Re: lynx-dev Handling of untranslatable charset

[Leonid Pauzner]

16-Jan-99 02:25 Klaus Weide wrote:
> Handling of untranslatable charsets was wrong.  I first noticed
> when a page with charset=ISO-2022-JP produced an alert message
> with a truncated "iso-2022-j".  Looking more closely, the code
> could also result in memory corruption in some cases.

Yes, it was a corruption from rudiment of "Other ISO Latin" (my fault),
but seems iso-2022-jp should be translated to euc-jp by UCGetLYhndl_byMIME():
why it wasn't done before? It should be converted to canonical name,
otherwise Set_HTCJK() will fail.

Or we should replace lines like  !strcmp(name, "euc-cn")
with  UCGetLYhndl_by_MIME(name) == UCGetLYhndl_byMIME("euc-cn")
in Set_HTCJK()?  But the problem may probably persists elsewhere.


> I looked at older versions, then changed the code to something
> somewhat like what was intended there. But we don't have an
> "Other ISO Latin" display character set any more; maybe those
> `else' blocks should just have disappeared completely.

Hope no heuristic, nothing 'else',
just leave some kind of warning message for very bad case
like CJK document and HTCJK = NOCJK (however, we couldn't check
whether the document is of 7bit chars only at this stage)

>     Klaus

> (Patch against recent dev code)