Re: LYNX-DEV Lynx-2.5 works but not 2.7 with Indian Script Terminal

[David Woolley]

> 
>     We have been  using lynx-2.5 on our Indian Script terminals.
>     it works satisfactorily. However lynx-2.6 
> 
>  lynx-2.7 do
>     not display the character  -> 
> #173; (soft hyphen), which stands

I would say that lynx-2.5 was broken and lynx-2.6 is behaving correctly.

>     for character "ai" in Indian Script.

No.  ­ stands for soft hyphen in all languages.  I lost the full
Unicode map in a disk crash and haven't reloaded it (try unicode.org),
but, depending on the language, your ai-kar should be somewhere in the
range ऀ to ൿ -

090 Devanagari (Northern Central India)
098 Bengali
0A0 Gurmukhi (Punjab)
0A8 Gujarati
0B0 Oriya
0B8 Tamil
0C0 Telugu (Andhra Pradesh?)
0C8 Kannada (Karnataka?)
0DO Malayalam (Kerala)

At a guess, the code you need is भ  (assuming that the offset from 0900
is the same as that from 0080).

This assumes that you are using entities, rather than directly embedding the
binary code, with an appropriate charset=.

I'm still waiting for the next released version so I've not experimented
with chartrans support, but it is almost certain that you will need to
add to the translate tables.  Note the theory is that the raw HTTP stream
is translated from the declared character set to unicode, entities are then
resolved into unicode, and finally the result is translated to the display
font.

If you are using binary codes, not entities and are not correctly declaring
the document and display character sets, you have to accept that you will
get behaviour consistent with the default document character set of
ISO 8859/1.

Note that it wouldn't surprise me if 2.7 has some ISO 8859/1 assumptions 
built in too deeply, but there are versions around which are mouch more
internationally aware.