I made a patch for this bug that causes a “beep” sound every time Alt hotkeys are used in windowed Windows apps:
https://bugzilla.libsdl.org/show_bug.cgi?id=2669
OSX: Removed joysticks wrongly marked as closed when they are not.
This caused a few problems for me when handling hot-pluged joystick closing.
Just filed the bug with a patch included:
https://bugzilla.libsdl.org/show_bug.cgi?id=2738
Cheers!------------------------
Rodrigo Cardoso Rocha
@RodrigoRodrigoR - twitter.com/RodrigoRodrigoR
Chibata Creations - chibatacreations.com
I’ve got a patch that would be nice to get in to 2.0.4:
https://bugzilla.libsdl.org/show_bug.cgi?id=2718
It’s for an annoying bug on OS X where, if you Cmd+Tab away from an
SDL window in relative mouse mode, you’re left with a broken relative
mouse mode when you return.
I tested the fix on 10.6 and 10.10.
Thanks,
Eric
I would love if someone take a look into this too https://bugzilla.libsdl.org/show_bug.cgi?id=2738------------------------
Rodrigo Cardoso Rocha
@RodrigoRodrigoR - twitter.com/RodrigoRodrigoR
Chibata Creations - chibatacreations.com
Also, This one is just a simple fix to build SDL for OSX if HAPTIC is disabled. https://bugzilla.libsdl.org/show_bug.cgi?id=2717------------------------
Rodrigo Cardoso Rocha
@RodrigoRodrigoR - twitter.com/RodrigoRodrigoR
Chibata Creations - chibatacreations.com
I’ll try to get all these patches in today.
–ryan.On 02/01/2015 12:18 AM, RodrigoCard wrote:
Also, This one is just a simple fix to build SDL for OSX if HAPTIC is
disabled. https://bugzilla.libsdl.org/show_bug.cgi?id=2717
Mouse Buttons X1/X2 don’t work as expected on Linux/X11 (don’t generate
SDL_BUTTON_X1/2, but the values 8 and 9):
https://bugzilla.libsdl.org/show_bug.cgi?id=2310
Not sure if this is a showstopper, but I think it is bad and it would be
great if someone with X11 knowledge could fix it.
Cheers,
DanielOn Sun, Feb 1, 2015 at 7:08 PM, Ryan C. Gordon wrote:
On 02/01/2015 12:18 AM, RodrigoCard wrote:
Also, This one is just a simple fix to build SDL for OSX if HAPTIC is
disabled. https://bugzilla.libsdl.org/show_bug.cgi?id=2717I’ll try to get all these patches in today.
–ryan.
SDL mailing list
SDL at lists.libsdl.org
http://lists.libsdl.org/listinfo.cgi/sdl-libsdl.org
2015-02-01 15:22 GMT-03:00, Daniel Gibson :
Mouse Buttons X1/X2 don’t work as expected on Linux/X11 (don’t generate
SDL_BUTTON_X1/2, but the values 8 and 9):
https://bugzilla.libsdl.org/show_bug.cgi?id=2310
Not sure if this is a showstopper, but I think it is bad and it would be
great if someone with X11 knowledge could fix it.
Dunno, but I believe that X reports those two buttons as 8 and 9
indeed (because 4-5 are the vertical wheel and 6-7 are the horizontal
wheel, if I recall correctly), so that sounds like SDL is just passing
the button number from X as-is.
Yeah, that’s what’s happening, but SDL should translate them to the
expected values.
(I think it happens because 4/5 is used for vertical scrolling and 6/7
for horizantal scrolling)
The fix sounds trivial, but I’m not 100% sure if it’s always 8/9 or only
if [xyz]axis-mapping is enabled explicitly in xorg.conf or implicitly by
whatever magic would do that.
Thus “it would be great if someone with X11 knowledge could fix it.” 
Cheers,
DanielOn 02/01/2015 07:32 PM, Sik the hedgehog wrote:
2015-02-01 15:22 GMT-03:00, Daniel Gibson <@Daniel_Gibson>:
Mouse Buttons X1/X2 don’t work as expected on Linux/X11 (don’t generate
SDL_BUTTON_X1/2, but the values 8 and 9):
https://bugzilla.libsdl.org/show_bug.cgi?id=2310
Not sure if this is a showstopper, but I think it is bad and it would be
great if someone with X11 knowledge could fix it.Dunno, but I believe that X reports those two buttons as 8 and 9
indeed (because 4-5 are the vertical wheel and 6-7 are the horizontal
wheel, if I recall correctly), so that sounds like SDL is just passing
the button number from X as-is.
Here’s one more: Patch for dead keys on Windows:
https://bugzilla.libsdl.org/show_bug.cgi?id=2834
(Off-topic: What’s up with the mailing list archive? I’ve posted about
this patch before but can’t link to it… even this reply I had to
manufacture from the digest e-mail.)
-EthanOn 02/01/2015 01:08 PM, Ryan C. Gordon wrote:
I’ll try to get all these patches in today.
–ryan.
Oh, is it ?mob Ryan with patches? day? Very well. 
In Linux, touch finger coordinates are in pixels rather than normalized to [0, 1] like they should be. Here?s a bugzilla, including a patch provided by someone: https://bugzilla.libsdl.org/show_bug.cgi?id=2307
I made an improved workaround patch for the window-focus-on-close issues in OS X. It fixes the focus transfer when the ?About? window is closed: https://bugzilla.libsdl.org/show_bug.cgi?id=1825
Here?s a small patch to prefer modern APIs (when available) rather than deprecated ones for a couple cases in OS X: https://bugzilla.libsdl.org/show_bug.cgi?id=2858
And of course my big iOS patch, which I expect will take more than a bit longer to look over: https://bugzilla.libsdl.org/show_bug.cgi?id=2798On Feb 1, 2015, at 2:08 PM, Ryan C. Gordon wrote:
I’ll try to get all these patches in today.
–ryan.
SDL mailing list
SDL at lists.libsdl.org
http://lists.libsdl.org/listinfo.cgi/sdl-libsdl.org
If we’re all posting patches then here are a couple that would be good
to get in before 2.0.4:
"Fix Key and Textinput event order on X11 (issue introduced post-2.0.3)"
https://bugzilla.libsdl.org/show_bug.cgi?id=2748
"X11+IBus: SDL2 should post SDL_TEXTEDITING when preedit is cancelled by
hitting Backspaces"
https://bugzilla.libsdl.org/show_bug.cgi?id=2844
(Off-topic: What’s up with the mailing list archive? I’ve posted about
this patch before but can’t link to it… even this reply I had to
manufacture from the digest e-mail.)
Dreamhost is having problems with it; not sure what specifically. It
seems to catch up eventually.
The eventual goal is to move all this infrastructure off Dreamhost to
the Digital Ocean server where everything else is now…this will solve
several problems, but doing so will be a lot of effort and risk, so
we’ve been stalling on it.
–ryan.
Also off topic:
I feel kind of guilty bumping my patches in the mailing list like
this, but I’m worried that the patch will be forgotten in bugzilla.
Is there anything I should do differently when contributing a patch?
I’m not very good with bugzilla, but it doesn’t seem very good as a
way of tracking patches/pull requests, because they blend in with
regular bugs.
Just brainstorming, but would any of the following help the sdl dev process?
- a policy that patches from outside contributors should be sent to a
particular mailing list - a separate code review web app for posting patches
- a new bug status in bugzilla, “please review” or “pull request” or something.
Cheers,
Eric