[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Bug-guile-ncurses] Unicode in form-driver
From: |
Mike Gran |
Subject: |
Re: [Bug-guile-ncurses] Unicode in form-driver |
Date: |
Fri, 10 Oct 2014 16:03:03 +0000 (UTC) |
Sorry. That was a long shot.
I uploaded to the Git repo an update to guile-ncurses
that will use the Unicode-enabled version of form-driver if
the underlying ncurses supports it. I tried with Polish characters;
it works for me. You need to build it on a version of ncurses
from the last few months, however.
I'll make a new release soon.
> On Friday, October 10, 2014 8:53 AM, Konrad Makowski <address@hidden> wrote:
> > Well, that doesn't work. I wrote something like that, and got only
> "??"
> instead of "?" when i write "ó".
>
> (map (lambda (i) (form-driver my-form (integer->char i)))
> (bytevector->u8-list (string->utf8 (string c))))
>
> --
> Konrad
>
>
>
> W dniu 09.10.2014 o 00:10, Mike Gran pisze:
>> I haven't tried it, but, you could try
>>
>> (map integer->char
>> (bytevector->u8-list (string->utf8 (string ch))))
>>
>> to convert a Unicode character to a list of bytes,
>> and then feed each byte to form-driver.
>>
>> But I can't test that today. Maybe tomorrow.
>>
>>
>>> On Wednesday, October 8, 2014 1:52 PM, Konrad Makowski
> <address@hidden> wrote:
>>>> Could you write me some example of second option (with UTF-8)?
>>> --
>>> Konrad
>>>
>>>
>>>
>>> W dniu 08.10.2014 o 23:48, Mike Gran pisze:
>>>> Konrad-
>>>>
>>>> I guess you're using UTF-8 and not ISO-8859-2.
>>>>
>>>> Support for characters with codepoints greater than U+00FF
>>>> is not available for form-driver because of limitations
>>>> in the upstream ncurses form library.
>>>>
>>>> A function was added this year into the upstream ncurses
>>>> form library that handles codepoints greater than U+00FF.
>>>> The "form_driver_w" function. I haven't wrapped
> that
>>> function
>>>> into guile-ncurses yet.
>>>>
>>>> For now, your options are
>>>> 1. use ISO-8859-2
>>>> 2. convert your character to set of UTF-8 bytes,
>>>> and feed each byte one-by-one into form-driver.
>>>>
>>>> I'll go ahead and add the new "form_driver_w"
> function
>>>> to guile-ncurses this week. But this will only work
>>>> with a version of NCurses from this year.
>>>>
>>>> Thanks,
>>>>
>>>> Mike Gran
>>>>
>>>>
>>>>
>>>>> On Wednesday, October 8, 2014 1:06 PM, Konrad Makowski
>>> <address@hidden> wrote:
>>>>>> I'm using guile-2.0.9, yes i called (setlocale LC_ALL
>>> "") and yes
>>>>> i
>>>>> built --with-ncursesw flag.
>>>>> In other part my app polish characters display correct.
>>>>>
>>>>> --
>>>>> Konrad
>>>>>
>>>>>
>>>>>
>>>>> W dniu 08.10.2014 o 23:02, Mike Gran pisze:
>>>>>> Hi-
>>>>>>
>>>>>> Three questions.
>>>>>>
>>>>>> Are you using guile-1.8 or guile-2.0 or the git
> version?
>>>>>>
>>>>>> Did you call (setlocale LC_ALL "") at the top
> of your
>>> program?
>>>>>> When you built guile-ncurses,
>>>>>> did you call ./configure with the --with-ncursesw flag
> so that it
>>>>>> is built with the Unicode-enabled version of ncurses
> forms?
>>>>>>
>>>>>>
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> Mike
>>>>>>
>>>>>>> On Wednesday, October 8, 2014 12:29 PM, Konrad
> Makowski
>>>>> <address@hidden> wrote:
>>>>>>>> Hello i try to write application with
> guile-ncurses and
>>> stuck.
>>>>>>> When i do something like that (form-driver my-form
> ch) and
>>> type some
>>>>>>> polish letter ie. "ł" or "ó"
>>>>>>> in form window apears ??.
>>>>>>>
>>>>>>> How can i avoid this issue?
>>>>>>>
>>>>>>> --
>>>>>>> Konrad
>>>>>>>
>