This set of patches changes struct term_event_keyboard to use UCS-4 if CONFIG_UTF_8 is defined. Thus, there will be just one event per character, rather than one per byte. The binary interlink protocol is completely unaffected, although the implementation changes quite a bit.
The patches supersede the ones I posted in these messages: [patch 0/5] mid:[EMAIL PROTECTED] [patch 1/5] mid:[EMAIL PROTECTED] [patch 2/5] mid:[EMAIL PROTECTED] [patch 3/5] mid:[EMAIL PROTECTED] [patch 4/5] mid:[EMAIL PROTECTED] [patch 5/5] mid:[EMAIL PROTECTED] These changes fix bug 782 and allow input fields to guard against invalid UTF-8. They also make it easier to fix bug 777 or add support for binding actions to non-ASCII keys. However, non-ASCII hotkeys do not yet work when CONFIG_UTF_8 is defined. Note the "Character Conversions" comments patches 09/11 and 11/11 in this series add to handle_interlink_event(). They describe how I think different combinations of CONFIG_UTF_8, UTF-8 I/O, and the UTF-8 charset should behave; I tried to write the code accordingly. If you think the behaviour described in these comments is not correct, now is a good time to post your opinion.
pgpxe5LSLzJdV.pgp
Description: PGP signature
_______________________________________________ elinks-dev mailing list [email protected] http://linuxfromscratch.org/mailman/listinfo/elinks-dev
