Hyperion Entertainment Message Boards

4 keys ... one to be typed direct and 3 mode keys for input method editing

*ALL* Japanese keyboards will need those extra keys mapped

I have tried the Allkeysprogram learning mode and that also failedto show anything

I have also got more to do with perception-ime to setup input events buffers so that I can feed edited sequences to the system

simply changing the font is not enough as Japanese uses 4 pictographic sets including English characters

Do you have or have you created a Japanese keymap?

You may already know this, but keymaps are in the SYS:Devs/KeyMaps/ drawer and they are text files which can be edited with a text editor. So you can add new keys just by editing the file.

For example, a defintion for the key Q is as follows (usa_int_ISO_8859-15 keymap):
In order to make a new definition for a key, you must know its raw keycode. For the key Q it's 10.

Spektro wrote:Do you have or have you created a Japanese keymap?

You may already know this, but keymaps are in the SYS:Devs/KeyMaps/ drawer and they are text files which can be edited with a text editor. So you can add new keys just by editing the file.

For example, a defintion for the key Q is as follows (usa_int_ISO_8859-15 keymap):

Code: Select all

KEY 10 ALONE "q" SHIFT "Q" ALT "å" SHIFT_ALT "Å"

In order to make a new definition for a key, you must know its raw keycode. For the key Q it's 10.

Yes I have written a new keymap for Sys:Devs/Keymaps called Japanese_ISO-2022-JP

Modified from the US keymap

in searching and mapping keys I have 4 keys that appear to have no rawkey values that I can see


the Japanese keyboard I am using is fully functioning except for those 4 keys

The Japanese keymap is available to everyone right now but remains incomplete
I have also tried using extra rawcodes to find ones that work without success of any kind

I am willing to leave the extra jp2,jp3,jp4 keys unmapped however
they still need to have a raw input event value for the system
I also willing to volunteer a keyboard to a developer to make any kind of Japanese support viable as I am also wondering what input events I need to buffer for Japanese to be supportable as any kind of input option

would you tolerate keys on your system remaining unused because the OS ignores the keyboard codes?

I will be posting pictures of the two keyboard kits that I am testing with

I know this is a software issue having used the same keyboards with linux on the 440 flex that I have

any further questions that need clarification?

An early copy of the keymap is http://code.google.com/p/perception-ime ... SO-2022-jp

I still need to update this and have a completely functioning keymap

If AllKeys doesn't 'see' you pressing the key, then it's a driver (USB HID driver?) or input.device issue (maybe keymap.library, but I think commodities intercepts higher up the chain than that).

I get the same problem with my UK Apple keyboard - several keys do not get picked up at all by AllKeys.

Belxjander wrote:I know this is a software issue having used the same keyboards with linux on the 440 flex that I have

Is there some tool on Linux which can show the raw keycodes sent by those keys? I'm thinking if we knew the exact codes not being sensed by (whatever module in) AmigaOS, it might be easier for someone with access/knowledge to find out why and where and whether it can be changed.

Best regards,

Niels

chris wrote:If AllKeys doesn't 'see' you pressing the key, then it's a driver (USB HID driver?) or input.device issue (maybe keymap.library, but I think commodities intercepts higher up the chain than that).

I get the same problem with my UK Apple keyboard - several keys do not get picked up at all by AllKeys.

Yes, Commodities deals with input events before intuition in the input event stream

I am looking at making an updated keyboard driver but where is keymap processing?

If processing is input device then it may need an update but if it is entirely the keybover then I can do something on my own

--- update ---
http:// dl.dropbox.com/u/11964747/AmigaOS4/Keyboard.jpg

one key left from the backspace key
one key left from the space key
two keys right and one key right from the space key

I will be trying a modified install later today...

nbache wrote:

Belxjander wrote:I know this is a software issue having used the same keyboards with linux on the 440 flex that I have

Is there some tool on Linux which can show the raw keycodes sent by those keys? I'm thinking if we knew the exact codes not being sensed by (whatever module in) AmigaOS, it might be easier for someone with access/knowledge to find out why and where and whether it can be changed.

Best regards,

Niels

There is such a tool and I will post what I can get though I would like to dual boot linux and Amiga OS from the existing partition layout

Belxjander, if you are interested, I have a program which reads the state of every key in the keyboard through the keyboard device. It shows which keys are down and what are the raw key codes. However, it won't work with USB keyboards.

Spektro wrote:Belxjander, if you are interested, I have a program which reads the state of every key in the keyboard through the keyboard device. It shows which keys are down and what are the raw key codes. However, it won't work with USB keyboards.

I only have USB as an option on my sam440 flex setup

so unless that is the issue I need to resolve other than dual booting linux?

I have grabbed the rkeyboard driver with source to test as well

Belxjander wrote:I only have USB as an option on my sam440 flex setup

Ok, my program won't help you then.

Niels' suggestion of using a Linux tool is good. The keycodes you get from the tool makes it possible to file an enhancement/bug report for the AOS4 developers.