. This approach minimizes the size of the patch and therefore allows for easy merging with future upstream changes. It is also client-friendly: the client can easily add or remove these attributes when they change their mind (note that this attribute only works with menu_bar, menu, menu_item_call, menu_item_separator and menu_item_check).
The Setting 'LightShareAllowed' is set to 2 (always allow LightShare being set by the region) and made non-persistent. Also, in the preferences panel the choice to change it has been removed.
Approach is the same as with normal menu's: The default value of the 'moonworld' attribute (when it is not there) is 'true', causing everything to be added as before. By adding the 'moonworld' attribute, the default for that node's children becomes 'false'. If the attribute is true, the node is added, if it is false (or absent, and the parent has the attribute) then the node and all it's children are skipped. This allows to restrict the pie menu's with only a minimal amount of changes to it's xml file and makes it flexible to enable/disable things later.
This required more changes to indra/newview/skins/default/xui/en-us/panel_toolbar.xml than adding 'moonworld="true"': that results in the two buttons using up the whole window width. So, I set the width of the buttons to a fixed size of 100 pixels now, and have them follow the right side of the window.
As per Dennis' email: Pie Chart on Land ==> not needed, please remove entirely Pie Chart on right click of another avatar ==> not needed, please remove entirely Pie Chart on right click of self ==> please remove all options except for "Take Off" Pie Chart on right click of attachments worn by self ==> please remove all options except for "More.../Detach"
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Nobody is REALLy sure about this patch, but it seems the most correct thing to do.
An alternative would be to keep the filenames as they are and just fix four include lines,
but I can't really see a reason to do that, as I don't think that the size of the patch matters.
What matters is the end result.