> > With Lynx 2.8.6dev.17, the string '&target_' > is translated into '%E2%8C%96_' in a link. > In the body test, it is not changed.
It should attract error recovery processing in both contexts as there is no ⌖ entity. There is a whole appendix to the HTML specification on this issue. The reccommended solution is to configure the server to treat ";" as equivalent to "&" and use ";" as the parameter delimiter. The problem arises because form style URLs were only intended to be created by form submission. You may find your server is already so configured. Failing that, you must escape (entity encode) the & in all contexts. > > Must I use '&target_'? If you don't want the browser to attempt to substitute the value of an entity called ⌖ > Content-Type: application/octet-stream; > name="strange_translation_in_link.html" Hint. Attaching text/ media types as application/octet-stream makes them unnecessarily difficult to read. > PGEgaHJlZj0iLz9tPXBjJnRhcmdldF89MyI+IGludmFsaWQhISA8L2E+DQo8YSBocmVmPSIv > P209cGMmYW1wO3RhcmdldF89MyI+IHZhbGlkIDwvYT4NCg== > _______________________________________________ Lynx-dev mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/lynx-dev
