![]() ![]() It might be worth asking on emacs-devel since the current maintainer of emacs is an OSX user. However I doubt such a solution would really be possible mostly because I doubt Apple provides an interface for emacs to access. As for using the OSX spellchecker I can't offer any advise since I don't use OSX. If you have qualms with how good it is you will have to take it up with those projects, which are completely separate from emacs. Emacs only ever uses an external program to do spell checking. If that's impossible, is there a way send Emacs input though OSX so I can at least make use of the superior OSX spellchecker?Īs you already pointed out, none of this is an emacs problem. Is there a better spellcheck algorithm I can use for Emacs, so that if I type ticke, Emacs can suggest ticket as the 1st, rather than the 8th, suggestion? Can't Emacs do better than this terrible, dumb algorithm? Huh? What the hell is "ch-acne"? The word I intended is the 10th word that Hunspell guesses. I know that this is technologically possible, because when I type the same thing in TextEdit, OSX automatically changes it to chance and I don't even have to think about it.īy contrast, here's the sequence I get from Hunspell:Ĭh acne ch acne ch-acne chicane Chicana Chicano chicken change Chance Chaney chance. What I would like is for Emacs to know what word I want, which is chance. The word I want, ticket, is the 8th word on the list, which means I have to do 8 keystrokes.īy contrast, in OSX TextEdit, I type ticke and OSX instantly autocorrects it to ticket. Hunspell gives me: ticker tick Tucker tucker toke tuck ticked ticket tickle tic. I realize I mistyped it, so I hit flyspell-auto-correct-word. In any case, Hunspell still isn't very good at guessing which word I mean. (Again I write only prose, no code.) I'm not sure if it's better. ![]() I switched from aspell to hunspell because I read that it had better spellcheck for English words. ![]() I need a better algorithm-ideally, an intelligent algorithm based on morphology and word distance. After all these years, I've realized that abbrev-mode is never going to learn all my typos. I'm aware that I can do what this blog post recommends and direct abbrev-mode to "learn" my corrections by adding them to the abbrev list and "sleep easier at night knowing I'll never see that mistake again (thanks to abbrev)." But I've been doing exactly this for 4 years, adding dozens of new typos my abbrev-list every day, and I still make a hundred new typos every day that are new and don't get corrected. What's so hard about automatically assuming that if I type onlyw I always want it corrected to only? I wish there were a way to get Emacs to automatically correct typos, without being prompted, as OSX does. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |