[no subject]

As for why do I "insist" on solving it at the keyboard level... this is a
system that already exists and works somewhat well, but I think it can
practically be improved to do a good job for most cases. All I'm trying to
do is create a simple set of mapping tables that will mean that most users
don't even have to think about setting up their inputs. If we can get 90%
of users to stop having the existing input setup problem, I think that's a
huge win.

The French keyboard is special in this respect having 3 ST keys with no
modern equivalent, most layouts seem to have 0 cases of this. The chosen
answer could be "we don't think French should have these keys mapped",
which would surprise me, but if that's consensus I won't fight it. I'm just
trying to review all such cases and point them out, and try to suggest a
reasonable solution. I just want to improve the system we already have to
do as good as it can do without getting more complicated than that.

I personally think "all keys are pressable but 3 of them have funny
mappings which you might have to hunt and peck to find" is better than "3
keys are not pressable because we didn't want any non-intuitive mappings".
I don't really care what 3 keys are chosen, I just suggested 3 that seemed
the best candidates to use.

-- Brad Smith

--0000000000001b72120623e13b42
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail=
_attr">On Mon, Oct 7, 2024 at 4:44=E2=80=AFAM Miro Krop=C3=A1=C4=8Dek &lt;<=
a href=3D"mailto:miro.kropacek@xxxxxxxxx";>miro.kropacek@xxxxxxxxx</a>&gt; w=
rote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0p=
x 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir=
=3D"ltr"><div class=3D"gmail_quote"><div>Excuse me if I&#39;m asking someth=
ing obvious but why do you guys insist on solving this on the keycode level=
? French, as many other national variants, use various input methods to get=
 the desired character AltGr+&lt;key&gt;, Alt+&lt;ASCII code on the keypad&=
gt; etc) so why not just take the final text representation and compare tha=
t with the ST symbols? AFAIK SDL provides &quot;text&quot; representation a=
s part of the SDL event, so why not just take that string and let the user =
handle the input method by the OS?<br></div></div></div></blockquote><div><=
br></div><div></div><div>From my undersatnding, text input events are a sin=
gle character event, followed by repeat if held. What we need is both a key=
 down when something is pressed and a key up when it is released. Text even=
ts are the correct thing to use for filling a text entry field, but it does=
n&#39;t really work for trying to map buttons to an emulated keyboard.</div=
><div><br></div><div>As for why do I &quot;insist&quot; on solving it at th=
e keyboard level... this is a system that already exists and works somewhat=
 well, but I think it can practically be improved to do a good job for most=
 cases. All I&#39;m trying to do is create a simple set of mapping tables t=
hat will mean that most users don&#39;t even have to think about setting up=
 their inputs. If we can get 90% of users to stop having the existing input=
 setup problem, I think that&#39;s a huge win.<br></div><div><br></div><div=
>The French keyboard is special in this respect having 3 ST keys with no mo=
dern equivalent, most layouts seem to have 0 cases of this. The chosen answ=
er could be &quot;we don&#39;t think French should have these keys mapped&q=
uot;, which would surprise me, but if that&#39;s consensus I won&#39;t figh=
t it. I&#39;m just trying to review all such cases and point them out, and =
try to suggest a reasonable solution. I just want to improve the system we =
already have to do as good as it can do without getting more complicated th=
an that.</div><div><br></div><div>I personally think &quot;all keys are pre=
ssable but 3 of them have funny mappings which you might have to hunt and p=
eck to find&quot; is better than &quot;3 keys are not pressable because we =
didn&#39;t want any non-intuitive mappings&quot;. I don&#39;t really care w=
hat 3 keys are chosen, I just suggested 3 that seemed the best candidates t=
o use.<br></div><div><br></div><div>-- Brad Smith<br><br></div></div></div>

--0000000000001b72120623e13b42--