On Tue, Apr 5, 2011 at 1:00 PM, Jason Gerecke <killert...@gmail.com> wrote:
> The prior patch has gestures use button mapping like buttons et.al.,
> but for some reason this results in the cursor being "stuck" at the
> top left of the screen when:
>
> 1) A gesture is performed
> 2) The gesture uses a button with an associated action
> 3) The action presses a button but does not release it
>
> After much testing, I've found that this patch solves the issue and
> also works for non-gesture code.
>
> To be honest, I'm not sure why the old code didn't work but this does.
> I see only see one difference in execution (the old code executed for
> all 256 actions, '0' or not), but that change alone doesn't appear to
> fix the bug...
>
> Signed-off-by: Jason Gerecke <killert...@gmail.com>
> ---
> src/wcmCommon.c | 61 ++++++++++++++++++++----------------------------------
> 1 files changed, 23 insertions(+), 38 deletions(-)
>
> diff --git a/src/wcmCommon.c b/src/wcmCommon.c
> index 0d3e8d8..ff70d64 100644
> --- a/src/wcmCommon.c
> +++ b/src/wcmCommon.c
> @@ -206,7 +206,7 @@ void sendAction(InputInfoPtr pInfo, int press,
> int i;
>
> /* Actions only trigger on press, not release */
> - for (i = 0; press && i < nkeys; i++)
> + for (i = 0; i < nkeys; i++)
> {
> unsigned int action = keys[i];
>
> @@ -219,6 +219,17 @@ void sendAction(InputInfoPtr pInfo, int press,
> {
> int btn_no = (action & AC_CODE);
> int is_press = (action &
> AC_KEYBTNPRESS);
> +
> + if (!press)
> + {
> + if (!is_press)
> + break;
> + if (countPresses(btn_no,
> &keys[i], nkeys - i))
> + is_press = 0;
> + else
> + break;
> + }
> +
> xf86PostButtonEventP(pInfo->dev,
>
> is_absolute(pInfo), btn_no,
> is_press,
> first_val, num_val,
> @@ -229,6 +240,17 @@ void sendAction(InputInfoPtr pInfo, int press,
> {
> int key_code = (action & AC_CODE);
> int is_press = (action &
> AC_KEYBTNPRESS);
> +
> + if (!press)
> + {
> + if (!is_press)
> + break;
> + if (countPresses(key_code,
> &keys[i], nkeys - i))
> + is_press = 0;
> + else
> + break;
> + }
> +
> wcmEmitKeycode(pInfo->dev, key_code,
> is_press);
> }
> break;
> @@ -239,43 +261,6 @@ void sendAction(InputInfoPtr pInfo, int press,
> break;
> }
> }
> -
> - /* Release all non-released keys for this button. */
> - for (i = 0; !press && i < nkeys; i++)
> - {
> - unsigned int action = keys[i];
> -
> - switch ((action & AC_TYPE))
> - {
> - case AC_BUTTON:
> - {
> - int btn_no = (action & AC_CODE);
> -
> - /* don't care about releases here */
> - if (!(action & AC_KEYBTNPRESS))
> - break;
> -
> - if (countPresses(btn_no, &keys[i],
> nkeys - i))
> -
> xf86PostButtonEvent(pInfo->dev,
> -
> is_absolute(pInfo), btn_no,
> - 0, first_val,
> num_val,
> -
> VCOPY(valuators, num_val));
> - }
> - break;
> - case AC_KEY:
> - {
> - int key_code = (action & AC_CODE);
> -
> - /* don't care about releases here */
> - if (!(action & AC_KEYBTNPRESS))
> - break;
> -
> - if (countPresses(key_code, &keys[i],
> nkeys - i))
> - wcmEmitKeycode(pInfo->dev,
> key_code, 0);
> - }
> - }
> -
> - }
> }
>
> /*****************************************************************************
> --
> 1.7.4.1
>
>
While this patch works, I don't like the idea of submitting what I
don't understand. If somebody else wants to take a stab at figuring
out why the last patch breaks things (or alternatively, why this patch
fixes things) you're more than welcome.
Jason
---
Day xee-nee-svsh duu-'ushtlh-ts'it;
nuu-wee-ya' duu-xan' 'vm-nvshtlh-ts'it.
Huu-chan xuu naa~-gha.
------------------------------------------------------------------------------
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming
smartphone on the nation's most reliable network.
And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
_______________________________________________
Linuxwacom-devel mailing list
Linuxwacom-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel
