[developers] [spr30362] bug using climxm

[Francis Bond]

G'day,

Nice to hear from you.

> > Japanese, however doesn't work:
> > "en_GB.UTF-8"
> > Error: Error reading mb-vector from
> >        #<BUFFER-INPUT-SIMPLE-STREAM  pos 8 @ #x50d429aa>
>
> I'm working on reproducing that. When does the error apppear - when you
> enter a Japanese character into the text box, or when you advance to the
> next text box?

The error appears after the Japanese has been entered, when I hit
return once more to enter the text.  I add a little more context
below:

Error: Error reading mb-vector from
       #<BUFFER-INPUT-SIMPLE-STREAM  pos 8 @ #x50d429aa>

Restart actions (select using :continue):
 0: Umlauts! top level
 1: Exit Umlauts!
 2: Return to Top Level (an "abort" restart).
 3: Abort entirely from this (lisp) process.
[1] CL-USER(5): :zoom
Evaluation stack:

   (ERROR "Error reading mb-vector from ~s"
          #<BUFFER-INPUT-SIMPLE-STREAM  pos 8 @ #x50d429aa>)
 ->(OCTETS-TO-STRING #(230 151 165 ...) :END ...)
   (EUC-TO-STRING #(230 151 165 ...))
   (XT:GET-VALUES #<XT::XM-TEXT-FIELD NIL #x807e7c8 @ #x51ca8faa> :VALUE)
   ((METHOD (SETF CLIM:GADGET-VALUE) :AFTER (T XM-SILICA::MOTIF-TEXT-FIELD))
      "日本語" #<XM-SILICA::MOTIF-TEXT-FIELD @ #x51c92762>)
   ((:INTERNAL (:EFFECTIVE-METHOD 2 T T NIL NIL) 0) "日本語"

#<XM-SILICA::MOTIF-TEXT-FIELD
                                                      @
                                                      #x51c92762>)
   ((METHOD CLIM-INTERNALS::INVOKE-WITH-OUTPUT-AS-GADGET (T T))
      #<CLIM-INTERNALS::ACCEPT-VALUES-STREAM @ #x51b62bc2>
      #<Closure # @ #xbfe5fae2> ...)
   ((:INTERNAL (:EFFECTIVE-METHOD 2 T T NIL NIL) 0)
      #<CLIM-INTERNALS::ACCEPT-VALUES-STREAM @ #x51b62bc2>
      #<Closure # @ #xbfe5fae2> ...)
   [... CLIM-INTERNALS::INVOKE-WITH-OUTPUT-AS-GADGET ]
   (CLIM-INTERNALS::MAKE-GADGET-FOR-TEXT-FIELD-VIEW
      #<CLIM-INTERNALS::ACCEPT-VALUES-STREAM @ #x51b62bc2>
      #<CLIM:TEXT-FIELD-VIEW @ #x4476ef8a> ...)

... more older frames ...
[1] CL-USER(6):

> > Further, although the display of Korean works in the input text window
> > and the window title bars (which it didn't before) it doesn't display
> > any of the lkb's output displays.  Instead, each character appears as
> > a latin character, some accented.  The internal representation is
> > fine: it parses and it saves as utf-8, this is purely a display
> > problem.
>
> Testing it, I see the issue too. While testing, I found out that the
> same problem appears with Korean button labels. I'm working on fixing
> that right now.
>
> Thanks for the report. I think I should have an update within the next
> two days.

Great.  I shall look forward to it.

-- 
Francis Bond <http://www2.nict.go.jp/x/x161/en/member/bond/>
NICT Computational Linguistics Group