On Tue, 06 Jan 2009 15:52:59 +0800, Sean wrote: > >>> Technically, nothing more can be done there, as the algorithm for >>> searching is now doing no scanning, no loop, and no cache at all. >> >> Maybe it can be done in this way? If search for, say, "w", >> then ChineseIME will just search the list that >> starting from "w" and consisting just one or two charactors long. >> And if search for "ww", it searches the lists >> that start with "ww" and just two or three charactors long > > Good idea! I just implemented it, and uploaded another new version :) > > Improved performance when searching by one or two alphabets: > (1) to limit search where short characters are typed > (2) the default setting is hard-coded to be 2 > (3) for example, a<C-^> only shows maps for a only, not 'aiqing'
Would you consider my following advices please? 1. If the charactor that cursor is on isn't non-black-space ascii, then <space> just do normal <space> i.e. insert a real <space> instead of saying "pattern not found" that is really useless. 2. add an autocmd to toggle IME off when escape from insert mode. Both of them aren't hard to implement I think. > > Sean > > > > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ --~--~---------~--~----~------------~-------~--~----~ You received this message from the "vim_use" maillist. For more information, visit http://www.vim.org/maillist.php -~----------~----~----~----~------~----~------~--~---
