[Ohrrpgce] SVN: james/5836 Add show_virtual_keyboard() and hide_virtual_keyboard() and use them in

[Ralph Versteegen]

On 16 June 2013 07:11, James Paige <Bob at hamsterrepublic.com> wrote:
> On Sun, Jun 16, 2013 at 04:42:42AM +1200, Ralph Versteegen wrote:
>> I haven't tried this yet, but why turn on the
>> "CompatibilityHacksTextInputEmulatesHwKeyboard" option? The
>> description seems to state it is only targetted for a single app.
>
> Without it, showing the keyboard caused the keyboard to display, and
> then vanish again in a fraction of a second, and send an ENTER key Then
> it would pop up again, and vanish again with another ENTER, and it would
> repeat until close the program.
>
> When I turned on CompatibilityHacksTextInputEmulatesHwKeyboard then the
> keyboard (at first glance) appeared to start working correctly.

Wow, that's pretty crazy. Surely that's a bug in SDL.

>> Also
>> I noticed in MainActivity.java that turning on this option disables
>> the hideScreenKeyboard method, because the _screenKeyboard member is
>> never set.
>
> Ah, that explains why it would not hide automatically.
>
> Also, after more testing this morning, I can see that the first call to
> show_virtual_keyboard shows it okay, but if I leave it in screen and
> rename a hero again, then it toggles off.
>
> So I am definitely not doing things right yet. I'll try again to get it
> working without the compatability hacks.
>
>> Also, why call SDL_ANDROID_CallJavaShowScreenKeyboard and
>> SDL_ANDROID_CallJavaHideScreenKeyboard directly instead of the public
>> functions in SDL_screenkeyboard.h?
>
> Oh, I had no idea that those existed. I found the
> SDL_ANDROID_CallJavaShowScreenKeyboard just by grepping the files in
> sdl-android/projects/jni/sdl-1.2 and I failed to notice the better
> functions. I'll try them next.

They're thin wrappers, so I don't expect any difference.

>> BTW, even after implementing io_show/hide_virtual_keyboard in
>> gfx_directx, you should still check whether those function pointers
>> are NULL before calling them, otherwise using an old version of
>> gfx_directx.dll will cause a segfault
>
> How do I check a sub pointer for nullity? I can see how to check
> functions that require >0 arguments, but how do I do
>
>   IF subpointer = NULL THEN
>
> Without executing the sub or argumentless function?

That works fine. Parentheses are required when calling a function or
sub pointer for that reason.

> ---
> James
>
>> On 15 June 2013 09:58,  <subversion at hamsterrepublic.com> wrote:
>> > james
>> > 2013-06-14 14:58:21 -0700 (Fri, 14 Jun 2013)
>> > 276
>> > Add show_virtual_keyboard() and hide_virtual_keyboard() and use them in renamehero()
>> >
>> > * Does nothing on platforms/backends that don't support it
>> > * Does nothing on non-Android
>> > * hide_virtual_keyboard does not work yet, I have not figured out why, but you can manually close it
>> > ---
>> > U   wip/allmodex.bas
>> > U   wip/allmodex.bi
>> > U   wip/android/AndroidAppSettings.cfg
>> > U   wip/backends.bas
>> > U   wip/gfx.bi
>> > U   wip/gfx_alleg.bas
>> > U   wip/gfx_console.bas
>> > U   wip/gfx_fb.bas
>> > U   wip/gfx_sdl.bas
>> > U   wip/moresubs.bas
>> > _______________________________________________
>> > Ohrrpgce mailing list
>> > ohrrpgce at lists.motherhamster.org
>> > http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
>> _______________________________________________
>> Ohrrpgce mailing list
>> ohrrpgce at lists.motherhamster.org
>> http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org
>>
> _______________________________________________
> Ohrrpgce mailing list
> ohrrpgce at lists.motherhamster.org
> http://lists.motherhamster.org/listinfo.cgi/ohrrpgce-motherhamster.org