On Wed, Aug 15, 2012 at 11:38:38PM +0200, Vincent Lefevre wrote: > retitle 673385 lynx-cur: in UTF-8 locales, lynx displays search text at the > wrong column if preceded by non-ASCII (multibyte) characters > thanks > > On 2012-08-15 16:46:03 -0400, Thomas Dickey wrote: > > On Wed, Aug 15, 2012 at 01:01:56PM +0200, Vincent Lefevre wrote: > > > On 2012-08-15 05:46:53 -0400, Thomas Dickey wrote: > > > > On Fri, May 18, 2012 at 10:37:06AM +0200, Vincent Lefevre wrote: > > > > > lynx displays search text at the wrong column if preceded by UTF-8 > > > > > characters. For instance, consider: > > > > > > > > This is addressed by setting > > > > > > > > XHTML_PARSING:true > > > > > > > > in lynx.cfg > > > > > > I don't see why a display problem should be affected by parsing. > > > > The character-set information is given only as an xml processing > > instruction, > > which in turn is used by lynx only when XHTML_PARSING is set. > > But lynx got the charset information right. Otherwise it couldn't > have output the ellipsis characters correctly!
not exactly (I think this would be a long explanation...) Running your example (default lynx.cfg, uxterm, Debian/testing), I pressed "=" to see the document charset is "iso-8859-1". I pressed "o" to see that the display charset is UTF-8, while the document charset is iso-8859-1. So far I'm not seeing any details that contradict my advice above. -- Thomas E. Dickey <dic...@invisible-island.net> http://invisible-island.net ftp://invisible-island.net
signature.asc
Description: Digital signature
