Previously it was possible for a game with the same path and name to be in the list as another. This is annoying because duplicates ae (obviously) no different from the initial item.
This prevents duplicates from entering the list.
The way this works is:
1. We get the final list of items to add to the list.
2. Loop through it using two indices, which, for this explanation I'll call [item] and [itemAfter]
We compare path name at item with index [item] and the path name at item with index [itemAfter]
To phrase this numerically comparison works like so:
for (int i = 0; i < listSize; i++)
{
if (i+1 < listSize)
item[i].getPath().equals(item[i+1].getPath())
}
3. For each path comparison that is true, remove item at [indexNext].
Since FolderBrowserItems have an 'isDirectory()' method, that's all we need to care about now. There's no need to check subtitles to determine if an item is a directory anymore.
SDL2.0 removed SDL_HAPTIC_SQUARE because of ABI issues (see comment #7 on issue
6491 by Ryan C. Gordon from the SDL project). It will be reintroduced again in
2.1, so keep the code and #ifdef it away.
This can happen if the user does not have an up to date version of
the DirectX SDK, as Microsoft intentionally broke it and requires
users to install the W8 SDK.
If we don't do this, then when the game starts we'll send out the
buffer size to clients being a super large value of junk, and they'll
hang forever trying to accumulate an input buffer a size that they'll
never ever reach in a million years.
This never manifested in release builds for some reason.
Made some class variables final, since they should convey that they cannot be changed after the first assignment.
Made the formatting consistent between files.
In some cases, it would fail to inflate correctly in the sense that it would only show the binding status and not the name of the actual control that was being binded.
Also, join a variable's declaration and assignment in the onMotionEvent() function.
If no explicit accessibility term (private, protected, public, etc) is given, then the permission level is set to something called 'package-private' which means it is set to the scope of the whole package. So technically any class could have access to this method and variable, which is likely not what we want.
This prevents inheritance of the classes (will throw a compiler error if you try and extend any of the classes).
This is mainly syntactical sugar and form. Nothing major.
This is what it now looks like: http://i.imgur.com/KOZgA1i.png
As usual, if any bugs arise from this rather large change. Please report it so I can fix it.
- Join variable declaration and assignments in function onTouchEvent()
- Change a for-loop into a foreach loop in dispatchGenericMotionEvent(). Makes the loop body a single statement.
[streamtype].flush() is called when [streamtype].close() is called.
No need to null the references either after calling close(), the garbage collection will take care of it.
Added an example translation (Japanese). So now the Android version can both display in English and Japanese, depending on what the Android device's system language is set to.
Also did a tiny clean-up of InputConfigItem.java so that the parameters are slightly more descriptive.
Now, to do a translation in [x] language, all you have to do is take the normal English strings.xml and translate the XML entries into said language, and simply make a folder in the /res/ sub-directory in the form of values-[region code]. IE) With the Japanese translation, it is in the folder /res/values-ja
No configuration other than that is needed. After doing the above, the language should load fine on any device when set to that specific system language.
By default, if a translation file does not exist for a given system language. The app will automatically fall back to using the English translation.
This *should* be bug-free since I did check everything multiple times. But if any issues occur, please report them so that I can fix them.
The MS INI parser and most other INI parsing libraries APIs only support
comments at the beginning of lines. Right now, some Game INI files use sections
like:
[OnFrame]#Add memory patches here
But these section headers are parsed separately, so this should not break
them.
This reverts commit cce809ac90.
The code was actually correct: "expr" is never allocated when an error is
returned. This means when the expression parser fails, deleting "expr" means
deleting an uninitialized pointer.
It's possible to configure to use the vertex color as lightning source without enabling the vertex color at all.
The old implementation will use zero, but it seems to be wrong (prooven by THPS3), more likely is to disable
the lightning and just return the global color.
This fixes THPS3 on OpenGL, but it isn't verifed on hardware
The current shader uses bit operations which aren't supported by glsl120.
A workaround with round + frac + lots of additions would be possible, but unreadable.
So I think it isn't worth
But this fixes the annoying shader compilation error message
This would allow a new socket to be created with the same port after
we close it. However, we can't reuse it immediately because of the TCP
TIME-WAIT state.
This should be transparent, but it may cause regressions.
The idea here is that now all players, including the host of the server,
talk to the server through TCP/IP networking. This significantly reduces
our codepaths through netplay, and will prevent strange local-only bugs
from happening.
The cleanup isn't 100% finished yet. The NetPlay dialog still drives the
server through private APIs. I eventually want to sanction off the server
entirely, so all communication is done through TCP/IP. This will allow us
to have high-traffic public servers that can relay multiple games and
lobbies at a time, and split off channel and game management to people
other than the host.
This is all still just a pipe dream, though.
Ryan Houdek
